Summary
In Rambus, however, the district court held that the doctrine of stare decisis bound it to follow the Federal Circuit's previous construction of the same term at issue, "integrated circuit device."
Summary of this case from In re National Collegiate Athletic Association Athletic Grant-In-Aid Cap Antitrust LitigationOpinion
Nos. C-05-00334 RMW, C-05-02298 RMW, C-06-00244 RMW.
July 10, 2008.
Burton Alexander Gross, Esq., Carolyn Hoecker Luedtke, David C. Yang, Erin C. Dougherty, Jennifer Lynn PoIse, Miriam Kim, Peter A. Detre, Rosemarie Theresa Ring, Esq., Munger Tolles Olson LLP, San Francisco, CA, Gregory P. Stone, Keith Rhoderic Dhu Hamilton, II, Sean Eskovitz, Steven McCall Perry, Munger Tolles Olson LLP, Rollin Andrew Ransom, Sidley Austin LLP, Los Angeles, CA, Craig N. Tolliver, Pierre J. Hubert, McKool Smith, Kevin S. Kudlac, Austin, TX, for Plaintiff.
Kenneth Lee Nissly, Susan Gregory Vankeulen, Karin Morgan Frenza, Tomomi Katherine Harkey, Thelen Reid Brown Raysman Steiner LLP, San Jose, CA, Kenneth Ryan O'Rourke, Patrick Lynch, Belinda Martinez Vega, O'Melveny Myers LLP, Yonaton M. Rosenzweig, Robert Jason Becher, Robert E. Freitas, Yonaton M. Rosenzweig, Quinn Emanuel Urquart Oliver Hedges, LLP, Los Angeles, CA, Theodore G. Brown, III, Daniel J. Furniss, Jordan Trent Jones, Joseph A. Greco, Townsend and Towsend and Crew LLP, Palo Alto, CA, Geoffrey Hurndall Yost, Thelen Reid Priest, LLP, Linda Jane Brewer, Robert Jason Becher, Quinn Emanuel Urquhart Oliver Hegdes, LLP, Eric R. Lamison, Kirkland Ellis LLP, San Francisco, CA, Jason Sheffield Angell, Kaiwen Tseng, Orrick, Herrington Sutcliffe, LLP, Menlo Park, CA, John D. Beynon, Weil, Gotshal Manges LLP, Redwood Shores, CA, Vickie L. Feeman, Chester Wren-Ming Day, Craig R. Kaufman, Kaiwen Tseng, Jan Ellen Ellard, Jason Sheffield Angell, Robert E. Freitas, Orrick Herrington Sutcliffe, Menlo Park, CA, Davin M. Stockwell, Jan Ellen Ellard, Jason Sheffield Angell, Orrick Herrington Sutcliffe LLP, Mark Shean, Irvine, CA, John D. Beynon, Claire Elise Goldstein, Edward Robert Reines, Weil, Gotshal Manges LLP, Redwood Shores, CA, Ana Elena Kadala, David J. Healey, Anita E. Kadala, Weil Gotshal Manges, Norma N. Bennett, Houston, TX, Matthew J. Antonelli, Robert S. Berezin, Weil Gotshal Manges, New York, NY, for Defendants.
Rambus has accused the Manufacturers of infringing various patents. Largely in accord with the local rules, the parties have submitted their joint claim construction statement showing 72 claim terms in dispute. Rambus has filed its opening and responding Markman briefs, motions for summary judgment of infringement, and oppositions to the Manufacturers' motions. The Manufacturers have filed their responsive Markman brief, oppositions to Rambus's summary judgment motions, and their own motions for summary judgment of invalidity under Rambus's proposed claim constructions and non-infringement under theirs. The court has reviewed the papers and considered the arguments of counsel and now sets forth its claim construction and rulings on the summary judgment motions dealing with the Farmwald/Horowitz patents. The claim construction pertaining to the Ware patents is set forth in a separate order.
The court collectively refers to the Hynix, Micron, Nanya, and Samsung entities in this suit as "the Manufacturers."
Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996).
I. THE FARMWALD/HOROWITZ PATENT FAMILY
Fifteen of the seventeen patents-in-suit descend from the original patent application no. 07/510,898 filed by Drs. Michael Farmwald and Mark Horowitz on April 18, 1990. Because these fifteen share substantially similar specifications, the court's discussion refers to U.S. Patent No. 6,182,184. Given the number of claim terms in dispute and complexity of the technology, the court begins by explaining the context of the invention and the contents of the specification relevant to the disputed issues. Cf. Phillips v. AWH Corp., 415 F.3d 1303, 1313, 1315-17 (Fed. Cir. 2005) (en banc). It then briefly recounts aspects of the prosecution history before turning to claim construction.
The table in Appendix 1 reflects the patents-in-suit, the claims asserted, and the Manufacturers against whom the claims are asserted.
Rambus's opening and reply briefs exclusively refer to U.S. Patent No. 6,426,916. The court cannot ascertain any substantive difference between the two patents' specifications. Because of the patents' different prosecution histories described at the beginning of the specification, however, citations to one patent's specification do not map to the other. Unexplainedly, the parties did not harmonize their briefing on this point.
A. Background of the Inventions and the Specification's Written Description
Drs. Farmwald and Horowitz began their collaboration in the fall of 1988. Tr. 4078:21-4079:7. Dr. Farmwald met with Dr. Horowitz over dinner to discuss how processor speeds and memory speeds were diverging and how memory systems needed to become faster to keep up. Tr. 4079:9-4082:9. Within the semiconductor industry, this problem was commonly referred to as the "memory bottleneck" or "memory gap." Tr. 4084:7-4091:8 (Dr. Horowitz); 4161:10-4163:3 (Carl Everett); 5498:9-5502:8 (Dr. Farmwald). Over the course of the next year and a half, Drs. Farmwald and Horowitz worked on a variety of ideas for closing the memory gap, and they eventually wrote up their ideas in a patent application. Tr. 4133:15-4134:14. Dr. Horowitz testified that he "took over" the drafting of the specification. Id. The following discussion walks through the patents' common specification to illustrate the scope of the written description and explain the technology.
This case has spawned a number of transcripts. Unless otherwise noted, citations to the transcript refer to the trial transcript from the bifurcated, consolidated trial the court held from January to March of 2008 regarding some of the Manufacturers' fraud and antitrust counterclaims.
1. The Prior Art and Objects of the Invention
Dr. Horowitz testified that with the specification, he and Dr. Farmwald were "trying to describe our inventions, all the innovations that we had come up with to build a very high speed interface." Tr. 4134:10-14. The court will summarize here, however, only the intrinsic evidence and not the inventors' self-serving testimony. The specification begins with the field of the invention, where it describes "an integrated circuit bus interface" as well as "a new method for physically implementing the bus architecture." `184 patent, col. 1, ll. 21-26. From these introductory sentences, it is clear that the Farmwald/Horowitz specification discloses more than one invention.
The background of the invention discusses general features of prior art memory devices, focusing on how prior art bus architectures were not as efficient as they could be. Id., col. 1, l. 30-col. 2, l. 5. The comparison with the prior art is extensive and illustrates some of the problems that Drs. Farmwald and Horowitz sought to address with their inventions. It starts by noting that "prior art memory systems have attempted to solve the problems of high speed access to memory with limited success." Id., col. 2, ll. 8-10. The first piece of prior art examined by the specification is "the earliest 4-bit micro processor." Id., col. 2, ll. 10-11 (citing U.S. Pat. No. 3,821,715 (Hoff et. al.)). The Hoff micro processor connected multiple memory devices to a single CPU over a 4-bit wide bus that multiplexed, i.e., carried both, address and control information. Id., col. 2, ll. 13-16. It also used point-to-point control signals to select which memory device the CPU sought to access. Id. Drs. Farmwald and Horowitz critiqued aspects of the Hoff micro processor, noting that it used fixed access times for transmitting information and did not permit the memory devices to send information in blocks. Id., col. 2, ll. 16-20. "Most important[ly]," its use of point-to-point device select lines meant that it did not send device control information over the bus. Id.
The next piece of prior art discussed is U.S. Patent No. 4,315,308 (Jackson), which described a bus architecture using a single 16-bit wide bus that multiplexed data, address, and control information. Id., col. 2, ll. 21-24. It implemented some block-mode operations and allowed some access time variation, but it could not handle multiple requests and did not bus all signals. Id., col. 2, ll. 24-30. Another patent described a DRAM with multiplexed address and data information inside the DRAM, but with a "conventional" bus comprised of separate data, address, and control lines to connect to the external device environment. Id., col. 2, ll. 31-34.
Shifting away from describing prior art bus architectures, Drs. Farmwald and Horowitz described prior art packaging technology. Id., col. 2, ll. 35-42. While others had attempted to use 3-D packages for DRAMS with connections along a single edge only, the need for point-to-point wiring to enable a master device to select the correct memory device posed complex geometrical problems. Id. No prior art packaging solution had considered using a new interface to do away with the need for point-to-point device select wiring.
Leaving DRAM packaging and returning to DRAM architecture, Drs. Farmwald and Horowitz discussed the "state-of-the-art DRAM interface" in U.S. Patent No. 3,969,706 (Proebsting, et. al.). This DRAM architecture used two-way multiplexed address signals and maintained separate pins for data and control information. Id., col. 2, ll. 43-47. One of its problems was that it required more and more pins as the DRAM grew, and many of these pins had to be connected point-to-point to various devices. Id., col. 2, ll. 47-50. As previously discussed, this interface complexity made using a 3-D package difficult.
The specification next discusses backplane buses, i.e., the circuit boards used for connecting multiple devices to each other. Some prior art backplane buses had multiplexed address and data information, others used relatively low-voltage signals, and some others used programmable registers and block mode operations. Id., col. 2, ll. 51-60. While all backplane buses used some arbitration scheme for prioritizing signals, none used the bus arbitration scheme devised by Drs. Farmwald and Horowitz. Id., col. 2, l. 61 — col. 3, l. 2. Furthermore, all prior art buses used some point-to-point connections. Id., col. 3, ll. 2-4.
Finally, the specification summarizes some prior art related to clocking a DRAM interface. Id., col. 3, ll. 8-21. Drs. Farmwald and Horowitz noted that "the clocking scheme used in this invention has not been used before and in fact would be difficult to implement in backplane buses due to the signal degradation caused by connector stubs." Id., col. 3, ll. 8-11. While a prior art patent had described a clocking scheme using two clock signals, it relied on unusual ramp-shaped clocking signals in lieu of the conventional square signals used by Drs. Farmwald and Horowitz. Id., col. 3, ll. 11-14.
After having surveyed various aspects of the prior art, the specification sets out seven "objects of the invention." See id., col. 3, ll. 22-48. First, the invention uses "a new bus interface built into semiconductor devices to support high-speed access to large blocks of data from a single memory device by an external user of the data, such as a microprocessor, in an efficient and cost-effective manner." Id., col. 3, ll. 22-26. Though the invention seeks to address six other objectives, Rambus acknowledges that the "primary object that's accomplish by the claims at issue is the very first one, which is the new bus interface built into semiconductor devices to support high-speed access to large blocks of data from a single memory device." Claim Construction Hrg. Tr. 114:7-20 (June 4, 2008). The invention's second objective is "to provide a clocking scheme to permit high speed clock signals to be sent along the bus with minimal clock skew between devices." Id., col. 3, ll. 27-29. The invention also seeks to provide a method for mapping out defective devices, distinguishing identical devices connected to the bus, and assigning unique identifiers to otherwise-identical devices. Id., col. 3, ll. 30-35. Yet another object of the invention is to allow a device to transfer address, data, and control information over a relatively narrow bus and provide a bus arbitration scheme to allow multiple devices to use the bus simultaneously. Id., col. 3, ll. 36-40. Finally, the invention seeks "to provide devices, especially DRAMs, suitable for use within the bus architecture of this invention." Id., col. 3, ll. 46-48.
2. The Summary of Invention
The summary of invention begins by describing the new bus interface with substantially fewer bus lines, multiplexed address, data and control information, and no point-to-point device select lines because device select information is included in the control signal. Id., col. 3, ll. 51-61. This new bus also includes clock signals and power along with the multiplexed address, data and control signals. Id., col. 4, ll. 1-2. While the preferred implementation of the bus interface uses 8 bus data lines and an Addressvalid bus line, "persons skilled in the art will recognize that 16 bus data lines or other numbers of bus data lines can be used to implement the teaching of this invention." Id., col. 4, ll. 5-8. With the new interface, memory devices can use the bus more efficiently by making large block data transfers and simultaneous transactions. Id., col. 4, ll. 10-20.
The summary next notes that the DRAMs that connect to the new bus differ from conventional DRAMs. The new DRAMs use registers to store control information, device information, and whatever else might be appropriate. Id., col. 4, ll. 21-26. They also require circuitry for creating an internal device clock to synchronize the devices on the bus. Id., col. 4, ll. 31-33.
Finally, the summary of invention explains that the constrained bus environment enables high clock speeds by shortening the necessary length of the bus. Id., col. 4, ll. 35-50. Taken as a whole, these innovations improve DRAM bandwidth while reducing manufacturing costs and power consumption. Id., col. 4, ll. 51-55.
3. The Detailed Description
The detailed description of the invention begins with an overview of the new technology, then splits into discrete sections addressing various features of the invention. These feature-specific discussions address: device address mapping (col. 7, l. 17), bus (col. 8, l. 16), protocol and bus operation (col. 8, l. 42), retry format (col. 12, l. 1), system configuration/reset (col. 14, l. 46), error detection and correction (col. 16, l. 20), low-power 3-D packaging (col. 17, l. 14), bus electrical description (col. 17, l. 62), clocking (col. 18, l. 62), multiple buses (col. 19, l. 45), device interface (col. 21, l. 23), electrical interface-input/output circuitry (col. 21, l. 41) and DRAM column access modification (col. 23, l. 42). Only some of these discussions bear on the construction and validity of the claims in dispute.
a. The Multiplexed Bus Architecture
The detailed description starts by noting that "the present invention is designed to provide a high speed, multiplexing bus for communication between processing devices and memory devices and to provide devices adapted for use in the bus system." Id., col. 5, ll. 29-31. The bus has a number of features. First, it is narrow, i.e., it has a relatively small number of bus lines. Id., col. 5, ll. 35-37. Second, it is multiplexed, meaning, it carries substantially all address, data and control information on the same bus data lines. Id., col. 5, ll. 37-45. This obviates the need for device-select lines because device-select information is carried over the bus. Id.
The devices connected by the bus include master devices like the CPU and slave devices like a DRAM. Id., col. 6, ll. 12-16. Every semiconductor device, whether master or slave, has a number of programmable registers for storing information, including device identification, device type, and control settings. Id., col. 6, ll. 27-52. For example, preferred embodiments of the patent's semiconductor devices contain access-time registers that store delay times that control when the device can send and receive data. Id., col. 6, ll. 32-37. These register settings can be modified when the device is turned on. Id., col. 6, ll. 38-52.
The specification's preferred bus architecture has 11 signal lines. Id., col. 8, ll. 17-34. There are eight bus data lines referred to as "BusData0" through "BusData7" and collectively described as BusData[0:7]. Id., col. 8, ll. 17-23. These data lines provide a multiplexed bus for address, data and control information that is one byte wide. Id., col. 8, ll. 24-25. There are also two clock lines (Clk1 and Clk2) and an AddrValid line. Id. These lines serve to synchronize all of the devices on the bus and to indicate whether the information on the bus includes an address request for all slave devices to decode. Id., col. 8, ll. 25-30. Finally, there is an additional line (ResetIn, ResetOut) that connects all devices attached to the bus. While the other signal lines on the bus connect to every device in parallel, the ResetIn, ResetOut line connects all of the devices in series so that it can program their registers and assign them unique device identification numbers. Id., col. 8, ll. 30-35.
b. The Packet Protocol
The bus is the sole courier of information between devices on the bus. Id., col. 6, ll. 53-55. This requires a communication protocol. Id., col. 6, ll. 55-60. The specification describes using a packet protocol. Id. It begins when a master device sends a request packet ("a sequence of bytes comprising address and control information"). Id. Every slave device must decode the packet to determine if it must respond to the request packet. Id., col. 6, ll. 62-64. If it does, the slave device begins its internal processes and responds to the request packet by sending or retrieving data after a programmed delay time. Id., col. 6, l. 64 — col. 7, l. 4.
A preferred embodiment of the protocol begins when a master device sends out a request packet of address and control information onto the bus. Id., col. 8, ll. 59-62. The AddrValid line is activated, which instructs all slave devices to decode the incoming request packet and decide whether the request packet applies to them. Id., col. 8, l. 66 — col. 9, l. 4. If the packet contains the slave device's address, the slave device responds by either transmitting data to the master for a read request or receiving data from the master for a write request. Id. This response occurs after a delay time specified in the memory device's access-time register. Id., col. 9, ll. 11-23.
Exhibit
Figure 4 shows the preferred embodiment of a request packet. The request packet is nine bits wide, corresponding to the AddrValid bus line plus the eight bus data lines BusData[0:7], and the packet is six bytes long. Id., col. 9, ll. 24-29. The first byte begins with the AddrValid bit equal to 1 to indicate the beginning of a request packet. Id., col. 9, ll. 29-31. The first byte then contains two four-bit fields sent over the bus data lines. Id., col. 9, ll. 39-46. The first four bits are known as AccessType[0:3]. Id., col. 9, ll. 39-41. AccessType[0:3] is an "op code" or "operation code" that specifies the type of access for the DRAM to perform. Id. The second four bits, known as Master[0:3], serve to identify the master device sending the request packet. Id., col. 9, ll. 41-46. Briefly, the forty bits of address data in Address[0:35] and Address[36:39] specify the slave device that should perform the requested operation. See id., col. 9, ll. 35-38. Finally, the last four bits of the request packet, known as BlockSize[0:3], tell the DRAM the amount of data involved in the requested operation. No actual data is sent or received in the request packet. The request packet simply allows each DRAM on the bus to: (1) recognize the next six bytes as a request packet, (2) understand what operation is requested, (3) recognize the source of the request packet, (4) determine whether the request packet is addressed to it, and (5) determine the amount of data to send or receive in response to the request packet.
In the preferred embodiment, AccessType[0], i.e, the first bit of the four bit AccessType[0:3] field, is a read/write switch. Id., col. 9, ll. 47-56. If AccessType[0] equals 1, the requested operation is a read and the DRAM should access the data in its memory array and output the data to the bus. Id., col. 9, ll. 50-56. If AccessType[0] equals zero, the requested operation is a write and the DRAM should prepare to receive data from the bus. Id. The other three bits of the AccessType field, AccessType[1:3], could be used to specify the delay time of the DRAM's response or trigger preprogrammed responses. See id., col. 9, ll. 56-65; col. 11, ll. 19-37. For example, AccessType[3] can be used to tell the DRAM whether to precharge the sense amplifiers in the memory device to an intermediate voltage between zero and one to allow for a faster response to the next request or to save the information already contained in the memory array. See id., col. 10, ll. 15-48; col. 11, l. 37.
The BlockSize[0:3] field determines the size of the following data block transfer. Id., col. 11, ll. 38-39. In the preferred embodiment, if BlockSize[0] equals zero, the size of the following data block corresponds to the binary value of BlockSize[1:3], enabling the data block to be 0 to 7 bits long. Id., col. 11, ll. 39-40. If BlockSize[0] equals one, the data block's length corresponds to the value of 2 to the power of the binary value of BlockSize[1:3], beginning with 23, or 8, bits in length. Id., col. 11, ll. 40-41. Hence, a BlockSize[0:3] equal to 0001 corresponds to a data block that is one bit long. By contrast, a BlockSize[0:3] equal to 1111 corresponds to a data block that is 1,024 bits long, i.e., 210. See id., col. 11, ll. 48-57. These interpretations of BlockSize[0:3] are only illustrative; other encoding schemes are possible simply by changing the meaning of the field's values.
Once the DRAM has decoded the request packet, it must wait the programmed delay time. Id., col. 11, 61-63. It then responds by reading or writing data over the bus lines BusData[0:7] while the AddrValid bus line remains set to zero to indicate that data, and not another request packet, are being transmitted on the bus. Id.
c. The Clocking Scheme — External and Internal Clock Signals
As a memory device gets faster, i.e., as the time between signals decreases (or, as the frequency of the system increases), the time a signal takes to propagate down the bus becomes significant compared to the time between signals. This error created by the time it takes for the signal to travel can disrupt the ability of the memory devices to operate synchronously with each other.
The specification discloses one way of minimizing this error by having each device receive two clock signals. Id., col. 18, ll. 63-66. The device can then use the two external clock signals to derive an internal device clock. Id. If every device uses the external clock signals to derive the same internal clock signal, each device's internal clock can reflect a true system clock shared by all devices, despite the devices being in different positions on the bus. Id.
Exhibit
Figure 8a shows the preferred embodiment of the clocking scheme devised by Drs. Farmwald and Horowitz. The clock generator (CLK) sends an early clock signal along the bus's CLOCK1 line, where the signal first contacts Chip O and later contacts Chip N. Id., col. 19, ll. 3-7. The clock signal then loops around on a second line (the CLOCK2 bus line), where it now contacts Chip N first, then contacts Chip O. Id., col. 19, ll. 7-10. An alternative embodiment uses only one bus clock line and leaves the end of the clock line unterminated, allowing the clock signal to reflect back when it reaches the end of the clock line. Id., col. 19, ll. 10-13. The reflected clock signal performs the same function as the CLOCK2 signal in the preferred embodiment while remaining confined to the same clock line. Id.
Graph
Figure 8b depicts the method used by each memory device to derive an internal clock signal from the two external clock signals. Figure 8a's example shows that Chip O is closer to the clock generator. Chip O therefore receives the clock signal on CLOCK1 before Chip N, as demonstrated by comparing when each chip detects an increase in voltage on CLOCK1 (represented by the solid black line, 55). See id., col. 9, ll. 14-32. Meanwhile, Chip N is closer to the source of CLOCK2 than Chip O. Therefore, Chip N receives the second clock signal, CLOCK2, earlier than Chip O. Id. CLOCK2 is shown in Figure 8b by the dashed line, 56.
Figure 8b lacks axis labels on its two graphs. To a person of ordinary skill in the art, it would be clear that the x-axis represents time. It would be similarly clear that the y-axis represents the voltage on the clock lines, as measured by each chip.
If each memory device is connected to the bus's clock signal lines properly, the average of the times at which each chip receives CLOCK1 and CLOCK2 should be identical. In Figure 8b, this is shown as the dashed vertical line labeled 59. Thus, despite the propagation delay caused by each device being placed at a different point on the bus, each device can use two external clock signals to derive an internal clock signal that is the same in every device.
Figure 8b demonstrates another feature of Drs. Farmwald and Horowitz's invention, namely, the use of both edges of the clock signal (also referred to as the rising and falling edges of the clock signal) to gain twice the amount of information out of a single period of the clock signal. See id., col. 9, ll. 33-44. As discussed, the two chips on the bus use the transition of the two clock signals from low to high to derive the midpoint depicted by the dashed line 59. The two chips also detect when each clock signal decreases, i.e., goes from high to low, as shown at 57 (CLOCK1) and 58 (CLOCK2). Each chip can calculate the midpoint of the two decreases in the two clock signals, shown by the dashed vertical line 60. As with the prior midpoint, the two chips are connected to the clock signal lines so that the midpoint measured by Chip N is the same as the midpoint measured by Chip O. By using the midpoint of the rise and fall of the two clock signals, the devices on the bus can respond at two different points in time for each period of the clock signal. Thus, the bus data frequency can be twice the clock signal frequency. See id.
d. The Clocking Scheme — Responding to Internal Clock Signals
The preceding discussion omits some of the complex details required to compensate for the lag between the internal clock signal and the device's interaction with the other bus lines. To begin, a device requires both input and output circuitry for responding to signals from the bus. The input circuitry includes an "input sampler" for measuring the voltage on the bus lines and detecting an incoming signal. See id., col. 22, ll. 21-35. The output circuitry contains an "output driver" for putting data onto the bus. See id., col. 22, ll. 1-20. For example, the input sampler has a slight delay between when it receives a clock signal and its sampling of the bus's data lines. Because of the system's high frequency, it is important to minimize any such delay. Id., col. 22, ll. 50-56.
Graph
Figure 12 reveals that omitted complexity. The first thing to recognize is the inputs for the first clock signal (early clock, or CLOCK1) and second clock signal (late clock, or CLOCK2). The connections between the device and the bus's clock lines are labeled CLK1 (100) and CLK2 (110). To help orient Figure 12 within the context of the prior discussion, the edited excerpt of Figure 8a below needs to be referenced. All of the circuitry shown in Figure 12 exists in each device attached to the bus.
The first element of the circuit is a DC amplifier (102), which is necessary for converting the bus clock line's relatively low voltage signal into a more powerful signal that the device can more readily recognize. See id., col. 22, ll. 60-61. The amplified clock signal then feeds into a variable delay line (103), which in turn feeds three different delay lines (104, 105, and 106). Id., col. 22, ll. 61-66. The first delay line (104) has a fixed delay time. Id. The second delay line (105) has the same fixed delay time plus a second variable delay time. Id. The third delay line (106) has the same fixed delay as the prior two lines plus a variable delay time that is half of delay line 105's second variable delay time. Id. In Figure 12, the fixed delay time is represented as T0 and the second variable time is represented as X. The third delay line's delay time equals the midpoint of the delay times of the first and second delay lines.
Exhibit
The outputs of the first two delay lines (104 and 105) are labeled 107 and 108. Id., col. 22, ll. 66-67. These outputs connect to the clocked input receivers on each of the two bus clock lines labeled as 101 and 111. Id., col. 22, l. 66 — col. 23, l. 4. To be clear, these input receivers receive two signals: the external clock signal from the bus's clock line and the internal delay signal. See id., col. 23, ll. 2-4 Fig. 11. The clocked receivers then run feedback lines (115, 116) back into the variable delay lines (103, 105). Id., col. 23, ll. 3-7. The timing diagram in Figure 13 assists in understanding the purpose of all of this circuitry.
Exhibit
As with Figure 8b, the unlabeled x-axis is time and the unlabeled y-axes of the graphs in Figure 13 are the voltage levels on the various lines. For context, labels 121 and 123 correspond to the falling edges of the first and second, or early and late, bus clocks. Id., col. 23, ll. 7-13. When properly calibrated, delay lines 107 and 108 output signals similar to the early and late bus clocks respectively, but shifted early by a slight amount (128). Id., col. 23, ll. 3-13. This slight amount (128) is ideally equal to the delay time before the input sampler can respond to the clock signal and sample the bus's data lines. Id.
As discussed, the third delay line has a fixed delay time and half of the variable delay time of the second delay line. This puts the output of the third delay line (also known as the internal clock signal 73) at the midpoint of the other two delay lines. Id., col. 23, ll. 13-18. This is simply the distance labeled 129 between the internal clock signal's falling edge (73) and the two internal delay lines' falling edges (labeled 120 and 122). This final output is the device's derived internal clock signal, which precedes the true midpoint of the two bus clocks by a small amount of time equal to the delay in the input sampler's operation. In the preferred embodiment, the device also generates an inverse of the internal clock signal, referred to as the complementary internal clock signal and labeled 74 in Figure 13.
To complete the story, the device has two internal clock signals that mirror each other. Each precedes the true midpoint of the bus clock signals by an amount of time equal to the delay in operating the input sampler. The preferred embodiment then harnesses half of the input samplers to one clock signal and has them sample the bus on the clock signal's falling edge, while the other half of the input samplers sample the bus on the complement's falling edge. Id., col. 23, ll. 26-34. In such a fashion, the device successfully samples the bus for data at the midpoint of the two external clock signals' rising and falling edges.
B. Prosecution History
1. The Original 11-Way Restriction Requirement
As mentioned, Drs. Farmwald and Horowitz filed their original patent application no. 07/510,898 with 150 claims on April 18, 1990. The PTO responded with a restriction requirement, forcing Drs. Farmwald and Horowitz to separate their claims into 11 distinct applications. See Detre Decl., Ex. A. The examiner defined a first group of claims (Group I, covering claims 1-45 and 56-72) as the combination of "a memory and bus subsystem including address registers, transceivers, and memory sections" and a "general mention of memory control." Id. at 3. He then proceeded to distinguish other groups of claims as follows: Group Original claims Basis for distinction Source Id. Id. Id. Id. See id. Id. Id. Id. Id. Id.
II 46-55 Group I related to memory control generally. Group II claims "a specific at 3. memory control and arbitration scheme." III 73-81 Group I related to timing generally. Group III claims "a specific clocking at 4. and timing scheme." IV 82-90 Group IV claimed "an unrelated DRAM device." at 1. V 91-94 Group V claimed "an unrelated semiconductor package." at 1. VI 95-105 Group VI claimed "a semiconductor device and connection means" and generally mentioned "bus characteristics." The examiner did not find a at 4-5. "combination/subcombination" relationship between Group I and Group VI. VII 106, 107 Group VI made only "a general mention of bus characteristics." Group VII at 5. claimed "specific bus characteristics." VIII 108-110 Group VI generally mentioned "timing and clocking." Group VIII included at 6. "specific clocking and timing considerations." IX 111-113 Group VI generally mentioned "storage." Group IX included "specific at 6. receiving and storage means." X 114-123 Group VI generally mentioned "I/O." Group X related to "a specific I/O at 7. and multiplexing scheme." XI 124-150 Group VI generally mentioned "input and output." Group XI related to "a at 8. specific input/output scheme using packets." In laying out the details of the original restriction requirement, the court recognizes its limited evidentiary significance. First, the examiner purportedly distinguished claims going to separate inventions. This clearly distinguished groups IV and V from the other inventions. What is unclear is the relationship, if any, between Group I and Group VI. The examiner used the same shorthand to refer to both groups as the combination A and subcombination Bbr, jointly "ABbr." Yet he defined the meaning of "A" differently for Group I and Group VI. For Group I, "A" represented "a memory and bus subsystem including address registers, transceivers, and memory sections." For Group VI, "A" represented "a semiconductor device and connection means." Because it appears that Group I and Group VI claim different inventions, it seems that the additional prosecution history of each is of little relevance to the other.Second, the examiner divided claims to a combination and a subcombination into distinct groups where (1) "the combination as claimed does not require the particulars of the subcombination as claimed for patentability" and (2) the subcombination has utility by itself. See, e.g., id. at 3. This is how the examiner distinguished groups II and III from group I and groups VII through XI from group VI. The examiner found the first requirement met whenever the subcombination's details appeared in a dependent claim. See id. Because all of the claims were assumed to be patentable, the presence of the subcombination in a dependent claim was evidence that the combination did not require the subcombination for the combination to be patentable. See id. To be clear, the examiner's restriction is not evidence that each separate group is supported by the specification and separately patentable. It is only evidence that the claims, as drafted, were directed at separate inventions. Accord Honeywell Int'l., Inc. v. ITT Industries, Inc., 452 F.3d 1312, 1319 (Fed. Cir. 2006) (assigning little weight to restriction requirement that did not construe claim terms or consider the specification).
2. Additional Restrictions
All of the patents-in-suit (and the vast majority of the Farmwald/Horowitz patent family) descend from Group I. Group I's prosecution history is rife with additional restrictions. One in particular is worth noting. On June 5, 1997, the examiner issued a restriction requirement for application no. 08/798,520 distinguishing claims to "a semiconductor device having at least one access-time register" and claims to "a memory device having a plurality of conductors being multiplexed for sequentially receiving an address." The examiner noted that the two groups of claims do not require each other and would require separate prior art searches. In response, Rambus elected to proceed with the claims to a semiconductor device with an access-time register.
II. PRIOR CLAIM CONSTRUCTION ORDERS
The Farmwald/Horowitz family of patents have been previously construed in various cases involving Rambus. Certain claim terms — "integrated circuit device," "read request," "write request," "transaction request" and "bus" — have been construed by the Federal Circuit. Rambus Inc. v. Infineon Techs. AG, 318 F.3d 1081, 1089-95 (Fed. Cir. 2003). This court construed additional terms in a previous case between Rambus and Hynix. Hynix Semiconductor Inc. v. Rambus Inc., C-00-20905-RMW, 2004 WL 2610012 (N.D. Cal. Nov. 15, 2004) ( Hynix I). The Manufacturers acknowledge these prior decisions, but argue about the amount of deference owed to them.
A. Stare Decisis and the Infineon Decision
In Markman v. Westview Instruments, Inc., the Supreme Court cited with approval authority that treated claim construction as a question of law, not fact, and held that claim construction is exclusively the responsibility of the court. 517 U.S. 370 (1996). The issue for the Court was whether the Seventh Amendment requires a jury to construe a patent. See id. at 372. Because history and precedent provided no clear answer, the Court examined "functional considerations" to determine the scope of the Seventh Amendment's right to a jury trial See id. at 388. The Court held that a judge, not a jury, would be better suited to interpret a patent as a written document, in part because jurors remain "unburdened by training in exegesis." Id. at 388-89. The Court also noted that a jury's ability to evaluate testimony would not be as helpful as a judge's ability to parse a complex document. Id. at 389-90.
The Markman Court buttressed its holding that functional considerations compel assigning claim construction to the court by emphasizing the "importance of uniformity in the treatment of a given patent." Id. at 390. The Court pointed out that issue preclusion cannot impose uniformity on infringement defendants who were not a party to a prior action. Id. at 391 (citing Blonder-Tongue Laboratories, Inc. v. University of Ill. Foundation, 402 U.S. 313 (1971)). After noting the difficulty of ensuring uniformity through issue preclusion, the Court stated that "treating interpretive issues as purely legal will promote (though it will not guarantee) intrajurisdictional certainty through the application of stare decisis on those questions not yet subject to interjurisdictional uniformity under the authority of the single appeals court." Id.
Since Markman, courts have treated this discussion as granting stare decisis effect to the Federal Circuit's claim constructions. E.g., Key Pharmaceuticals v. Hercon Laboratories Corp., 161 F.3d 709, 716 (Fed. Cir. 1998) ("We do not take our task lightly in this regard, as we recognize the national stare decisis effect that this court's decisions on claim construction have."); Hynix I, 2004 WL 2610012, at *4; Wang Laboratories, Inc. v. Oki Electric Industry Co., Ltd., 15 F. Supp. 2d 166, 176 (D. Mass. 1998). A district court must apply the Federal Circuit's claim construction even where a non-party to the initial litigation would like to present new arguments. E.g., Wang Laboratories, 15 F. Supp. 2d at 176 (applying the Federal Circuit's prior claim construction to a non-party whose motion to intervene as amicus at the Federal Circuit had been denied). Here Hynix, Micron, Nanya, and Samsung — none of whom were parties to the Infineon proceedings — ask this court to deviate from the Federal Circuit's claim construction in Infineon.
Hynix and Micron did submit an amicus brief in support of panel rehearing or rehearing en banc regarding the portion of the Infineon decision dealing with fraud. See Rambus Inc. v. Infineon Techs. AG, 2003 WL 24027551 (Fed. Cir. Mar. 7, 2003). This brief did not address the claim construction issues in the Infineon decision.
The Manufacturers do not dispute that stare decisis applies to a claim construction order generally. Instead, the Manufacturers suggest that a court may depart from binding precedent where the legal basis for the prior ruling has been sufficiently eroded. The Manufacturers quote the First Circuit for the principle that " stare decisis is neither a straightjacket nor an immutable rule; it leaves room for courts to balance their respect for precedent against insights gleaned from new developments, and to make informed judgments as to whether earlier decisions retain preclusive force." Carpenters Local Union No. 26 v. U.S. Fidelity Guar. Co., 215 F.3d 136, 142 (1st Cir. 2000). The Manufacturers argue that the Federal Circuit's recent cases on claim construction have altered the inquiry so much that this court should not follow Infineon. In urging this court to "not follow outdated decisions," the Manufacturers overlook a fundamental principle of stare decisis: only the court that issued the decision in question can elect not to follow it. See State Oil Co. v. Khan, 522 U.S. 3, 20 (1997) ("[I]t is this Court's prerogative alone to overrule one of its precedents."). The Federal Circuit recently illustrated this deference in Independent Ink, Inc. v. Illinois Tool Works, Inc., a case where the criticism of the existing precedent was overwhelming. See 396 F.3d 1342, 1349-51 (Fed. Cir. 2005), rev'd by Illinois Tool Works Inc. v. Independent Ink, Inc., 547 U.S. 28 (2006). Nonetheless, the Federal Circuit abided by "the duty of a court of appeals to follow the precedents of the Supreme Court until the Court itself chooses to expressly overrule them." Id. at 1351. This duty applies with equal force to this court and its relationship to the Federal Circuit.
The Manufacturers argue that the Phillips case has so altered claim construction that the Federal Circuit would have decided Infineon differently had it heard the case today. The court does not read Phillips as working so great a change. See Phillips v. AWH Corp., 415 F.3d 1303, 1330 (Fed. Cir. 2005) (Mayer, J., dissenting) ("Again today we vainly attempt to establish standards by which this court will interpret claims. But after proposing no fewer than seven questions, . . . we say nothing new, but merely restate what has become the practice over the last ten years."). Nonetheless, it cannot entertain that question because Phillips did not overrule Infineon or change the rules of claim construction such that Infineon's constructions need to be revisited. Accordingly, the court adopts the Federal Circuit's construction of "integrated circuit device," "read request," "write request," "transaction request" and "bus" from Infineon. For ease of reference, the court reproduces these constructions below: Claim term Construction Citation 318 F.3d at 1091 Id. Id. Id. Id.
"integrated circuit A circuit constructed on a single monolithic substrate, commonly called . device" a `chip.' "read request" A series of bits used to request a read of data from a memory device at 1093. where the request identifies what type of read to perform. "write request" A series of bits used to request a write of data to a memory device. at 1093. "transaction request" A series of bits used to request performance of a transaction with a at 1093. memory device. "bus" A set of signal lines to which a number of devices are connected, and at 1095. over which information is transferred between devices. In applying the Infineon claim construction, the court is sensitive to the fact that three of the Manufacturers have never been heard on this subject. Their arguments seeking to interpret terms in the claims such that the alleged inventions are limited to use with the inventors' narrow, multiplexed bus architecture has appeal, and if the court were construing the claim language on a clean slate, it might well so limit the claims. However, the slate is not clean. The Federal Circuit reviewed the intrinsic evidence and concluded that "[a]lthough some of Rambus's claimed inventions require a multiplexing bus, multiplexing is not a requirement in all of Rambus's claims." Infineon, 318 F.3d at 1094. This court cannot, no matter how appealing, contradict the Infineon claim construction.B. Stare Decisis and the Hynix I Decision
The parties next dispute how much deference this court owes to its prior claim construction in the earlier Hynix proceedings. The Markman Court commented that treating claim construction as a matter of law would "promote (though it will not guarantee) intrajurisdictional certainty through the application of stare decisis on those questions not yet subject to interjurisdictional uniformity under the authority of the single appeals court." 517 U.S. at 391. The Court's reference to promoting, but not guaranteeing, "intrajurisdictional certainty" suggests that a district court should follow another court's claim construction, though it does not have to.
Since Markman, various district courts have taken slightly different approaches to other courts' claim constructions, but despite the Court's suggestion, none has applied stare decisis. One court held that "considerable deference should be given to those prior decisions unless overruled or undermined by subsequent legal developments, including intervening case law." Sears Petroleum Transport Corp. v. Archer Daniels Midland Co., 2007 WL 2156251, at *8 (N.D.N.Y. 2007). It then proceeded to consider arguments that it had not heard during the prior claim construction. Id. at *12. Similarly, another court held that it would defer to its prior claim construction, but only "to the extent the parties do not raise new arguments." KX Industries, L.P. v. PUR Water Purification Products, Inc., 108 F. Supp. 2d 380, 387 (D. Del. 2000). Another court feared that refusing to consider a new party's claim construction arguments raised due process concerns and therefore granted the party's request for a Markman hearing. Texas Instruments, Inc. v. Linear Technologies Corp., 182 F. Supp. 2d 580, 589-90 (E.D. Tex. 2002). Meanwhile, courts in this district have been willing to consider a prior claim construction, but have stressed the importance of conducting an independent inquiry. Visto Corp. v. Sproqit Techs., Inc., 445 F. Supp. 2d 1104, 1108-09 (N.D. Cal. 2006) (Chen, M.J.); see also Townshend Intellectual Property, L.L.C. v. Broadcom Corp., 2008 WL 171039 (N.D. Cal. Jan. 18, 2008) (Fogel, J.) (modifying prior claim construction in light of a new party's arguments).
This general practice accords with the insight that a fresh look at a claim construction can hone a prior court's understanding and construction of a patent. See e.g., Finisar Corp. v. DirecTV Group, Inc., 523 F.3d 1323, 1329 (Fed. Cir. 2008). Indeed, the Federal Circuit has said that it "would be remiss to overlook another district court's construction of the same claim terms in the same patent as part of [a] separate appeal." Id. In Finisar, the Federal Circuit found a second district court's claim interpretation particularly helpful where it referred back to the prior construction and noted where it disagreed. Id. The lesson from Finisar is that additional litigation can refine and sharpen the courts' understanding of an invention and that a second court should not defer to a prior court's claim construction without questioning its accuracy.
On the other hand, this practice appears to conflict with the Supreme Court's understanding of stare decisis. The Supreme Court has stated that "[c]onsiderations in favor of stare decisis are at their acme in cases involving property and contract rights, where reliance interests are involved." Payne v. Tennessee, 501 U.S. 808, 828 (1991). As early as 1851, the Supreme Court explained in dictum that " stare decisis is the safe and established rule of judicial policy, and should always be adhered to" when dealing with cases establishing rules of property. The Genesse Chief, 53 U.S. 443, 458 (1851). The Court applied similar thinking in Minnesota Mining Co. v. National Mining Co., 70 U.S. 332 (1865), to reject a repeated attempt to litigate the possession of an estate. The Court reasoned that rules of property engender reliance, and that changing such rules damages any interests that have grown up based on those rules. Id. at 334. Contemplating the issue of mistaken decisions, the Court mused that "[d]oubtful questions on subjects of this nature, when once decided, should be considered no longer doubtful or subject to change." Id. Such firmness would discourage parties from "speculat[ing] on a change of the law" and free the courts from "the infliction of repeated arguments by obstinate litigants, challenging the justice of [the court's] well-considered and solemn judgments." Id.
" Stare decisis is usually the wise policy, because in most matters it is more important that the applicable rule of law be settled than that it be settled right." Burnet v. Coronado Oil Gas Co., 285 U.S. 393, 406 (1932) (Brandeis, J., dissenting). However, the prevailing notion among the district courts and the Finisar court is that it is better to get a claim construction right than it is to get a claim construction settled. Why so many courts have intuitively reached this conclusion is not immediately clear. It may stem, however, from the imperfect analogy between a patent right and other property rights, or from the difficulty of construing complex technical terms without the benefit of the arguments of parties not involved at the time of the first construction. In explaining when to apply stare decisis, Justice Brandeis stressed that it should apply "even where the error is a matter of serious concern, provided correction can be had by legislation." Burnet, 285 U.S. at 406. Unlike a common law rule of contract or property, a patent interpretation cannot be readily narrowed by a statute. It is doubtful Congress would take the time to do such a thing, and doing so could raise Fifth Amendment concerns. When approaching stare decisis from the perspective of what institution could best fix a mistaken interpretation, a court rather than the legislature is probably in a better position to do so.
The Supreme Court has emphasized that "legislatures may alter or change their laws, without injury, as they affect the future only; but where courts vacillate and overrule their own decisions on the construction of statutes affecting the title to real property, their decisions are retrospective and may affect titles purchased on the faith of their stability." Minnesota Mining Co., 70 U.S. at 334. This basis for stare decisis reflects concerns about society's reliance on a rule of law and the harm associated with upsetting society's expectations. See Payne, 501 U.S. at 828 (modulating the impact of stare decisis based on the weight of reliance interests). It is unclear how many people develop reliance interests in any given claim construction. At its grandest, a claim construction could shape an entire industry's research and development efforts (by leading people to avoid research in an area already patented) and licensing relationships (by leading people to take licenses or buy a favorably construed patent), as well as various company's stock prices (the patent holder's could increase on a favorable claim construction while an infringing firm's might collapse). But this is rather speculative, and even at its greatest, a claim construction does not engender the same scope of reliance on property rights that animated the Court's holding in United States v. Title Insurance Trust, 265 U.S. 472 (1924), where a reversal would have redistributed huge tracts of land overnight. While a change in the scope of a patent may dry up a royalty stream, it would not lead to ejectments and repossessions.
Reliance interests could be harmed if a patent was narrowly construed, industries developed, and then a later court broadened a claim to cover those investments. While this is a concern in patent law, see, e.g., Miller v. Bridgeport Brass Co., 104 U.S. 350 (1881), it does not apply here. A patent holder that suffered a narrow claim construction might appeal that claim construction, but would likely be estopped in the future from attempting to broaden its claims.
Thus, while "most" matters benefit from being settled rather than being settled right, claim construction appears to be an exception, at least as among district courts. Further, since the constructions in Hynix I are not final, the court is free to revisit them. Accordingly, this court will initially treat its prior construction as correct, but consider the Manufacturers' arguments as to why a construction in Hynix I should be modified.
C. Collateral Estoppel as to Hynix
Putting aside Micron, Nanya, and Samsung, Rambus argues that Hynix should be precluded from challenging the court's prior claim construction. Preclusion does not attach lightly. As stated by the Ninth Circuit, a court's decision only bars relitigation of an issue where:
(1) the issue necessarily decided at the previous proceeding is identical to the one which is sought to be relitigated;
(2) the first proceeding ended with a final judgment on the merits; and
(3) the party against whom collateral estoppel is asserted was a party or in privity with a party at the first proceeding.Hydranautics v. FilmTec Corp., 204 F.3d 880, 885 (9th Cir. 2000); see also Kourtis v. Cameron, 419 F.3d 989 (9th Cir. 2005).
Rambus points out that a court has precluded a party from relitigating claim construction in situations where there appears to have not been a final judgment. E.g., TM Patents, L.P. v. Int'l Bus. Machines Corp., 72 F. Supp. 2d 370, 376-80 (S.D.N.Y. 1999). The prior action in TM Patents ended during trial when the parties settled. Id. at 375. Nonetheless, the court in the second action precluded TM Patents from relitigating the prior court's claim construction because it held that the decision was sufficiently "final." Id. at 375-76 (citing Lummus Co. v. Commonwealth Oil Ref. Co., 297 F.2d 80, 89 (2d Cir. 1961)). Indeed, the law of issue preclusion uses an unusual definition of "final judgment," namely, any adjudication that is "sufficiently firm" to be accorded preclusive effect. Luben Industries, Inc. v. United States, 707 F.2d 1037, 1040 (9th Cir. 1983); see also Rest. 2d of Judgments § 13.
Still, the court is unpersuaded that its Hynix I claim construction order is "sufficiently firm" to grant it preclusive effect. There has not yet been a final judgment in the earlier proceedings, preventing Hynix from obtaining appellate review of this court's claim construction. The inability to obtain appellate review bars the application of issue preclusion. Matter of Lockard, 884 F.2d 1171, 1176 (9th Cir. 1989) (citing Rest. 2d of Judgment § 28(1)). While the claim construction in TM Patents also could not be appealed, that was caused by the patentee's decision to settle, not because the law prevented appellate review. See Rest. 2d of Judgments § 28, cmt. a (noting that 28(1) "applies only when review is precluded as a matter of law. It does not apply in cases where review is available but is not sought."). Because Hynix has not yet been able to exercise its right to appeal this court's prior claim construction, it cannot be precluded from continuing to litigate the scope of Rambus's patent claims.
III. CLAIM CONSTRUCTION
The parties dispute the construction of a number of terms and so focused their arguments on groups of terms. After briefly addressing the claim construction legal standard, the court's inquiry follows the parties' groupings of the disputes.
A. Claim Construction Legal Standard
The court begins its claim construction inquiry by considering the ordinary and customary meaning of the terms in the claims from the perspective of a person of ordinary skill in the art. Phillips, 415 F.3d at 1312-13. This ordinary meaning is influenced by the use of language in the asserted and unasserted claims in the patents. Id. at 1314-15. For example, the phrase "steel baffles" implies that a "baffle" need not be made of steel. See id. at 1314. More generally, the appearance of a limitation in a dependent claim creates a presumption that the independent claim lacks that limitation. Id. at 1315; but see Curtiss-Wright Flow Control Corp. v. Velan, Inc., 438 F.3d 1374, 1380-81 (Fed. Cir. 2006) (discussing limits of claim differentiation).
This "objective baseline" of ordinary meaning is also formed in part by the patents' specification's use of the claim language because a person of ordinary skill understands a claim's language in the context of the entire patent. Id. at 1313. For example, the specification is often helpful in defining the scope of technical terms. See id. at 1315. The specification can also reveal a "special definition" of a claim term or demonstrate "an intentional disclaimer" or "disavowal" of claim scope. Id. at 1316. The specification's statutory role in determining a claim's validity justifies placing this heavy weight on it. Id. at 1315-16.
Similarly, the prosecution history of a patent can inform the proper construction of a patent's claims. Id. at 1317. It provides evidence of how both the Patent Office and the inventor understood the claims. Id. The prosecution history's back-and-forth nature often makes it unclear, however, and therefore it is rarely as helpful in construing a patent's claims as the claims themselves and the specification. Id. Still, the prosecution history can be invaluable for demonstrating the inventor's understanding of the claims and checking "whether the inventor limited the invention in the course of prosecution, making the claim scope narrower than it would otherwise be." Id.
Finally, a court may consider extrinsic evidence like inventor and expert testimony, dictionaries, and treatises. Id. at 1318. Such evidence can be helpful, but courts must be mindful that it is not necessarily created contemporaneously with the patent application and that the litigation context can produce biased evidence. Id. Accordingly, extrinsic evidence is rarely more probative of the proper construction of the claims than the intrinsic record, id. at 1319, but extrinsic evidence is particularly helpful "[w]hen the intrinsic evidence is silent as to the plain meaning of a term." Helmsderfer v. Bobrick Washroom Equipment, Inc., 527 F.3d 1379, 1381-83 (Fed. Cir. 2008) (construing a claim term in light of three dictionary definitions); see also Phillips, 415 F.3d at 1322-23. Nonetheless, to avoid "the risk of systematic overbreadth" the court begins its inquiry focused "on how the patentee used the claim term in the claims, specification, and prosecution history, rather than starting with a broad definition and whittling it down." Phillips, 415 F.3d at 1321.
All four sources of evidence discussed above have their advantages and drawbacks and varying amounts of persuasive force. All of the sources are appropriately considered, and contrary to Samsung's argument at the claim construction hearing, there is no "rigid algorithm" or "sequence of steps" which the court must follow. Id. at 1324. All that matters "is for the court to attach the appropriate weight to be assigned to those sources in light of the statutes and policies that inform patent law." Id.
B. The "Device" Terms and Multiplexing
The first dispute between the parties is whether Rambus's claims should be limited to DRAMs that use multiplexed bus data lines. This dispute touches on the construction of "bus" from Infineon and influences the construction of "device" and variants of "device." Rambus argues that there is no need to construe the term "device" in isolation, but that the court should adopt the Infineon court's construction of "integrated circuit device" and the construction of "memory device" from Hynix I. The Manufacturers proffer a detailed and limited construction of the term "device," which they then incorporate into their proposed constructions of "integrated circuit device" and "memory device." The proposed constructions appear below: Claim term: Rambus's construction: The Manufacturers' Construction:
"device" (no separate construction) Electronic circuits or components with a bus interface that is adapted to connect to a multiplexed set of signal lines, each for carrying address, data, and control information, that has substantially fewer connections to signal lines than the number of bits in a single address, and that does not connect to a separate device select line. "integrated A circuit constructed on a single A device that includes one or more integrated circuits. circuit device" monolithic substrate, commonly called a "chip." "memory An integrated circuit device in A device in which information can be stored and retrieved device" which information can be stored electronically. and retrieved electronically. 1. "Device"At bottom, the Manufacturers' proposed construction of "device" is an attempt to evade the Federal Circuit's Infineon construction of the terms "integrated circuit device" and "bus." The Manufacturers' proposed construction of device, when incorporated into the phrase integrated circuit device, conflicts with the Federal Circuit's construction of that phrase, which lacks the narrow, multiplexed bus limitation of the Manufacturers' proffered construction. See Infineon, 318 F.3d at 1089-91. The Manufacturer's construction of "device" also imports a multiplexing limitation into every claim, an interpretation the Federal Circuit already considered and rejected when it construed the term "bus." Id. at 1094-95. The court also rejected any attempt to read a multiplexing limitation into other claim terms when it held that "multiplexing, if necessitated by the claims, is applicable to the construction of the term `bus,' not `read request.'" Id. at 1093. Hynix previously made this argument regarding the construction of "device," and the court rejected it because "[t]his end result, a multiplexed bus as part of the claim limitations, was rejected by the Federal Circuit." Hynix I, 2004 WL 2610012, at *6.
The Manufacturers argue that the court should ignore the Federal Circuit's interpretation of "bus" because the claims-in-suit do not use the term. From this, the Manufacturers argue that the Federal Circuit's interpretation of "bus" should not influence the interpretation of "device" because "bus" is not in dispute. Even if the term "bus" were not in dispute here, the court cannot ignore the Federal Circuit's construction of the term because this court cannot construe a claim to blatantly conflict with the Federal Circuit's prior interpretation. See Hynix I, 2004 WL 2610012, at *6.
Rambus requests that the court construe "bus," and construe it in light of Infineon, because the term appears in the Manufacturers' proposed constructions. Hence, while the claims do not recite the term "bus," the term has been put in dispute by the Manufacturers' use of the term in their proposed constructions.
Nor is the court necessarily persuaded that the Manufacturers' construction of "device" is correct, even if the court could ignore the Infineon claim construction. The Manufacturers' construction of "device" would import a host of limitations, specifically that the device (1) be adapted to connect to a multiplexed set of signal lines, (2) have substantially fewer connections to signal lines than the number of bits in a single address, and (3) lack a separate device select line. That a person of ordinary skill in the art would read the term "device" and necessarily understand the term to have all of these limitations is questionable.
The court does note that the specification is dense with discussion of the advantages of the multiplexed bus architecture at the heart of Drs. Farmwald and Horowitz's invention. The most compelling piece of evidence presented by the Manufacturers is that the detailed description of the invention begins with "the present invention is designed to provide a high speed, multiplexing bus for communication between processing devices and memory devices and to provide devices adapted for use in the bus system." `184 patent, col. 5, ll. 29-31. Furthermore, the specification's defined objectives include "provid[ing] devices, especially DRAMs, suitable for use with the bus architecture of the invention." Id., col. 3, ll. 46-48. The Manufacturers therefore argue that the specification necessarily limits the claimed devices to ones "adapted for use in the bus system."
But these modifications do not all require the multiplexed bus architecture. For example, programmable registers can contain device identification information, device-type information, and access-time settings, among other things. See id., col. 6, ll. 28-37. To be sure, a register for device information is unnecessary where the bus architecture uses device-select lines. However, nothing about the use of an access-time register hinges on the use of a multiplexed bus architecture. While using programmable delay times is useful in the multiplexed bus environment, it is not limited to that environment. This fact is reflected in the prosecution history. As discussed above, at one juncture the examiner required Rambus to split its claims because the use of access-time registers did not require a multiplexed bus, and vice versa. Accord Infineon, 318 F.3d at 1095. To construe "device" to require the use of a multiplexed bus (among other things) would run counter to the Patent Office and the inventors' disclosed understandings of the claims.
2. "Integrated circuit device"
Putting aside the dispute over multiplexing, the Manufacturers argue that an "integrated circuit device" should be construed to include "one or more integrated circuits," while Rambus argues that the court should adopt the Federal Circuit's narrower Infineon construction that limited an "integrated circuit device" to a circuit built on a single chip.
It is unclear why this claim construction dispute must be resolved. The joint case management order in this case set a deadline for filing summary judgment motions that depend on claim construction issues. See C-05-00334-RMW, Docket No. 174, at Attachment C (N.D. Cal. Apr. 24, 2007). The Manufacturers have moved for summary judgment of non-infringement based on the court's adoption of their various claim constructions, and have also moved for summary judgment of invalidity based on the court's adoption of Rambus's proposed claim constructions. However, the Manufacturers' bring no motion that adoption of any of their proposed claim constructions would compel summary judgment of invalidity.
This prevents the court from fully grasping the nature of the claim construction dispute. For example, the Manufacturers' motion suggests that construing the term "device" to include a multiplexing limitation might support a summary judgment of non-infringement. But their motion does not advance a non-infringement argument based on the "one or more chips" construction of "integrated circuit device." Indeed, it would be hard to imagine how requesting a broader claim construction than that of the patent holder could lead to a finding of non-infringement. This leaves the court to speculate as to why this dispute is material. Either this dispute is about nothing, or the Manufacturers are suggesting that the court construe a claim so broadly that it is invalid. This uncertainty defeats the purpose of the court's case management order.
That aside, the court begins with the Federal Circuit's claim construction from Infineon holding that an integrated circuit device is limited to a single chip. In reaching its decision, the Federal Circuit considered the lower court's claim construction, Rambus's position, and the ordinary meaning of the term "integrated circuit" based on two computing dictionaries. See Infineon, 318 F.3d at 1091. The Manufacturers now argue that while the Federal Circuit construed an "integrated circuit device" as being limited to a single chip, it did not explicitly consider whether the claim must be limited to a single chip or could encompass multiple devices. The Manufacturers are correct; Infineon does not appear to have argued that an "integrated circuit device" could be more than one integrated circuit. Instead, it argued that an "integrated circuit device" had to contain three types of circuitry based on Rambus's specification. See Infineon, 318 F.3d at 1089-90. Rambus meanwhile advocated its view of the ordinary meaning of the phrase.
The Manufacturers also point out that the Federal Circuit relied on definitions of integrated circuit, while Rambus's claims refer to an integrated circuit device. The Manufacturers argue that adding "device" means the phrase is broader than just one integrated circuit, and therefore the phrase should be construed as comprising one or more integrated circuits. The Manufacturers also identify an instance in the prosecution history of Rambus's claims where it referred to "a single semiconductor substrate" and argue that this distinction shows that an "integrated circuit device" is broader than a single chip. In reply, Rambus characterizes the Manufacturers' construction as "absurdly broad" such that "many modern washing machines would be `integrated circuit devices.'" Rambus also argues that the Manufacturers do not explain why "integrated circuit device" should be construed more broadly than "integrated circuit." On the other hand, Rambus does not credibly explain why "integrated circuit device" should not be construed more broadly than "integrated circuit." Nor does Rambus point to anything in the claim language, specification, or prosecution history to support its argument that an "integrated circuit device" must necessarily exist on a single chip.
While the Manufacturers' argument has some appeal, the court is bound by the claim construction in Infineon. Stare decisis would be largely meaningless if a lower court could change an appellate court's interpretation of the law based only on a new argument. Here, the Federal Circuit has definitively interpreted the phrase "integrated circuit device" in the context of the Farmwald/Horowitz patent family as "a circuit constructed on a single monolithic substrate, commonly called a `chip.'" This court may not revisit it.
3. "Memory device"
In Hynix I, this court interpreted a "synchronous memory device" as "a memory device that receives an external clock signal which governs the timing of the response to a transaction request." Hynix I, 2004 WL 2610012, at *6-*8. The court's analysis focused on the "synchronous" aspect of the phrase and did not independently focus on the phrase "memory device." The reason is that the term "memory device" was not contested in Hynix I — the parties agreed that it meant "an integrated circuit device in which information can be stored and retrieved electronically." See C-00-20905, Docket No. 326, at 1 (N.D. Cal. Sept. 15, 2003).
Rambus emphasizes repeatedly that it is proposing the same construction that it and Hynix agreed to in 2003. The Manufacturers propose that a "memory device" is simply "a device in which information can be stored and retrieved electronically." The difference is merely whether a "memory device" incorporates the definition of an "integrated circuit device," and with it, the single chip limitation from Infineon.
It must be noted that since Hynix and Rambus disagreed about the meaning of "integrated circuit device," which was embedded within their definition of "memory device," the parties did not really agree on the meaning of "memory device" — they only agreed that the definition turned on the definition of "integrated circuit device."
Rambus points to a number of illustrative references to a memory device in the patents' common specification. First, Rambus notes that the specification refers to DRAMs, SRAMs, and ROM devices as "memory devices" and assumes that these memory devices are confined to a single chip. `184 patent, col. 1, ll. 49-56. Next, Rambus shows that "one object of the present invention is to use a new bus interface built into semiconductor devices to support high-speed access to large blocks of data from a single memory device by an external user of the data, such as a microprocessor, in an efficient and cost-effective manner." Id., col. 3, ll. 23-27. It is unclear why this sentence would compel a person of ordinary skill in the art to interpret a "memory device" as residing on a single chip. Somewhat more helpfully, Rambus shows that in the summary of invention, Rambus described the present invention as "a memory subsystem comprising at least two semiconductor devices, including at least one memory device, connected in parallel to a bus. . . ." Id., col. 3, ll. 51-61. Another portion of the specification describes how the DRAMs claimed in the patent differ from other DRAMs, namely, because they include "registers . . . which may store control information, device identification, device-type and other information appropriate for the chip such as the address range for each independent portion of the device." Id., col. 4; ll. 22-27 (emphasis added). This sentence suggests that a DRAM must exist on a single chip, but it is not clear that it requires a "memory device" to be a single chip. Rambus also refers to a paragraph of the detailed description where the patent states that "the bus architecture of this invention connects master or bus controller devices, such as CPUs, Direct Memory Access devices (DMAs) or Floating Point Units (FPUs), and slave devices, such as DRAM, SRAM, or ROM memory devices." Id., col. 6, ll. 12-17. The description reinforces the linkage between "memory devices" and chips likes DRAMs when it states that "a memory device will typically have only slave functions, while a DMA controller, disk controller or CPU may include both slave and master functions." Id., col. 6, ll. 20-23. Lastly, the patents' drawings also support Rambus's argument; for example, figure 8a shows two DRAMs attached to a bus and refers to them as "chips."
As Rambus often points out, a court must "avoid importing limitations from the specification into the claims." Phillips, 415 F.3d at 1323. However, where a specification clearly states the scope of an invention and the scope of the invention is "described as the advantage and distinction of the invention," the specification may narrow the ordinary scope of the claims. E.g., On Demand Machine Corp. v. Ingram Indus., Inc., 442 F.3d 1331, 1340 (Fed. Cir. 2006). The On Demand case is illustrative of how explicitly a specification must define the scope of the invention for the specification to limit the scope of the claims. For example, the patentee in On Demand clearly distinguished its customer-oriented book printing machine from "general purpose machines." Id. at 1340. From this, the Federal Circuit held that the claimed machine that enabled a customer to select a book to print on demand required a physically present consumer instead of a remote purchaser.
Rambus's references to the specification show that a DRAM, SRAM or ROM may be a memory device, but the specification does not clearly limit the term "memory device" to a single chip. Because the specification does not clearly limit the scope of the invention to a single chip, the court declines to read the phrase "memory device" so narrowly. Accordingly, the court adopts the Manufacturer's construction. A "memory device" is "a device in which information can be stored and retrieved electronically." It need not be on a single chip.
Similar to its arguments about the need to limit "integrated circuit device" to not read on a washing machine, Rambus argues that the Manufacturers' construction of "memory device" could cover a tape recorder because it is a "device in which information can be stored and retrieved electronically." Rambus's hyperbolic argument ignores that its claims contain more limitations than just "a memory device." Putting aside other limitations, the claims often (but not always) modify the phrase "memory device." For example, Rambus asserts claim 43 of U.S. Pat. No. 6,314,051. That claim specifically limits itself to "[t]he memory device of claim 34 wherein the memory device is a synchronous dynamic random access memory." Rambus also asserts claims 1 and 34 of U.S. Pat. No. 6,584,037. These claims are drawn to "a method of operation of a synchronous memory device" and "a method of operation of a synchronous dynamic random access memory device." These additional limits on the scope of "memory device" ameliorate fears that Rambus's claims could read on a "tape recorder," while also demonstrating that Rambus knew how to limit its claims to a single chip when it wished to do so. See Curtiss-Wright Flow Control Corp. v. Velan, Inc., 438 F.3d 1374, 1380-81 (Fed. Cir. 2007); cf. Mark A. Lemley, The Limits of Claim Differentiation, 22 Berkeley Tech. L. J. 1389, 1396 (2007).
C. The "Transaction" Terms and the Packet-Based Protocol
The next significant dispute between the parties turns on whether Rambus's claims extend to DRAMs that do not use a packet-based protocol for transmitting information. This dispute influences the construction of "read request," "write request," "operation code," "block size information," and a host of additional terms incorporating these or similar terms.
Like the dispute over the "device" terms, the "transaction" terms appear throughout Rambus's claims. Also like the "device" terms, the Federal Circuit has already construed some of the terms at issue. As these interpretations control this court's decision making, the inquiry begins with them.
1. "Read Request," "Write Request," and "Transaction Request" Claim term: Rambus's Construction: The Manufacturers' Construction:
"write request" A series of bits used to request a One or more bits in a write request packet to request a write write of data to a memory device of data to a memory device, where the write request packet where the request identifies what identifies the type of write to perform. type of write to perform. Rambus's construction of write request is illustrative of its constructions of "read request" and "transaction request," all of which reflect the Federal Circuit's construction of those terms. Infineon, 318 F.3d at 1093. The Manufacturers' construction is essentially identical, except it requires that a request be contained in a request packet.Again, the court begins with the Federal Circuit's construction in Infineon. There, Infineon argued that a "read request" must also include address and control information. Id. The Federal Circuit characterized Rambus as arguing that "address and control information is part of the `request packet,' not the `read request'" and that the read request was simply "an instruction to the memory device to perform a read action." Id. The Federal Circuit's opinion accurately summarizes Rambus's argument then. Rambus characterized the four bits AccessType[0:3] as the "read request." See Brief of Plaintiff-Appellant Rambus Inc., Rambus Inc. v. Infineon Techs. AG, 2001 WL 34773495, *17 (Fed. Cir. Nov. 6, 2001) ("Specifically, the `read request' consists of 4 digital `bits,' called the AccessType field, which are contained within the larger `request packet,' as shown in Fig. 4."). So there can be no mistake, Rambus's brief repeated that "a `read request' (and, correspondingly, a `write request') is determined by and consists of the AccessType field, 4 digital bits of control information contained within the `request packet.'" Id. (emphasis in original). Nowhere in Rambus's appellate briefing did Rambus argue that a read request was not part of a request packet. See generally id. at *15-*19; Reply Brief for Rambus Inc. as Plaintiff-Appellant, 2002 WL 32817343, *16-*18 (Fed. Cir. Mar. 7, 2002). Instead, Rambus (and as shown above, the specification) consistently referred to the "request" as the command contained in AccessType[0:3], the second through fifth bits of Rambus's preferred embodiment of the request packet.
The Federal Circuit nonetheless construed "read request" more broadly than Rambus proposed. It held that a "read request" was "a series of bits used to request a read of data from a memory device where the request identifies what type of read to perform." Infineon, 318 F.3d at 1093. In so doing, the court seems to have rejected the argument that a read request must be contained within a request packet. Id. The court quoted two dependent claims from U.S. Patent No. 6,034,918 that recited the need to include a read request and block size information "in a request packet" and then "in the same request packet." Id. at 1093. The court then criticized the district court's interpretation (and Infineon's) for being irreconcilable with those two claims. Id.
These claims were not cited to the court in Rambus's appellate briefing. See generally id. at *15-*19; Reply Brief for Rambus Inc. as Plaintiff-Appellant, 2002 WL 32817343, *16-*18 (Fed. Cir. Mar. 7, 2002).
The Manufacturers acknowledge the Infineon decision, but argue that their construction requiring a request to be inside a request packet is consistent with it. The Manufacturers point out that Infineon rejects the notion that a request must include address and control information, id. at 1092, but that Infineon does not foreclose the construction they now proffer. The court disagrees. For better or for worse, the Infineon decision is quite clear. Its claim differentiation reasoning rejects the Manufacturers' proposed construction requiring that a request be sent in a request packet. The court therefore adheres to the construction of "read request," "write request" and "transaction request" from Infineon.
2. Additional "Transaction" Terms
The parties also dispute the proper construction of various "transaction" terms that were not construed in Infineon, but the disputes largely turn on the same issue: whether or not the "transaction" requires the use of a request packet. The parties use the term "operation code" to illustrate their dispute.
The other "transaction" terms with similar construction disputes are: block size information, control information, precharge information, address information, read operation, write operation, request for a write operation and set register request.
Rambus construes an operation code as "one or more bits to specify a type of action." The Manufacturers construe operation code as "one or more bits in a request packet used to specify a type of action to be performed." Semantics aside, the substantive difference in the proposed construction is whether an operation code must be sent in a request packet.
In prior proceedings, this court construed "operation code" as "one or more bits to specify a type of action," which is now Rambus's proposed construction. See Hynix I, 2004 WL 2610012, at *9-*13. The Manufacturers argue that this court erred in relying on the dictionary definition to determine the ordinary meaning of "operation code." The court's analysis did cite the definition of "operation code" contained in The Authoritative Dictionary of IEEE Standard Terms. See id. at *11 (citing Tex. Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed. Cir. 2002)).
And had the court stopped there, its prior construction might have run afoul of the Federal Circuit's approach to claim construction. See Phillips, 415 F.3d at 1320-23. But the court did not stop with the ordinary meaning conveyed by the dictionary. The court then construed "operation code" in light of various dependent claims that recited the use of the operation code as part of a request packet. See Hynix I, 2004 WL 2610012, at *9-*13. To be sure, the court also considered the specification, which states that "the first byte contains two 4 bit fields containing control information, AccessType[0:3], an op code (operation code), which, for example, specifies the type of access[.]" `184 patent, col. 9, ll. 39-41. This demonstrates that AccessType[0:3] — a portion of Rambus's preferred packet embodiment — is an operation code. It does not, however, foreclose the possibility that an operation code could exist outside of a packet. Hence, the specification's ambivalence and the explicit structure of the dependent claims confirm that the phrase "operation code" receives its ordinary meaning and is not limited to being sent as part of a request packet.
The Manufacturers argue that the court should not rely on claim differentiation because Rambus's original claims in 1990 did not include dependent claims explicitly claiming the use of an operation code as part of a request packet, and therefore the current structure of Rambus's claims should be "ignored as a litigation-driven attempt to broaden the scope of the Farmwald patents beyond the scope of the invention set forth in the specifications." The Manufacturers cite no support for this argument. Perhaps the Federal Circuit's concerns about expert testimony — that it is "generated at the time of and for the purpose of litigation and thus can suffer from bias that is not present in intrinsic evidence" — should also reach intrinsic evidence that is generated as litigation occurs. See Phillips, 415 F.3d at 1318. On the other hand, the only value of a patent claim is to exclude, and it is naive to think that a patent is prosecuted without an eye to litigation. Absent more compelling support, the court cannot accept the Manufacturers' invitation to ignore intrinsic evidence because it might have been "litigation-driven."
It is also worth noting that an "operation code" specifies an action generically. For example, an operation code like AccessType[0:3] specifies a read request, a write request, or certain other possible operations. See '184 patent, col. 9, l. 47 — col. 10, l. 14 (explaining additional possible operations, like control register access, which permits the device's control register to be reprogrammed). In other words, an operation code includes the concepts of a read request or write request. It would be inconsistent with Infineon for this court to hold that a read or write request need not be sent in a packet, but that the general family of requests (operation codes) must all be sent in a packet. Accordingly, the court adopts the constructions for the "transaction" terms that do not include a "request packet" limitation, consistent with its claim construction order in Hynix I and the reasoning of Infineon. 3. "Data" Claim term: Rambus's construction: The Manufacturers' Construction:
"data" Plain meaning. No construction is One or more bits written to/read from the memory array. necessary. Alternatively, if Court deems construction necessary, "one or more bits representing information." Apart from the dispute over importing a "request packet" limitation into various transaction terms, the parties dispute the construction of the term "data" in various limitations dealing with transactions within the memory device. The dispute turns on whether "data" refers to any bit of information, or whether "data" refers only to bits of information written to or read from a memory array. The court previously declined to construe the term "data," see Hynix I, 2004 WL 2610012, at *20, and it is not clear why the term must be construed here as no motion for summary judgment of non-infringement or invalidity turns on the construction of the term "data."The Manufacturers base their argument on the specification's summary of invention, where Rambus describes the invention as ". . . includ[ing] a plurality of bus lines for carrying substantially all address, data, and control information needed by said memory devices . . ." `184 Patent, col. 3, ll. 54-56. The Manufacturers argue from this that Rambus explicitly distinguished "data" information from "address" or "control" information. The Manufacturers claim that Rambus "consistently and repeatedly" made this distinction, though they supply no additional citations to the specification to support this assertion.
Meanwhile, Rambus argues that data does not need construction or should be construed to mean "one or more bits representing information." Rambus emphasizes that in discussing a preferred implementation, the specification describes a protocol for accessing and reprogramming a control register. See id., col. 9, l. 66 — col. 10, l. 14. In that situation, the request packet's AccessType field indicates that the master device seeks access to the control register, the Address field then specifies the address of the control register, and some of the Address field can be used to "represent or include data to be loaded into that control register." Id., col. 10, ll. 6-8. Rambus argues that since a control register is not a memory array, the Manufacturers' definition of "data" is too narrow because the specification describes writing "data" to a control register.
It is clear that the specification distinguishes three types of information in the context of the multiplexed bus architecture: address, control and data. Beyond the single example cited by the Manufacturers, the specification makes clear that in a preferred implementation of Rambus's new bus, "8 bus data lines and Addressvalid bus line carry address, data, and control information for memory addresses up to 40 bits wide." Id., col. 4, ll. 2-5. This sentence in the specification shows the impossibility of construing the term "data" outside of the context of the claims. The bus is composed of eight data lines that carry three different types of information. Where "data" is used as an adjective to modify a bus line, only Rambus's proposed construction makes sense. But in discussing the types of information carried over the bus's data lines, the specification distinguishes between three kinds: address information, control information, and data information. Where "data" is used as an adjective to modify "information," the Manufacturers' construction makes sense. Finally, the specification refers to writing "data" to a control register. Here, the specification uses "data" as a noun, and the Manufacturers' construction cannot be reconciled with the fact that a control register is not a memory array. As shown, the word "data" takes on differing meanings in the Farmwald/Horowitz specification, and the court does not believe one construction of the term is necessarily appropriate.
Accordingly, the court turns to the claims, where the term "data" appears dozens of times. As used in the claims, "data" most often appears as a noun, not as the adjective that often appears in the specification. For example, Claim 14 of the `184 patent uses the term "data" to refer to information to be written to a memory array. See '184 patent, col. 25, l. 57 — col. 26, l. 6. The claim covers a method of operating a memory device including "receiving first block size information from a master, wherein the first block size information defines a first amount of data to be sampled by the memory device in response to a write request" and "sampling a first portion of the first amount of data synchronously." In this method claim, the memory device receives a write request, and the block size information defines the amount of data for the memory device to receive in response to that write request. In response to that information, the memory device then samples the bus to obtain data to write to its memory array. Here, the Manufacturers' construction is the most precise (though Rambus's is still accurate). Similarly, in U.S. Patent 6,715,020, claim 1 recites a controller device featuring "first output driver circuitry to output block size information to the memory device, wherein the block size information defines an amount of data to be output by the memory device[.]" See col. 24, ll. 57-65. Here, a controller transmits block size information to the memory device telling the memory device how much data to output. This use of data refers to information to be read from the memory array.
On the other hand, U.S. Patent No. 6,314,051's claim 27 demonstrates the shifting contextual uses of the term "data." It recites a memory device that includes "a programmable register to store a value which is representative of a number of clock cycles of the external clock signal to transpire before sampling a first portion of data, wherein the first portion of data is sampled in response to an operation code" and " data input receiver circuitry to sample the first portion of data synchronously with respect to the external clock signal." In this claim, "data" appears in both its noun and adjective forms. To be sure, where it is used as a noun, it refers to data to be sampled from the bus after a set time period. Where it modifies "input receiver," however, its meaning seems ambiguous. In the preferred embodiment, the bus is multiplexed and the data lines carry address, data, and control signals to the input receivers. Thus, the input receivers receive more than just data information to be written to the memory array, they also receive address and control information. In this claim, the "data input receiver circuitry" samples data information, but it should not be construed to sample only data information. Instead, a person of ordinary skill in the art would recognize that "data input receiver circuitry" could sample any type of information from the bus.
That aside, the asserted claims predominantly use the term "data" as a noun and in those contexts, the Manufacturers' construction seems more precise. Accordingly, the court adopts the Manufacturers' construction of "data," namely, as "one or more bits written to/read from the memory array" where the term appears as a noun. The court is mindful, however, that it is being asked to interpret "data" without the context of any invalidity or non-infringement arguments. In the event that this construction of "data" conflicts with the context of a specific claim, the court will revisit the issue.
C. The "Clocking" Terms
The next cluster of disputes between the parties involves the scope of claims related to the clocking features of Rambus's inventions. This dispute influences the construction of "internal clock signal," "external clock signal," "sample" and various additional terms incorporating these. Like the dispute over the "device" and "transaction" terms, the "clocking" terms appear throughout Rambus's claims. Unlike the "device" and "transaction" families of terms, the Federal Circuit has not construed the "clocking" terms. The only prior construction of some of these terms was rendered by this court in Hynix I. See 2004 WL 2610012, at *20-*21 (construing "first external clock" and "second external clock"). The dispute largely turns on the extent to which the clocking scheme described in the Farmwald/Horowitz specification limits the scope of Rambus's claims.
1. The Manufacturers' Compliance with the Patent Local Rules
As a preliminary matter, Rambus points out that the Manufacturers' proposed constructions of various clocking terms are not the constructions they proposed in their joint claim construction statement. Indeed, the Manufacturers' revisions to their constructions of "external clock signal" and "sample" trickle down and affect their construction of 42 of the 72 disputed terms in this case. Rambus therefore requests this court to strike the entirety of the Manufacturers' responsive brief on claim construction. Meanwhile, the Manufacturers claim that they "revised some proposed constructions . . . in an effort to narrow the issues in dispute and also for clarity."
For example, the Manufacturers had previously submitted that an "external clock signal" was "a continuously periodic signal from a source external to the device, which travels along a single bus clock line where the signal is sampled by the device at two separate points along the bus clock line." See, e.g., Joint Claim Construction Statement, C-05-00334-RMW, Appendix A at 15 (N.D. Cal. Jul. 11, 2007). The Manufacturers now argue that an "external clock signal" is properly construed as "a continuously periodic signal, from a source external to the device that provides an early clock and a late clock from which timing information can be derived."
The Patent Local Rules in effect at the time the Manufacturers changed their proposed claim constructions do not explicitly forbid this shift. See generally Patent Local Rule 4, compare with Patent Local Rule 3-7 (requiring a showing of good cause and an order of the court to amend or modify infringement or invalidity contentions). However, such changes in position violate the spirt of the applicable Rule 4. Rule 4-2 requires the parties to exchange preliminary claim construction proposals, as well as the basis for those proposals. Rule 4-3 then requires the parties to jointly file a claim construction and prehearing statement containing a list of agreed-upon constructions, a list of the parties' proposed claim constructions where they disagree, a disclosure of the evidence supporting those proposals, and additional information necessary for conducting the claim construction hearing. Rule 4-4 explicitly closes discovery with respect to claim construction thirty days after the filing of the joint statement. Rule 4-5 then sets a schedule for filing claim construction briefs once discovery has closed.
The Patent Local Rules were revised effective March 1, 2008. The rules discussed here have not significantly changed, though Rule 3-7 regarding amendments is now Rule 3-6.
While no provision of the Patent Local Rules explicitly forbids a party from shifting its claim construction position, the rules strongly suggest that a party is not supposed to do so. Rule 4-4 is designed to permit the parties to conduct discovery in a timely and orderly manner. It therefore closes discovery with respect to claim construction thirty days after the submission of the joint claim construction statement. One clear inequity of the Manufacturers' shift in position is that Rambus was not able to depose the Manufacturers' expert on their new proposed claim constructions prior to filing its opening and reply briefs.
The Northern District's jurisprudence interpreting the Patent Local Rules has been clear regarding their purpose. "The rules are designed to require parties to crystallize their theories of the case early in the litigation and to adhere to those theories once they have been disclosed." Nova Measuring Instruments Ltd. v. Nanometrics, Inc., 417 F. Supp. 2d 1121, 1123 (N.D. Cal. 2006) (discussing infringement contentions). In preventing a party from submitting new infringement contentions after a claim construction ruling, the court explained that "infringement cases would fall prey to a vexatious shuffling of positions-a kind of legal musical chairs serving no purpose other than to entertain highly paid lawyers and to thwart the very intention behind the patent local rules." Atmel Corp. v. Information Storage Devices Inc., 1998 WL 775115, *2 (N.D. Cal. 1998). The court further noted that the local rules are "designed to prevent the `shifting sands' approach to claim construction." Id. at *3.
In addition to being stymied in discovery, Rambus also lost the opportunity to use their opening brief to criticize the Manufacturers' new construction of "external clock signal" and "sample." Indeed, in its opening brief, Rambus pointed out that Manufacturers' original construction of "external clock signal" used the term "sample," while their construction of "sample" included the term "external clock signal." Rambus argued that because the Manufacturers' proposed constructions were circular, they should be rejected. The Manufacturers' revision of their claim construction proposals to fix this circularity appears opportunistic.
The Northern District's jurisprudence does not instruct on how to remedy a shift in claim construction positions after the parties have filed their joint claim construction statement. The nearest case on point involved two parties that had agreed not to construe a claim term, thus granting it a very broad meaning. Ultratech, Inc. v. Tamarack Scientific Co., 2005 WL 856408, *2 (N.D. Cal. 2005). The defendant therefore conceded infringement. Id. After securing a summary judgment of infringement, the patent holder then appears to have argued that the claim term should be construed more narrowly to avoid the defendant's argument that the claim was invalid based on prior art. Id. at *2-*3. In that case, Magistrate Judge Larson permitted the defendant to submit an expert report addressing the new claim construction after discovery had closed. Id. at *3. The opinion does not say whether the court later set aside the summary judgment of infringement based on the prior claim construction.
Rambus's proposed remedy — striking the Manufacturers' claim construction brief — is too harsh. Nor is the court willing to ignore the Manufacturers' arguments if they help the court to construe the claims in dispute. Yet the Manufacturers' conduct is not conducive to the orderly progress of this case, and the court disapproves of it.
2. "External Clock Signal" Claim term: Rambus's Construction: The Manufacturers' Construction:
"external clock A periodic signal from a source A continuously periodic signal, from a source external to [signal]" external to the device to provide the device that provides an early clock and a late clock from timing information. which timing information can be derived. Turning to the claim construction dispute, the parties first disagree on the meaning of "external clock signal." The central dispute is whether Rambus's claims are limited to clock signals generated by the clocking scheme disclosed in the specification. The parties also appear to disagree on whether the signal must be continuous.The Manufacturers argue that the specification discloses only one clocking scheme, and that it requires the external clock signal to provide both an "early clock" and a "late clock" signal. See `184 Patent, col. 18, l. 62 — col. 19, l. 44. Accurately synchronizing multiple devices in a high speed system is difficult because of the delay caused by the time it takes for the clocking signal to propagate to all of the system's devices. The specification explains that accurate clocking is possible if each device on the bus monitors two bus clock signals and then derives an internal clock from the midpoint of those two signals. Id., col. 18, ll. 63-66. As discussed, the specification discloses two similar methods for doing this: the early and late clocks and the unterminated clock. See id., col. 18, l. 66 — col. 19, l. 13. Rambus contests that the Farmwald/Horowitz specification discloses only one clocking scheme by citing various snatches of the specification's text. The court disagrees. None discloses another clocking scheme. Accord Hynix I, 2004 WL 2610012, at *21 (noting the specification's description of only one clocking scheme).
Based on the specification containing only one clocking scheme, the Manufacturers argue that the proper construction of "external clock signal" must include both an early clock signal and a late clock signal. The Manufacturers note that the specification emphasizes that "another object of this invention is to provide a clocking scheme to permit high speed clock signals to be sent along the bus with minimal clock skew between devices." `184 Patent, col. 3, ll. 28-30. The specification also trumpets that "the clocking scheme used in this invention has not been used before and in fact would be difficult to implement in backplane buses due to signal degradation caused by connector stubs." Id., col. 3, ll. 8-11. Drs. Farmwald and Horowitz distinguished their clocking scheme from a prior art clocking scheme based on the unusual ramp-shaped waveform used in the prior art, "in contrast to the normal rise-time signals used in the present invention." Id., col. 3, ll. 11-14. The Manufacturers therefore rely on cases like On Demand Machine for the proposition that where the specification clearly defines the limits of the invention and the limits are "described as the advantage and distinction of the invention," the specification may narrow the ordinary meaning of a claim. See 442 F.3d at 1340.
Rambus takes issue with the Manufacturers' claim construction for a number of reasons. First, Rambus counters this argument by pointing out that the Farmwald and Horowitz specification contains a number of inventions, and that only some of the claims are directed to the clocking scheme described in the specification. Rambus argues that the court should not construe every claim in its patents that requires an external clock signal to use the novel clocking scheme Rambus invented. Rambus bolsters this argument by submitting that a person of ordinary skill in the art could have implemented other clocking schemes upon reading Rambus's specification. See Murphy Reply Decl. ¶ 131. Because a person of ordinary skill in the art could implement other clocking schemes, Rambus argues it would be improper to limit its claims to just the novel loop-back clocking scheme. The Manufacturers do not respond to this argument because it was raised in Rambus's reply (though this was partially caused by the Manufacturers' shift in claim construction proposals).
Nevertheless, the Manufacturers and Rambus do disagree on the level of ordinary skill in the art that the court must consider when construing the patents. See Phillips, 415 F.3d at 1313 ("The inquiry into how a person of ordinary skill in the art understands a claim term provides an objective baseline from which to begin claim interpretation."). Rambus submits that a person of ordinary level of skill in the art possesses a B.S. in electrical engineering and three to five years of experience designing memory devices likes DRAMs. Murphy Decl. ¶ 42. The Manufacturers submit that a person of ordinary skill in the art has a bachelors degree in electrical engineering, mechanical engineering, physics, or an equivalent distinction and four to five years of experience in the design of "electromechanical" products or device design generally. McAlexander Decl. ¶ 9. The parties do not expand on these arguments in their supplemental briefs filed after Rambus's reply. Based only on the scant argument, the court agrees with Rambus that the level of ordinary skill in the art at the time the patents-in-suit were originally filed was an electrical engineering degree and experience in the field of memory design specifically. The Manufacturers' conception of the level of ordinary skill is vague and consequently much too broad. It is implausible that an "electromechanical" device designer, which would seem to include people who design everything from cars to blenders, would be "ordinarily skilled" in this art.
At the two-day claim construction hearing, the parties focused their arguments on two recent Federal Circuit cases dealing with this quandary, Golight, Inc. v. Wal-Mart Stores, Inc., 355 F.3d 1327 (Fed. Cir. 2004) and Akeva L.L.C. v. Adidas-Salomon AG, 208 Fed. Appx. 861 (Fed. Cir. 2006). In Golight, the dispute turned on whether a claim to a searchlight with a means for "rotating" required the searchlight to rotate 360° (the infringer's searchlight rotated about 345°). 355 F.3d at 1329-30. The court emphasized that it saw "no clear definition or disavowal of claim scope in the written description of the `989 patent that would limit claim 11 to horizontal rotation through 360°." Id. at 1331. While the specification did list the ability to rotate through 360° as an advantage, it was "but one feature of the invented search light." Id. Because a claim does not have to capture every significant or important feature disclosed in the specification, the court held that it would be improper to limit "rotating" to rotating 360°. Id. Despite the fact that the specification disclosed only the 360°-rotating preferred embodiment, the "absence of a clear disclaimer of particular subject matter" led the court to broadly interpret "rotating."
Civil Local Rule 3-4(e) prohibits parties from citing an unpublished opinion of another court if that court's rules prohibit citation. The Federal Circuit does not prohibit or restrict parties "from citing nonprecedential dispositions issued after January 1, 2007." Fed. Cir. Rule 32.1(c) (emphasis added). The Akeva opinion predates January 1, 2007, and should not be considered. Nevertheless, the court will comment on its holding since "an opinion or order which is designated as nonprecedential is one determined by the panel issuing it as not adding significantly to the body of law." Fed. Cir. Rule 32.1(b).
The Akeva case involved a patent to shoes with detachable heels. See 208 Fed. Appx. at 862-64. The claim language required "a rear sole secured below a portion of [the upper part of the shoe]." Id. at 863. The alleged infringer argued that "secured" had to be interpreted in light of the specification as not covering shoes that lacked detachable or rotatable heels. Id. The Federal Circuit agreed, and distinguished Golight because "the detachable sole of the [patent] is not one of several features, it is the primary feature of the invention." Id. at 864. The whole point of the invention was to provide extended life, versatility, and better performance by interchanging rear soles. Id. In Golight on the other hand, there was no disclaimer where the specification described several significant or important features. Id.
Reconciling Akeva with Golight, and taking into account that "an opinion or order which is designated as nonprecedential is one determined by the panel issuing it as not adding significantly to the body of law," it seems that a feature of a single embodiment discussed in the specification should not limit the interpretation of a claim unless it is the sole inventive feature disclosed in the specification. If there are multiple inventive features, the specification must clearly disclaim or disavow subject matter for the specification to limit a claim term's ordinary meaning. Here, the loop-back clocking scheme is but one of many inventions disclosed in the Farmwald/Horowitz specification; indeed, the patent lists seven distinct objectives in its summary. The new clocking scheme is a discrete object of the invention, separate from providing an improved bus architecture and new memory devices. See '184 patent, col. 3, ll. 22-48. Under Golight, it would be inappropriate to limit every claim reciting an "external clock signal" to Rambus's novel clocking scheme, and unlike Akeva, Rambus's specification discloses multiple inventions. The court therefore rejects the Manufacturers' construction.
Rambus's other arguments bolster and confirm this interpretation. Rambus points to a claim it does not assert in these cases. In U.S. Patent No. 6,807,598 (featuring the same Farmwald/Horowitz specification), Rambus claimed a method of operating an integrated circuit device requiring a first external clock signal and a second external clock signal. See '598 patent, col. 27, ll. 56-65; see also Hynix I, 2004 WL 2610012, at *20-*21 (construing "first external clock" and "second external clock"). Rambus argues that under the Manufacturers' claim construction, this claim would require a first external clock signal comprised of an early clock and a late clock and a second external clock signal comprised of another early clock and another late clock. Rambus notes that such a construction would prevent that claim from covering the preferred embodiment disclosed in the specification due to the plethora of clocks required, while its construction of external clock signal would allow the claim to be reconciled with the "early clock/late clock" embodiment from the specification.
Finally, Rambus points out that Hynix has previously agreed with Rambus's construction of "external clock signal." In Hynix I, the parties agreed that an "external clock signal" was "a periodic signal from a source external to the device to provide timing information." Id. at *20. The parties had disagreed on the construction of "first external clock signal," which Rambus argued should be construed the same as "external clock signal." The court instead construed "first external clock signal" as "a periodic signal received by the memory device from an external source to provide first timing information." Id. at *21.
Meanwhile, in litigation with Rambus in the District of Delaware, Micron argued that an "external clock signal" was "a periodic signal received by the memory device from an external source to provide first timing information," i.e., Micron advanced this court's construction of "first external clock signal" as its interpretation of "external clock signal." This is not the same construction that Rambus asserts now because Micron's interpretation might imply a need for a second external clock signal, but it is different from Micron's current construction.
While arguing that they should not be bound by these prior positions, the Manufacturers do not explain why these prior positions were wrong other than to continue to argue that the specification discloses only one clocking scheme and that the claims should be limited to that scheme. Whether Rambus's specification would allow a person of ordinary skill in the art to devise an "external clock signal" other than the early clock/late clock embodiment disclosed in the specification goes to whether Rambus successfully enabled its claims that require only an "external clock signal." If in fact a person of ordinary skill in the art could not implement an external clock signal other than the early clock/late clock, Rambus's claims will fail. But if a person of ordinary skill could implement other external clocks as Rambus argues, then it would be improper to limit all of Rambus's claims to the novel clocking embodiment disclosed. Doing so would effectively punish Rambus for having disclosed an additional clocking scheme instead of remaining silent and relying on the prior art. It is thus clear from Rambus's claim differentiation argument, the ordinary meaning of the words "external clock signal," and the parties' prior positions that an "external clock signal" is "a periodic signal from a source external to the device to provide timing information."
What is unclear is why an "external clock signal" must also be construed to be "continuous." While the Manufacturers propose this construction, their argument is only developed in footnote 30 on page 29 of their brief. Rambus only rebuts the argument by suggesting that "continuous" adds "unnecessary verbiage" to the construction of "external clock signal" in footnote 26 on page 19 of its reply brief. Neither side has developed this argument to the extent the court feels comfortable determining whether an "external clock signal" must be continuous. Nor must the court countenance substantive arguments raised only in footnotes. See SmithKline Beecham Corp. v. Apotex Corp., 439 F.3d 1312, 1320 (Fed. Cir. 2006). Accordingly, the court will retain the stipulated construction of "external clock signal" from Hynix I. iii. "Internal Clock Signal" Claim term: Rambus's Construction: The Manufacturers' Construction:
In the joint claim construction and prehearing statement, the Manufacturers proposed that an "internal clock signal" was "an internally created clock signal that is aligned with the midpoint of the external clock signal as sampled at two separate points along the bus clock line by the device."
Rambus's proposed construction comes from the joint construction of the term in Hynix I. See C-00-20905, Docket No. 326, at 3 (N.D. Cal. Sept. 15, 2003). While Rambus amply illustrates why the Manufacturers' construction must fail, Rambus does not explain why the court should adopt its construction. Specifically, Rambus does not explain why an "internal clock signal" may be periodic or "gated periodic." The requirement that a clock signal be periodic comes from the definition of "clock" shared by persons of ordinary skill in the art. See Murphy Reply Decl. ¶ 141 (quoting technical dictionaries to define a clock as a "device that generates periodic signals used for synchronization"). Indeed, the parties agreed that an "external clock signal" must be periodic. But Rambus's briefing never explains why a person of ordinary skill in the art would understand that a clock signal could also be "gated periodic." See McAlexander Decl. ¶ 106.
On the second day of the claim construction hearing, Rambus's counsel explained that many devices include a gate for blocking the clock signal to conserve energy when the device is not needed. He urged that persons of ordinary skill in the art would recognize that an internal clock signal could also be gated. The Manufacturers do not respond to this argument; indeed, their expert Mr. McAlexander's opinion is limited to pointing out that the specification does not use the word "gated."
A final version of the hearing transcript for the second day is not yet available. The colloquy between the court and Mr. Detre regarding gated periodic clock signals occurred around 10:21 AM. In the final version of the hearing transcript, each line of text has a time stamp. Using these time stamps, one can locate the cited discussion.
While the issue of whether an "internal clock signal" can be a gated periodic signal has been addressed only tangentially, the court will adopt Rambus's construction of an "internal clock signal" from Hynix I because the Manufacturers have not demonstrated why the prior stipulated construction should not be adopted. Accordingly, the court construes "internal clock signal" to mean "a periodic or gated periodic signal generated in a device to provide timing information for internal operation."
The court notes that Rambus's construction of "internal clock signal" does not define its relationship to the "external clock signal." But it appears clear from the specification that a device's internal clock signal must bear some relationship to an external clock signal received by all of the other devices to ensure that the devices remain synchronized. No such clarification of the term "internal clock signal" is required here, however, because the asserted claims all describe the relationship between the external and internal clock signals necessary for each claim. For example, claim 32 of `051 patent requires circuitry "to generate an internal clock signal having a predetermined phase relationship with respect to the external clock signal." U.S. Patent No. 6,314,051, col. 27, ll. 1-4. On the other hand, a few claims may appear ambiguous in this regard. For example, claim 36 of the `8,020 patent requires "a clock alignment circuit to receive the external clock signal" and that "the clock alignment circuit generates an internal clock signal." U.S. Patent No. 6,378,020, col. 28, ll. 35-38; see also U.S. Patent No. 6,426,916, col. 28, ll. 19-23. This language may suggest a break in the linkage between the external clock signal and internal clock signal. This ambiguity is dispelled, however, by the proper construction of "clock alignment circuit." See infra III-C-7.
4. "Synchronous" Terms Claim term: Rambus's Construction: The Manufacturers' Construction:
"synchronously Having a known timing Having a fixed timing relationship with respect to. with respect to" relationship with respect to. "synchronous A memory device that receives an A memory device that receives an external clock signal. memory device" external clock signal which governs the timing of the response to a transaction request. "synchronize" Establish a known timing No construction is necessary. relationship between. The Manufacturers argue that "synchronously" must be interpreted to mean "having a fixed timing relationship," as opposed to a known timing relationship, to reflect the limits of the early clock/late clock system disclosed in the Farmwald/Horowitz specification. The Manufacturers' brief does not further elucidate this argument. At the hearing, counsel for Hynix stated that "the Manufacturers don't dispute Rambus's proposed construction." No one representing Micron, Nanya or Samsung disagreed with this statement or presented arguments in favor of the Manufacturers' proposed construction. Accordingly, the court adopts Rambus's constructions of "synchronously with respect to" and "synchronize."The parties next dispute the proper construction of "synchronous memory device." Both parties agree that a synchronous memory device receives an external clock signal. Rambus further suggests that a synchronous memory device must allow the external clock signal to govern its transaction requests. The Manufacturers disagree that the term imposes any additional limit.
Hynix and Rambus previously disputed the construction of this phrase in Hynix I, and the court construed a "synchronous memory device" as "a memory device that receives an external clock signal which governs the timing of the response to a transaction request." See id. at *9. This is the construction that Rambus now proposes. The Manufacturers' briefing is silent as to why their proposed construction of "synchronous memory device" is correct. Rambus points out that it does not make sense to a call a memory device "synchronous" simply because it receives an external clock signal — it must do something in response to that clock signal to be considered synchronous. Rambus's argument makes common sense. Furthermore, the Manufacturers fail to explain why the court should deviate from its prior construction. Accordingly, the court adopts its prior construction of "synchronous memory device."
5. "Sample" Terms Claim term: Rambus's Construction: The Manufacturers' Construction:
The Manufacturers previously construed "sample" as "obtain from the bus at intervals defined by the internal clock signal."
Rambus submits that the word "sample" as used in the claims is unambiguous and does not require construction. Where a term's ordinary meaning is readily apparent, the court may simply refer to the dictionary to nail down the "widely accepted meaning of commonly understood words." Phillips, 415 F.3d at 1314. Nevertheless, where the parties actually dispute the proper construction of a claim term, the court must construe it rather than let an issue of claim interpretation be decided by the jury. O2 Micro Int'l Ltd. v. Beyond Innovation Tech. Co., Ltd., 521 F.3d 1351, 1361 (Fed. Cir. 2008). Accordingly, the court declines Rambus's request that the court not construe the term "sample."
The term "sample" is ubiquitous in the asserted claims. For example, in claim 14 of the `184 patent, the verb "sample" appears in two limitations. First, "the first block size information defines a first amount of data to be sampled by the memory device in response to a write request." Then, the memory device must " sampl[e] a first portion of the first amount of data synchronously with respect to a first transition of an external clock signal." The claim language reveals the awkwardness of the Manufacturers' construction. Claim 14 requires the memory device to sample synchronously with respect to an external clock signal. The Manufacturers' construction would cause the claim to require the memory device to obtain the data synchronously with the internal clock signal as well. The Manufacturers explain that any difficulty in reconciling these issues is moot because their construction of internal clock signal explains its relationship to the external clock signal, and therefore it is not a problem for the memory device to obtain data synchronously with respect to both the internal clock and the external clock because the two are related. However, the court has rejected the Manufacturers' construction of internal clock signal, leaving this problem to linger.
Rambus has provided various dictionary definitions of "sample." For example, the 1988 Standard Dictionary of Electrical and Electronics Terms defines "sampled data" as "data in which the information content can be ascertained only at discrete intervals of time." The Manufacturers point to nothing in the claims or specification that conflicts with this ordinary meaning. They do urge that their construction of "sample" follows from limiting the scope of the claims to the early clock/late clock embodiment. As the court has declined to limit the claims to that clocking scheme, the Manufacturers' rationale for their construction of "sample" also fails. Accordingly, the court construes "sample" to mean "to obtain at a discrete point in time," "samples" as "obtains at discrete points in time," and "sampling" as "obtaining at discrete points in time."
6. Input Receiver(s) Circuitry and Output Driver(s) Circuitry Claim term: Rambus's Construction: The Manufacturers' Construction:
"Input An element/circuitry on the device A circuit/circuitry on the device to sample one or more receiver(s) to receive one or more signals signals from the bus synchronously with the internal clock circuitry" from an external source. signal of the device. "Output driver(s) An element/circuitry that outputs A circuit/circuitry on the device to output data to the bus circuitry" information from the device. synchronously with the internal clock signal. The only major difference between the parties' proposed claim constructions is whether the input receivers and output drivers must respond "synchronously with the internal clock signal." To be sure, this dispute also incorporates the parties' dispute regarding the scope of the clocking terms through the word "internal clock signal," which the court resolved above. Because it sheds additional light on the argument regarding "synchronously with respect to the internal clock signal," the court addresses it briefly.The Manufacturers note that the specification states that "[o]ne important part of the input/output circuitry generates an internal device clock based on early and late bus clocks." `184 Patent, col. 22, ll. 51-52. The specification also states that input receivers and output drivers must "operate as close in time as possible to midway between the two bus clocks." Id., col. 22, ll. 54-56. From this, the Manufacturers argue that the terms "input receiver" and "output driver" must be interpreted to reflect the need for the early and late bus clocks, and that their construction of "internal clock signal" accounts for this requirement. The Manufacturers overlook the context of the discussion containing these quotes. This discussion relates to the need to account for the delay time between the input samplers and output drivers receiving the device's clock signal and their responding by sampling the bus for data or outputting data to the bus. If this delay time is comparable to the frequency of the system, the delay could cause these circuits to miss their "window" for sending or receiving data. But this need for specialized clocking to compensate for the circuit's delay time is only necessary in very high-speed environments. In slower environments, the circuits' delay times are irrelevant. The Manufacturers' proposed construction overlooks that the proper construction of "input receiver" and "output driver" must account for their use in any system, not just high-speed systems.
The claims reflect this distinction. In U.S. Patent no. 6,715,020, Rambus first claims "a controller device [comprising] first output driver circuitry to output block size information to the memory device, . . .; and input receiver circuitry to receive the amount of data output by the memory device." Col. 24, ll. 57-65. In the tenth claim of the `5,020 patent, Rambus claims "the controller device of claim 1 further including delay locked loop circuitry, coupled to the first output driver circuitry, to generate an internal clock signal, wherein the first output driver circuitry outputs the block size information synchronously with respect to the internal clock signal." Id., col. 25, ll. 29-33. Under the Manufacturers' construction of "output driver," the final limitation of claim 10 requiring the output driver to output data synchronously with respect to the internal clock signal becomes redundant. A person of ordinary skill reading the `5,020 patent would clearly see that "output driver circuitry" should not be construed to require the output of data "synchronously with respect to the internal clock signal" because claim 10 adds that limitation. Claim 9 of the `5,020 patent similarly undermines the Manufacturers' construction of "input receiver circuitry." See id., col. 25, ll. 24-28. Where a synchronous relationship to the internal clock signal is required to compensate for the circuits' delay times, the claims expressly dictate that requirement.
Unlike the Manufacturers' deficient constructions of these terms, Rambus's constructions are straightforward and appear to reflect the ordinary meaning of the terms. The construction of "input receiver" and "output driver" do, however, deviate slightly from Rambus and Hynix's stipulated construction of these terms: Claim term: Rambus's Construction Now: Hynix and Rambus's Construction Then: See Id.
"Input receiver(s) An element/circuitry on the device to Circuitry on the device to receive one or more circuitry" receive one or more signals from an signals from an external source. C-00-20905, external source. Docket No. 326, at 4 (N.D. Cal. Sept. 15, 2003). "Output driver(s) An element/circuitry that outputs Circuitry that outputs information from the device. circuitry" information from the device. Rambus does not explain the significance of broadening the meaning of "input receiver(s) circuitry" or "output driver(s) circuitry" to include an "element" as well as "circuitry." Rambus submits no evidence that an "element" is synonymous with "circuitry," nor does Rambus explain why the construction of these terms should be broadened beyond what it agreed to in Hynix I. At the hearing, the court asked, "What's the significance of element/circuitry?" Rambus's counsel responded that "some of the claims use the term output driver and some of the claims use the term output driver circuitry. When it's output driver, we say it's an element. When it's output driver circuitry we say it's circuitry." Rambus's counsel noted that "There's no significant difference in my view."Because inserting "element" into the construction of these terms does not appear to create any "significant difference," the court adopts the simpler construction of these terms that Rambus and Hynix agreed to in Hynix I. 7. Clock Alignment Circuit Claim term: Rambus's Construction: The Manufacturers' Construction:
"Clock A circuit for adjusting the timing A circuit in each device that generates an internal clock alignment relationship between a clock signal signal that is aligned with the midpoint of the early clock circuit" and another signal. and late clock from the external clock signal. Rambus proposes the claim construction stipulated to by Rambus and Hynix previously. See C-00-20905, Docket No. 326, at 3 (N.D. Cal. Sept. 15, 2003). The Manufacturers propose a construction to incorporate the early clock/late clock scheme from the specification. Neither side offers any argument on the merits of these proposed constructions outside of their broader arguments about whether the Farmwald/Horowitz patents are limited to the loopback clocking scheme.The phrase "clock alignment circuit" does not appear in the specification, and is used in only two of the claims in suit. Claim 32 of the `051 patent recites a memory device that, among other things, includes "a clock alignment circuit coupled to the clock receiver circuitry, to generate an internal clock signal having a predetermined phase relationship with respect to the external clock signal." Claim 36 of the `8,020 patent recites an integrated circuit device that features "a clock alignment circuit to receive the external clock signal" and further requires the clock alignment circuit to "generate an internal clock signal." The court has already rejected the Manufacturers' contention that the internal clock signal must result from the midpoint of an early clock signal and a late clock signal. There is no reason now to imply such a limitation here. Instead, the previously stipulated construction accurately reflects the ordinary meaning of a clock alignment circuit. It is a circuit "for adjusting the timing relationship between a clock signal and another signal." In the two claims-in-suit using the phrase, each claim defines the clock signal and the other signal at issue.
Samsung advances a separate argument in the joint claim construction hearing statement's appendix and in a footnote. Samsung argues that the phrase "clock alignment circuit" is a means-plus-function claim and must be restricted to the embodiments disclosed in the specification, effectively limiting these two claims to the early clock/late clock embodiment. See 35 U.S.C. § 112, ¶ 6. The claims at issue do not use the term "means" and are therefore presumed to not fall under the requirements for means-plus-function claims. Linear Tech. Corp. v. Impala Linear Corp., 379 F.3d 1311, 1319-20 (Fed. Cir. 2004). A "circuit" though is a combination of electrical components for accomplishing a function. Id. at 1320. Thus, the term "circuit" can require further definition under section 112, paragraph 6, but "when the structure-connoting term `circuit' is coupled with a description of the circuit's operation, sufficient structural meaning generally will be conveyed to persons of ordinary skill in the art, and § 112 ¶ 6 presumptively will not apply." Id.
In Linear Technology, the claim at issue required "a first circuit for monitoring a signal from the output terminal to generate a first feedback signal." Id. The court held that the claim language described the circuit's "objective" (monitoring the output terminal's signal) and the circuit's "output" (a feedback signal). Id. It therefore held that because the claim recited the "respective circuit's operation in sufficient detail to suggest structure to persons of ordinary skill in the art," the claim was presumptively not a means-plus-function claim. Id. at 1320-21. Here, the word "circuit" is prefaced by "clock alignment" and each claim requires that the clock alignment circuit "generate an internal clock signal." The claims dictate the circuit's objective (to align at least two signals) and its output (an internal clock signal). Accordingly, section 112, paragraph 6 presumptively does not apply. Samsung failed to make any argument to support its interpretation at the hearing and the single footnote in the Manufacturers' brief addressing this point does not explain why "clock alignment circuit" fails to suggest sufficient structure to a person of ordinary skill in the art. Since Samsung has failed to rebut the presumption against reading these claims as means-plus-function claims, the court concludes that the term "clock alignment circuit" does not fall under section 112, paragraph 6.
8. Delay Locked Loop Claim term: Rambus's Construction: The Manufacturers' Construction:
"Delay lock(ed) Circuitry on the device, including a variable A circuit in each device that uses fixed and loop" delay line, that uses feedback to adjust the variable delay lines and feedback to align the amount of delay of the variable delay line internal clock signal with the midpoint of the and to generate a signal having a controlled early clock and late clock from the external clock timing relationship relative to another signal. signal. Like clock alignment circuit, the terms "delay lock loop," "delay locked loop," and "DLL" do not appear in the written description. This term does appear, however, in five of the claims-in-suit. In claim 4 of the `446 patent, the delay lock loop "synchronize[s] the output of the first and second portions of the data with the external clock signal." U.S. Patent No. 6,546,446, col. 25, ll. 17-20. The other claims-in-suit require a delay locked loop to "generate an internal clock signal using the external clock signal" or to generate multiple internal clock signals. See, e.g., U.S. Patent No. 6,697,295, col. 28, ll. 13-19 (claim 45); U.S. Patent No. 6,426,916, col. 28, ll. 19-23 (claim 40); U.S. Patent No. 5,915,105, col. 29, ll. 9-15 (claim 34); U.S. Patent No. 6,034,918, col. 27, ll. 26-31 (claim 33).Rambus submits the same construction of the term that it and Hynix agreed to in Hynix I. See C-00-20905, Docket No. 326, at 3 (N.D. Cal. Sept. 15, 2003). Hynix and the other manufacturers do not explain why the prior stipulated construction is wrong, except for their argument that the specification discloses only an early clock/late clock clocking scheme and that the terms must be construed to reflect such a scheme. As the court has rejected this argument, the Manufacturers do not further explain why their construction should be adopted. Indeed, the Manufacturers' expert, Mr. McAlexander, has retreated at a recent deposition from the Manufacturers' proposed construction in favor of a prior construction of his that lacked a midpoint limitation. See Detre Supp. Decl., Ex. A at 148:11-149:19.
But Mr. McAlexander's prior construction is not the one Rambus proposes. Mr. McAlexander previously explained that the ordinary meaning of a "delay locked loop" was "a circuit loop that uses feedback signals to adjust the timing relationship between two signals." Id. Both constructions require a circuit to use feedback to create a timing relationship between two signals. The meaningful differences appear to be that Rambus's construction also requires that the circuit contain a variable delay line and that the circuit use the feedback signal to adjust the variable delay line. Neither party submitted any technical dictionaries defining the phrase, nor do the parties' experts' declarations meaningfully explain the basis for either side's construction.
The court must give "delay locked loop" its ordinary meaning at the time of the invention, here, 1990. Phillips, 415 F.3d at 1313. Prior art patents, whether cited by any patent-in-suit or not, can assist in determining the ordinary meaning of a term. Arthur A. Collins, Inc. v. Northern Telecom Ltd., 216 F.3d 1042, 1044-45 (Fed. Cir. 2000). Such patents provide better insight than expert testimony because they cannot be shaped by the pending litigation and were initially written for the person of ordinary skill in the art. The most relevant patent uncovered by the court's search is U.S. Patent No. 4,338,569, issued July 6, 1982, for a "Delay Lock Loop." For the purposes of this discussion, it is significant that the "delay lock loop" describes the need to use a feedback signal to control a variable or programmable delay device. See U.S. Patent No. 4,338,569, col. 2, l. 40 — col. 3, l. 2; compare with id., col. 11, l. 15 — col. 14, l. 19 (every claim recites a "variable delay means" and a "feedback means" for providing an input to the variable delay means).
Based on this prior art description of a delay lock loop and Rambus and Hynix's prior stipulation, the court concludes that Rambus's proposed construction more accurately captures the ordinary meaning of "delay locked loop" to a person of ordinary skill in the art in 1990. The Manufacturers' proposed construction is too narrow. It suffers from focusing solely on the clocking scheme disclosed in the specification. Mr. McAlexander's prior construction is too broad. It encompasses too much by including any and all circuit loops that use feedback. Rambus's construction appears accurate. Accordingly, the court adopts the stipulated construction of "delay locked loop" from Hynix I. A delay locked loop as used in the claims means "circuitry on the device, including a variable delay line, that uses feedback to adjust the amount of delay of the variable delay line and to generate a signal having a controlled timing relationship relative to another signal."
D. Other Terms
Rambus requests construction of "coupled to," "precharged automatically" and "automatically precharged." In the parties' joint claim construction hearing statement, the Manufacturers submitted constructions of these terms that differ from Rambus's. The Manufacturers' brief is silent with respect to these terms, despite Rambus raising them in its opening brief. The Manufacturers' expert declares that he has considered the Manufacturers' proposed constructions, and opines that "the Manufacturers' constructions reflect what one skilled in the art would have understood each term to mean in view of the teachings of the Ware Patents-in-Suit." McAlexander Decl. ¶ 44. The "Ware Patents-in-Suit" refer to the other two patents involved in this litigation that arise from a different application, and the court has construed those claims in a separate order. The three terms addressed here that Rambus raised as needing construction only appear in the Farmwald/Horowitz patents. See Joint Claim Construction and Prehearing Statement, C-05-00334, Docket No. 254, Appx. A at 4, 8-9 (N.D. Cal. Jul. 11, 2007). The Manufacturers' expert's declaration, already conclusory, is entitled to no weight based on its failure to discuss the correct patents. Collectively, the Manufacturers' briefing and their expert's declaration fail to explain the basis for disputing these terms with Rambus.
With respect to the precharge terms, the court can discern no difference between "precharged without an additional command" (Rambus's proposal) and "precharg[ed] without additional instruction" (the Manufacturers' proposal). Because the Manufacturers do not explain any difference, the court adopts Rambus's construction of these terms.
For "coupled to," Rambus proposes "electrically (or otherwise) connected to allow the transfer of signals." The Manufacturers propose that no construction is necessary. Rambus provides various dictionary definitions that suggest that "coupled" is synonymous with "connected," and the Oxford English Dictionary agrees, noting that "to couple" may mean "to join or connect in any way." Moreover, "the term coupled is a term of art in patent parlance that means electrically (or otherwise) connected to allow the transfer of signals." Intel Corp. v. Broadcom Corp., 172 F. Supp. 2d 478, 490-91 (D. Del. 2001) (holding that a node "does not as part of its own definition have to be coupled, or electronically connected, to the communication medium"). In light of these definitions, case law, and the Manufacturers' failure to make any contrary argument, the court adopts Rambus's construction. Accordingly, the court construes "coupled to" as "electrically (or otherwise) connected to allow the transfer of signals."
IV. NON-INFRINGEMENT
The Manufacturers have moved for summary judgment of non-infringement contingent on the court's adoption of their proposed constructions of "device," the "transaction" terms, and the "clocking" terms. As the court has declined to adopt the Manufacturers' constructions of those terms, the court denies their motion for summary judgment of non-infringement as moot.
V. INVALIDITY BASED ON WRITTEN DESCRIPTION
Contingent on the court adopting certain claim constructions proposed by Rambus, the Manufacturers move for summary judgment of invalidity. The Manufacturers argue that various claims are invalid for failing to comply with the written description requirement in 35 U.S.C. § 112, ¶ 1 and for failing to claim the subject matter the inventors regarded as their invention under 35 U.S.C. § 112, ¶ 2.
A. The Manufacturers' Burden on Summary Judgment
A patent must contain a written description of the invention. 35 U.S.C. § 112, ¶ 1. This requirement ensures that "the patentee had possession of the claimed invention at the time of the application, i.e., that the patentee invented what is claimed." LizardTech, Inc. v. Earth Resource Mapping, Inc., 424 F.3d 1336, 1344-45 (Fed. Cir. 2005). Whether a patent satisfies the written description requirement is a question of fact. Vas-Cath Inc. v. Mahurkar, 935 F.2d 1555, 1563 (Fed. Cir. 1991) (reversing summary judgment of invalidity for want of written description because whether drawings adequately described the invention could not be decided as a matter of law). The jury must determine whether a person of ordinary skill in the art would recognize that the inventor possessed the claimed invention. Id. at 1563-64. The Manufacturers' burden of persuasion on this question is clear and convincing evidence. Ralston Purina Co. v. Far-Mar-Co, Inc., 772 F.2d 1570, 1574 (Fed. Cir. 1985).
The Federal Circuit has recently explained that "[c]ompliance with the written description requirement is a question of fact but is amenable to summary judgment in cases where no reasonable fact finder could return a verdict for the non-moving party." PowerOasis, Inc. v. T-Mobile USA, Inc., 522 F.3d 1299, 1307 (Fed. Cir. 2008) (citing Invitrogen Corp. v. Clontech Labs., Inc., 429 F.3d 1052, 1072-73 (Fed. Cir. 2005)). It must be noted that Invitrogen involved a motion for partial summary judgment by the patentee that its patent complied with the written description requirement, which is not the case here. 429 F.3d at 1072-73. If the alleged infringer is challenging the patent's validity, the movant must show that no reasonable jury could conclude that its evidence fails to clearly and convincingly show that the written description requirement has not been met. Accord Vas-Cath, 935 F.2d at 1567. Plainly, this is a high burden to meet, though it can be met. See University Of Rochester v. G.D. Searle Co., Inc., 358 F.3d 916, 927 (Fed. Cir. 2004) (rejecting the argument that "a patent may not be held invalid on its face" on summary judgment).
B. The Written Description Requirement
The Manufacturers move for summary judgment, arguing that the Farmwald/Horowitz specification does not satisfy the written description requirement for any claim featuring a "device" limitation, any of the "transaction" terms, or any of the "clocking" terms.
The parties generally agree on the legal framework of this dispute. The parties agree that the Farmwald/Horowitz specification discloses a memory device that uses a narrow, multiplexed bus interface, a packet-based transaction protocol, and a early clock/late clock scheme for synchronizing devices and reducing clock skew. The parties also agree that the Federal Circuit's case law recognizes two situations in which a specification that adequately describes a species might fail to adequately describe a claim to the genus. See Bilstad v. Wakalopulos, 386 F.3d 1116, 1125-26 (Fed. Cir. 2004). The first situation applies where the art is too unpredictable for the disclosure of a species to show that a patentee also possessed the genus. See id. at 1125 (citing In re Curtis, 354 F.3d 1347 (Fed. Cir. 2004)). The Manufacturers' motion for summary judgment does not assert this first argument. Reply at 10.
The Manufacturers' argument rests on the second scenario described in Bilstad. The Federal Circuit has held a claim invalid under the written description requirement where the specification emphasizes the species while disclaiming the genus. Bilstad, 386 F.3d at 1126 (citing and explaining Tronzo v. Biomet, Inc., 156 F.3d 1154 (Fed. Cir. 1998)). At this point, the parties disagree about what a patentee must say in the specification to limit the patent's written description to the species disclosed. Rambus relies on the Bilstad court's explanation of Tronzo. The Bilstad court characterized the standard as requiring the patentee to "specifically distinguish" a species from its genus or "expressly disclaim" other species for the written description not to provide support for a claim to a genus. Id. at 1125. The "specifically distinguish" language was quoted from Tronzo, where the court explained that, "instead of suggesting that [the patent] encompasses additional shapes, the specification specifically distinguishes the prior art as inferior and touts the advantages of the [claimed shape]." 156 F.3d at 1159. Based on this reasoning, the court held that "such statements make clear that [the patent] discloses only [the claimed shape] and nothing broader." Id. Accordingly, the written description could not support a claim to a device with a generic shape. Id.
The Manufacturers disagree with Rambus's characterization that the law requires the specification to "specifically distinguish" or "expressly disclaim" other species. The Manufacturers point out that under Phillips a claim may not be interpreted to cover material that the specification "expressly disclaims." See 415 F.3d at 1316 ("In other cases, the specification may reveal an intentional disclaimer, or disavowal, of claim scope . . . In that instance . . . the inventor has dictated the correct claim scope, and the inventor's intention, as expressed in the specification, is regarded as dispositive."). Given Phillips' guidance on claim construction, it seems to logically follow that no claim should be invalid for violating the written description requirement as in Tronzo because no claim should be construed broadly enough to raise that issue. The Manufacturers argue that this demonstrates that Tronzo is broader than the situation of "express disclaimer" because the Federal Circuit could not have meant to conflate the written description requirement and claim construction. The court is not so sure. See Lizardtech, Inc. v. Earth Resource Mapping, Inc., 433 F.3d 1373, 1376-78 (Fed. Cir. 2006) (discussing tension between written description and claim construction) (Rader, J., dissenting from denial of rehearing en banc). The rule that a specific disclaimer should limit the construction of a claim (and hence avoid a written description problem like the one in Tronzo) accords well with the general canon that claims should be construed to preserve their validity. See Phillips, 415 F.3d at 1327-28 (discussing the values and limits of the preserve validity canon).
Furthermore, the panel opinion in Lizardtech demonstrates where Tronzo might apply despite Phillips' limits on claim construction. In Lizardtech, the specification described a single way of accomplishing the claimed computer process. 424 F.3d 1336, 1344 (Fed. Cir. 2005). It also explained the "disfavored" prior art. Id. at 1345. Faced with a generic claim to any method of accomplishing the claimed computer process, the court held the claim invalid for failing to satisfy the written description requirement because the specification would not have demonstrated to a person of ordinary skill in the art that Lizardtech possessed the ability to perform the method other than by the disclosed process. Id. at 1345-46. This issue only arose, however, because the patent explicitly contained separate claims to Lizardtech's specific, disclosed method and to the generic method. See id. at 1343 (comparing the scope of the claims). Because the court could not narrowly construe the generic claim in the face of the more specific claims, it had to reach the issue of whether the claim satisfied the written description requirement. In situations where clear claim differentiation prevents the court from construing a claim to exclude explicitly disclaimed material, the Tronzo rule provides a backstop.
Accordingly, the court interprets the Tronzo line of the Federal Circuit's written description case law as invalidating claims to a genus where the written description specifically distinguished its embodiment from the genus or expressly disclaims other members of the genus. These clear limits on the Tronzo line of authority take into account that Tronzo is an outlier in the Federal Circuit's jurisprudence. The touchstone of the written description requirement is that the specification must demonstrate to a person of ordinary skill that the patentee possessed what it claimed. Pandrol USA, LP v. Airboss Ry. Products, Inc., 424 F.3d 1161, 1165 (Fed. Cir. 2005). The court in Lizardtech applied the same standard, noting that after reading the specification, a person of ordinary skill would have only understood Lizardtech as possessing the specific method described. 424 F.3d at 1345. The Tronzo decision, however, does not accord with this principle. By suggesting that claims covering generic shapes did not satisfy the written description requirement because the patentee specifically distinguished them, it seems inescapable that the patentee actually did, in fact, possess devices of other shapes. See Tronzo, 156 F.3d at 1159. How else could the patentee have denigrated the other shapes if it did not possess them? It seems that the patentee simply did not consider the other shapes its invention. By contrast, the other lines of cases recognized in Bilstad — those involving unpredictable arts — support the thrust of the other written description jurisprudence because if an art is unpredictable, it is unlikely that the patentee possessed all of the variants within the art simply because it possessed one of them. Whether a specification like the one in Tronzo should invalidate a claim under the written description requirement or not is beside the point; Tronzo sets forth a principle that has been repeatedly cited by the Federal Circuit. The court points out the inconsistency between Tronzo and other written description cases only to attempt to cabin a divergent and confusing doctrine and provide structure for this court's review of the Manufacturers' motion. Accord University of Rochester v. G.D. Searle Co., Inc., 375 F.3d 1303 (Fed. Cir. 2004) (opinions of Newman, Lourie, Rader, Linn, Dyk, JJ., expressing concerns regarding the written description requirement); Lizardtech, 433 F.3d 1373 (opinions of Lourie, Rader, JJ.).
C. The Level of Ordinary Skill
As discussed above, the written description inquiry is made from the perspective of one of ordinary skill in the art. In re Ruschig, 379 F.2d 990, 995-96 (C.C.P.A. 1967); see also Bilsatd, 386 F.3d at 1126 (vacating with instructions to make findings regarding the level of ordinary skill in the art). Here, the parties dispute the level of ordinary skill in the art. See supra, n. 15. For the purposes of conducting claim construction, the court rejected the Manufacturers' proffered level of skill in the art and accepted Rambus's. The court cannot make such credibility determinations with respect to a motion for summary judgment.
The Manufacturers submit that the difference in opinion regarding ordinary skill in the art should not preclude summary judgment. First, the Manufacturers argue that there is no difference between the two sides' proposed levels of skill in the art. The court disagrees, as discussed above. The Manufacturers' definition of the level of skill in the art is markedly lower than Rambus's.
The Manufacturers next contend that any difference is immaterial because Rambus's expert, Robert Murphy, does not explain the significance of any difference in the levels of skill in the art. To be sure, Mr. Murphy's declaration is sparse on testimony, but Mr. Murphy does testify that the person of ordinary skill under Mr. McAlexander's definition "could very well have no experience with memory devices." Murphy Decl. ¶ 47. This lack of experience supports the inference that Mr. McAlexander's person of ordinary skill would lack familiarity with the prior art and fail to recognize well-known alternative ways of implementing the patents' teachings.
Nevertheless, Rambus is wrong to suggest that the Manufacturers' motion for summary judgment must be denied simply because of this difference of opinions. Rambus cites a single case denying a motion for summary judgment of obviousness because the court found a "genuine issue of material fact as to the level of ordinary skill in the art." Emerson Elec. Co. v. Spartan Tool, LLC, 223 F. Supp. 2d 856, 913 (N.D. Ohio 2002). Rambus overlooks that the record in that case was riddled with issues of fact that precluded granting summary judgment. See id. at 917 (denying summary judgment "particularly in light of the conflicting evidence concerning the differences between the prior art and the asserted claims and defendant's failure to present evidence of some teaching, suggestion or reason to combine the references"). As with all motions for summary judgment, the court must draw all reasonable inferences in favor of the non-movant. Here, the parties dispute the level of ordinary skill in the art; therefore, the court must draw an inference in favor of Rambus's submission. But if the record demonstrates that even a person of Rambus's level of ordinary skill would not recognize that Drs. Farmwald and Horowitz possessed the claimed inventions, the court must still grant summary judgment. In that event, the dispute over the level of ordinary skill in the art ceases to be a dispute of material fact and cannot prevent summary judgment.
D. Claims Using the "Device" Terms
As a preliminary matter, the court observes that the Manufacturers' motion for summary judgment proceeds by clusters of terms, and not claim by claim. This presentation makes it difficult for the court to " separately analyze whether the `written description' requirement has been met as to the subject matter of each claim." Vas-Cath, 935 F.2d at 1567 (emphasis in original). That aside, the Manufacturers argue that the specification "expressly disclaimed" or "specifically distinguished" any claim to a device that lacks the limitations of their proposed construction, i.e., a narrow, multiplexed bus interface and the absence of a separate device-select line.
To begin, the Manufacturers emphasize the first object of the invention, namely, that the invention uses "a new bus interface built into semiconductor devices to support high-speed access to large blocks of data from a single memory device by an external user of the data, such as a microprocessor, in an efficient and cost-effective manner." Mot. at 23 (quoting to `184 patent, col. 3, ll. 22-26) (emphasis in original). Indeed, Rambus concedes that all of the claims-in-suit implement this objective. See Claim Construction Hrg. Tr. 114:7-20 (June 4, 2008). The Manufacturers therefore argue that Drs. Farmwald and Horowitz disclaimed any coverage of devices that do not use "a new bus interface." But the phrase "new bus interface" does not encompass or compel the limitations that the Manufacturers suggest. The bus's interface consists of more than device selection, the number of bus lines, and whether the bus lines are multiplexed. Indeed, those specific aspects of bus interface technology are described as distinct objects of Dr. Farmwald and Horowitz's invention. See '184 patent, col. 4, ll. 33-40. Furthermore, those aspects of a bus interface are largely aspects of the bus and are not "built into semiconductor devices." On the contrary, one aspect of the bus interface that is built into the semiconductor device is the use of programmable registers for controlling access to the bus included in many of the claims-in-suit. See id., col. 21, ll. 28-31. While this object may distinguish claims that fail to use "a new bus interface," the phrase "new bus interface" does not require the claimed devices to interface with a narrow, multiplexed bus or to lack a device-select line.
The Manufacturers next point out the following discussion from the summary of the invention:
In the system of this invention, DRAMs and other devices receive address and control information over the bus and transmit or receive requested data over the same bus. Each memory device contains only a single bus interface with no other signal pins. Other devices that may be included in the system can connect to the bus and other non-bus lines, such as input/output lines.Id., col. 4, ll. 10-16. The Manufacturers argue that this passage specifically distinguishes the invention from claims to a "memory device" with more signal pins than a single bus interface. Similar language appears in the detailed description of the invention, where the specification states that "in the preferred implementation, memory devices are provided that have no connections other than the bus connections described herein and CPUs are provided that use the bus of this invention as the principal, if not exclusive, connection to memory and to other devices on the bus." Id., col. 6, ll. 58-63. This reduces the number of pins on the DRAM interface, and the specification points out that the state-of-the-art DRAM interface requires too many pins. See id., col. 2, ll. 44-50.
Nonetheless, the court cannot agree that these passages indicate as a matter of law that the specification discloses only memory devices that connect solely to a bus that carries device-select information. Compare with Tronzo, 156 F.3d at 1159. Here, the specification described the preferred memory device as connecting to the bus and nothing else. It "touted" this feature as reducing the number of pins needed on the DRAM, which in turn permits a number of advances. But this language is not as extreme as the disclaimer in Tronzo, and Tronzo is an extreme case. In Tronzo, the patentee "specifically distinguish[ed] the prior art as inferior and tout[ed] the advantages" of the new implant shape. Id. Rambus points out, and the court agrees, that the language used here is not so extreme that Drs. Farmwald and Horowitz claimed a species of "memory devices" while disclaiming the genus as a matter of law.
The Manufacturers point to a number of other passages of the specification to support their argument that Drs. Farmwald and Horowitz specifically distinguished and disclaimed devices lacking the limitations of the Manufacturers' proposed construction of "device." Those passages do not impose the limitations the Manufacturers want incorporated. For example, the Manufacturers point to the discussion of how "the DRAMs that connect to this bus differ from conventional DRAMs in a number of ways." See '184 patent, col. 4, ll. 21-31. The first way in which the new DRAMs differ is that they use programmable registers to store control information. This passage supports the various claims to a memory device with a register; it does not undercut them. The next passage discusses various aspects of the new bus. It says little about the invention's memory or integrated circuit devices, though it does suggest that they do not need device-select lines because such information is sent over the bus. This discussion does not take the form of an "express disclaimer," however. Finally, the Manufacturers quote that "many of these details have been implemented selectively in certain fast memory devices, but never in conjunction with the bus architecture of this invention." Id., col. 22, ll. 62-65. This passage appears in a discussion of Drs. Farmwald and Horowitz's modifications to accessing the columns of a DRAM's memory arrays. The passage does not concern the broader discussion in the specification, and does not disclaim devices that lack an interface to the narrow, multiplexed bus.
E. Claims Using the "Transaction" Terms
The Manufacturers next move for summary judgment of invalidity on the claims using the "transaction" terms (i.e., all of the claims) by arguing that the specification does not support claims that fail to incorporate the specification's packet-based protocol. This argument consists of a string cite of portions of the specification discussing the packet protocol and a reference to a single (misplaced) paragraph of Mr. McAlexander's declaration. See Mot. at 25; compare McAlexander Decl. ¶ 104 with ¶¶ 138-143. Moreover, the Manufacturers agree with Rambus that non-packetized transfer protocols were known in the art when Drs. Farmwald and Horowitz filed their application. See Reply at 17 fn. 10. By this court's reading of the written description jurisprudence, the only way for the Manufacturers to establish that Rambus's disclosure of a species of transfer protocols (the packet protocol) does not disclose the entire genus of transfer protocols is for the Manufacturers to show that Rambus disclaimed or distinguished the other existing transfer protocols.
The Manufacturers' briefing cites to chunks of the specification. Mot. at 25 (citing the `184 patent at 8:59-62; 8:66-9:4; 9:11-13; 12:45-48; 12:51-55; 13:4-6; 16:34-38; 16:62-65; 20:17-21; 20:32-38; 20:46-49; 21:28-31; and 22:21-23). None of these passages disclaims other transfer protocols. Instead, these portions of the specification describe the preferred embodiment and how it uses the packetized transfer protocol. Under Tronzo and Bilstad, this is insufficient to establish as a matter of law that non-packetized protocols are not part of the patents' disclosure because the patentee did not expressly disclaim or distinguish other transfer protocols.
F. Claims Using the "Clocking" Terms
Finally, the Manufacturers move for summary judgment that every claim incorporating a "clocking" term is invalid under the written description requirement because the court's construction of those terms does not incorporate the early clock/late clock limitation of the preferred embodiment. The Manufacturers' arguments largely rehash their claim construction dispute, but a few points merit further discussion.
The Manufacturers first emphasize that one of the objects of the invention is to provide a novel clocking scheme "to permit high speed clock signals to be sent along the bus with minimal clock skew between devices." This passage of the specification is not very relevant, however, because Rambus's various claims are directed at accomplishing different objects of the invention, and the claims-in-suit are not directed at providing a novel clocking scheme to reduce clock skew. They are directed at providing a new bus interface instead. Compare '184 Patent, col. 3, ll. 22-27 with id., col. 3, ll. 28-30. As discussed above, it would be inappropriate to limit independent interface claims to the novel clocking scheme simply because Drs. Farmwald and Horowitz also invented a new clocking scheme.
The second cited passage of the specification is more apt. Drs. Farmwald and Horowitz did explain that "one important part of the input/output circuitry generates an internal device clock based on early and late bus clocks" and that it is "important" to control clock skew in systems operating at very high frequencies. See id., col. 22, ll. 50-56. To be sure, Drs. Farmwald and Horowitz "touted" the role of their early clock/late clock system in achieving this. But in so doing, they did not distinguish, disclaim, or denigrate other methods of controlling clock skew by generating an internal device clock. Simply "touting" a species is not enough under Tronzo to prohibit claims to the genus. Indeed, discussing a "preferred" embodiment necessarily involves "touting" one solution to the problem, but a patentee's description of its invention is not limited to the preferred embodiment. The Federal Circuit's written description case law requires some additional criticism or distinction of the other members of species to invalidate claims covering those members, and that criticism is lacking here. See Reply at 19-20.
Accordingly, the court denies the Manufacturers' motion for summary judgment of invalidity based on want of written description under Tronzo. By this holding, the court does not foreclose further inquiry into the written description support of the various claims in the Farmwald and Horowitz patents including what was known by one skilled in the art at the time the original application was filed. For example, the Manufacturers first raise in their supplemental brief arguments specific to claims reciting a DLL and suggesting that a person of ordinary skill in the art would not recognize from the specification that Drs. Farmwald and Horowitz possessed those claimed inventions. Indeed, Dr. Horowitz testified that a DLL was a "scarier technology than others" and "ooh, expert, only experts touched that technology." Tr. 4380:6-9. He also characterized a DLL as "one of the trickiest circuits on the world." See Tr. 4210:5-10. Other DLL patents hint at this complexity. See U.S. Patent No. 4,338,569 ("Delay Lock Loop"); U.S. Patent No. 5,614,855 ("Delay Locked Loop," assigned to Rambus). However, because this argument was not properly raised in the Manufacturers' motion, the court does not reach it now.
G. Section 112, Paragraph 2
The Manufacturers also move for summary judgment pursuant to 35 U.S.C. § 112, ¶ 2, which requires claims to set forth "what `the applicant regards as his invention'" and to be definite. Allen Eng'g Corp. v. Bartell Indus., Inc., 299 F.3d 1336, 1348 (Fed. Cir. 2002) (quoting Solomon v. Kimberly-Clark Corp., 216 F.3d 1372, 1377 (Fed. Cir. 2000)). The Manufacturers argue that a careful study of the specification reveals that Drs. Farmwald and Horowitz did not consider the claims as they are presently construed to be their invention, and that therefore the claims are invalid.
The section 112, paragraph 2 inquiry is tied up in notions of public notice, and therefore, the most probative evidence with respect to this inquiry is the specification. See Solomon, 216 F.3d at 1379-80. Because of this public notice focus, inventor testimony has no relevance, even though the text of the statute suggests an inquiry into the inventor's subjective knowledge. Id. at 1380. As most recently framed by the Federal Circuit, a claim is invalid under section 112, paragraph 2 if "it would be apparent to one of skill in the art, based on the specification, that the invention set forth in a claim is not what the patentee regarded as his invention." Allen Eng'g, 299 F.3d at 1349. Despite the need to consider the perspective of one of ordinary skill in the art, the case law clearly states that unlike the written description inquiry this issue is a question of law, not fact. Solomon, 216 F.3d at 1377.
Curiously, the central piece of legislative history regarding the codification of the Patent Act says very little about section 112. See Senate Report No. 1979, 82d Cong., 2d Sess. (1952). The only comment regarding the provision at issue is: "The clause relating to the claim is made a separate paragraph to emphasize the distinction between the description and the claim or definition, and the language is modified." The committee report does suggest that this provision results from codifying Revised Statute 4888, which required the inventor to "particularly point out and distinctly claim the part, improvement, or combination which he claims as his invention or discovery." It appears that this "requirement" of section 112 may in fact be nothing more than a requirement that a patent application conclude with claims.
There is scant case law applying section 112, paragraph 2. Where it has invalidated a claim, the claim contradicted the specification and the patentee "admit[ted] as much." Allen Eng'g, 299 F.3d at 1349. This court has interpreted Allen Engineering as holding that a claim is invalid if the claim cannot be logically reconciled with the specification. TV Interactive Data Corp. v. Microsoft Corp., 2004 WL 2823231, *3 (N.D. Cal. Oct. 13, 2004) (White, J.).
Applying this standard, the Manufacturers' motion must fail. The Manufacturers point to nothing in the claims that logically contradicts the specification. Instead, the Manufacturers emphasize that Drs. Farmwald and Horowitz repeatedly referred to "the present invention" in describing certain embodiments. This is insufficient to show a violation of section 112, paragraph 2. Accordingly, the Manufacturers' motion for summary judgment on this ground is denied.
VI. ORDER
For the reasons set forth above, the court construes the claims as described (and as summarized in Appendix 2). The court denies the Manufacturers' motions for summary judgment of non-infringement and invalidity. Appendix 1: Information regarding the claims-in-suit. U.S. Issuance Defenda Asserted Claim Language Patent nt in No.; Hynix? Claim No.
Bracketed text indicates that the claim is dependent on a prior claim. The bracketed text represents the prior claim's language that is incorporated in the dependent claim.
Counsel are responsible for distributing copies of this document to co-counsel that have not registered for e-filing under the court's CMlECF program in each action.