[137]*137OPINION
SUE L. ROBINSON, District Judge.
I. INTRODUCTION
This action arises out of a dispute between Micron Technology, Inc., Micron Electronics, Inc., and Micron Semiconductor Products, Inc. (collectively, “Micron”) and Rambus Inc. (“Rambus”) over Micron’s alleged infringement of multiple Rambus patents.1 The case has been trifurcated (D.I. 739), and the court has held a bench trial on the issues of Ram-bus’ alleged spoliation of evidence and unclean hands and the appropriate sanction, if any, arising from those allegations. These issues have been fully briefed post-trial. (D.I. 1074, 1076, 1079, 1083, 1084, 1086, 1087)
II. FINDINGS OF FACT
1. Rambus Inc. (“Rambus”) was founded in March 1990 by Professors Mike Farmwald and Mark Horowitz. (MTX 48 at 0048-0003; RAMTX 212-0004) Rambus has described itself as a technology company employing semiconductor, system architecture, and system packaging technologies. (MTX 48 at 0048-0003) At the outset, Rambus focused on solving the so-called “memory bottleneck” problem — a scenario then foreseen in the industry where microprocessors’ speed would far outstrip the speed at which memory components, such as dynamic random access memory (“DRAM”), transferred data to and from microprocessors. (D.I. 1061 at 1244:6-14; D.I. 1062 at 1670:16-1673:10; RAMTX 212; see also RAMTX 259; D.I. 1061 at 1242:19-1245:7) Farmwald and Horowitz ultimately developed a potential DRAM solution.2 (D.I. 1061 at 1240:14-1241:22, 1257:6-1260:3; RAMTX 211)
2. In April 1990, Farmwald and Horowitz filed a patent application on behalf of Ram-bus containing the DRAM inventions.3 (D.I. 1061 at 1226:6-17) Rambus’ plan was to license its DRAM technology, “RDRAM,”4 to DRAM manufacturers. (D.I. 1062 at 1656— 3:9; D.I. 1061 at 1124:12-23) Rambus’ goal, by making RDRAM widely compatible, was to achieve industry-wide adoption. (D.I. 1061 at 1247:4-1248:4,1123:3-24:23)
3. In 1991, representatives from chip technology owners (including Rambus), chip manufacturers (including Micron), and chip purchasers began meeting through the Joint Electronic Design Engineering Council (“JE-DEC”) (MTX 203 at 2) to discuss and adopt industry-wide standards for computer memory chips.5 (D.I. 1058 at 160:16-161:4) One [138]*138standard these representatives sought was for what became SDRAM.6 (D.I. 1058 at 171:15-18)
4. Starting around this period, Rambus became concerned that DRAM manufacturers were using Rambus’ technology to develop their own competing DRAMs.7 (D.I. 1061 at 1133:5-19, 1135:7-13, 1082:2-1083:9; D.I. 1060 at 798:12-22, 686:2-13) From 1991 to 1995, Rambus representatives took information learned at JEDEC meetings and passed it along to Rambus’ patent prosecution counsel in an effort to solidify and extend Ram-bus’ patent claims to cover SDRAM, DDR SDRAM, and other potentially competing memory types.8 (D.I. 1060 at 693:23-695:21; see also id. at 724:23-725:18, 736:9-21, 798:12-799:8, 806:24-807:14; RAMTX 069; RAMTX 85; MTX 40) Statements made by Rambus employees in 1996 and 1997 reveal that Rambus planned to create a patent “minefield” that it could use to its advantage in dealing with other companies in the industry.9 (MTX 253; MTX 183; MTX 235) Ram-bus wanted to be able to “[g]et all infringers to license our IP with royalties [greater than] RDRAM ... OR sue.” (MTX 279 at ¶ 18(C); see also D.I. 1060 at 751:3-13)
5. On April 30, 1996, the PTO issued to Rambus United States Patent No. 5,513,327 (“the '327 patent”). (MTX 200) Also in 1996, Intel agreed to use RDRAM with its microprocessors. (D.I. 1061 at 1128:22-29:5; D.I. 1062 at 1619:17-1620:2; RAMTX 241) The Intel agreement was “critieal[ly]” significant to Rambus because, at that time, “almost all [personal computers] used Intel’s microprocessor and Intel’s chipset” (D.I. 1061 at 1128:13-18), and personal computers represented approximately half of the total DRAM market. (D.I. 1061 at 1127:14-17) Around that same time, Rambus entered into licensing agreements for RDRAM with eleven of the twelve major DRAM manufacturers, including Micron, with the goal of developing a new version of RDRAM (Direct RDRAM) in conjunction with these manufacturers. (D.I. 1061 at 1148:3-8, 1129:13-1130:6; RAMTX 67)
6. On August 12,1997, the PTO issued to Rambus United States Patent No. 5,657,481 (“the '481 patent”). (MTX 256) A few weeks later, in October 1997, Rambus hired Joel Karp as Vice President of Intellectual Property. (D.I. 1058 at 154:1-4, 291:23-292:1) Prior to joining Rambus, Karp had worked for several years at Samsung, where he participated in patent license negotiations. (D.I. 1059 at 405:5-12) Rambus hired Karp to work on a licensing program for non-Rambus technologies, including SDRAMs and DDR10 DRAMs, that CEO Tate said infringed the Rambus patents.11 (D.I. 1058 at 171:19-175:1) Karp was responsible for “assessing [the Rambus] patent portfolio, determining when chips infringe [the Rambus] patent [139]*139portfolio, setting licensing strategies for infringing chips, and for negotiations with companies that build and sell infringing chips.” (RAMTX 6) When Rambus hired Karp, Tate told him that any company wanting to license Rambus’ present and future patents for infringing DRAM would have to pay a royalty greater than the royalty for RDRAM. (MTX 263) Tate also advised Rambus executives to “have their spin control ready” to “downplay the whole infringemenf/lP issue” that could be raised in the industry by Karp’s hiring. (MTX 263) Tate advised other Ram-bus employees similarly and reminded them that “[asserting the Rambus] patents [is] really just a backup line of defense. [T]he # 1 strategy for winning must continue to be having the best solution____[I] think [Ram-bus is] well positioned to do this but we have a LOT to do to complete our programs/commitments on Direct Rambus and Concurrent Rambus over the next several quarters.” (RAMTX 6; see also D.I. 1061 at 1145:5-1147:3; D.I. 1071 at 154:23-155:14; D.I. 1059 at 405:21—406:24)
7. On January 7, 1998, Tate met with Karp and instructed him to prepare a presentation for the March 1998 Board of Directors (“the Board”) meeting discussing, inter alia, a licensing framework and a litigation strategy. (MTX 385-0022; MTX 270-0006; D.I. 1058 at 182:5-24, 184:9-185:14) Accordingly, Karp called Diane Savage, an attorney practicing in the Cooley Godward law firm’s technology transactions group with whom he had previously worked, and asked to be referred to a litigator. (D.I. 1059 at 597:9-598:15, 441:7-19) Savage then contacted Dan Johnson, a litigation partner at Cooley Godward, and arranged a meeting. (Id. at 599:1-7) Another Cooley Godward attorney, John Girvin, had primary responsibility for assisting Rambus in its engagement — which Cooley Godward identified as concerning “licensing activities” — but Johnson and attorneys Peter Leal and William Winters were also staffed on the Ram-bus matter.12 (RAMTX 83)
8. On January 13, 1998, Karp and Tate met with Leal. (D.I. 1058 at 187:18-188:2; see also MTX 285) Leal’s notes of that meeting indicate that Rambus wanted the following: “litigation strategy by [the] March board meeting,” “[n]o negotiations [without] full strategy and preparation],” “[g]o in quickly [and] proceed to either a license or litigation,” “[t]ry win-win first; do not prejudice [good faith] for litigation,” “[Booking for royalty rate that tells them it costs to infringe,” and “[g]o to first meeting but be ready (in advance) to go to litigation.” (MTX 285-0001; D.I. 1059 at 615:9-616:5, 616:22-618:15, 622:12-623:20, 624:9-625:9, 628:24-629:15) Karp also communicated to Leal Rambus’ belief that the '481 and '327 patents were being infringed. (MTX 287)
9. On January 15, 2008, Karp met with Leal again. (Id.) They discussed Rambus patents that allegedly were being infringed. (Id.; see also D.I. 1058 at 191:2-192:3) Prior to this meeting, Karp had prepared a “claim chart”13 outlining his evidence of infringement. (MTX 287-0003; D.I. 1058 at 175:12-18,178:5-179:12)
10. On February 12, 1998, Karp met with Leal, Johnson, and Girvin. (MTX 290; see also MTX 304-0001; D.I. 1058 at 196:24-197:6, 199:23-200:16) Karp and the attorneys discussed both licensing and litigation strategies,14 but discussion of litigation strategy predominated. (MTX 290; MTX 506) Specific to litigation, Karp and the attorneys discussed preparing trial graphics and claims; retaining experts; gathering critical documents and implementing a document retention policy;15 and building a case against [140]*140potential litigation targets, including Micron,16 Fujitsu, Samsung, and Hyundai. (MTX 290; MTX 506; D.I. 1058 at 198:24-199:16) They also discussed that Rambus would “[njeed to litigate against someone to establish [a] royalty rate and have [a] court declare [the Rambus] patent[s] valid” and that the planned royalty rate of five percent would “probably push [Rambus] into litigation quickly.”17 (MTX 290; D.I. 1058 at 202:2-19,212:18-213:15)
11. On February 20, 1998, Johnson prepared a memorandum outlining for Rambus proposed licensing and litigation strategies.18 (MTX 293-0001; MTX 304-0002) The memorandum discusses Rambus’ competitors, features of the licensing program and a hierarchy of potential licensees and, in the event that licensing efforts failed, a tiered litigation strategy contemplating litigation in fora that “proceed at an accelerated schedule,” making early preparation advantageous for Rambus. (MTX 293; see also D.I. 1059 at 444:17-445:9; D.I. 1062 at 1501:20-1503:7)
12. On March 2, 1998, Karp presented the licensing and litigation strategies to the Rambus Board. (MTX 295; D.I. 1058 at 214:2-19; D.I. 1061 at 1269:14-1270:11) Karp proposed seeking an up-front payment and a five percent running royalty19 from licensees of non-compatible products, with incentives and penalties tied to licensees’ production rates, and outlined the mechanics of the licensing negotiation. (MTX 295-0001-0002, 0006) Karp also explained the litigation strategy to be implemented if licensing discussions failed, including Rambus’ available causes of action and potential defendants and fora. (Id. at 0003-0005) Finally, he proposed a timeline for the strategies to play out and identified needed near term actions, including creating a document retention policy,20 preparing a discovery database, and organizing the prosecuting attorney’s files for issued patents. (Id. at 0007-0008; see also D.I. 1058 at 246:5-247:3, 260:14-261:14) Neither the Board nor Tate voiced disapproval of the licensing and litigation strategies.21 (D.I. 1058 at 259:23-260:13)
[141]*14113. Rambus then proceeded to implement Karp’s suggested licensing and litigation strategies. On March 16, 1998, after discussions with Karp about “the email back-ups [containing] discoverable information,” Roberts sent an email to Joseph Lau, manager of the computer systems, instructing him to add a Quarter 2 goal about a back-up policy and to implement that policy in Quarters 2 and 3.22 (MTX 299; D.I. 1061 at 1076:19-1078:23) Roberts instructed that the back-up policy should address how often files are backed up, how long they are kept, and where they are stored. (MTX 299) One of the reasons for implementing the policy was to allow Ram-bus to purge documents, including emails, from its files that might be discoverable in litigation. (MTX 299; D.I. 1061 at 1076:19-1078:23, 1181:19-1182:8) Also in March 1998, in response to Karp’s inquiry regarding document retention policies, Savage sent to Karp a memorandum outlining general considerations for creating a document retention policy. (MTX 300; D.I. 1059 at 599:8-600:9)
14. On April 14, 1998, Tate and a colleague met with an Intel executive to discuss the two companies’ future relationship. (MTX 307) Tate summarized Intel’s message at this meeting thus: “[I]ntel says they are basically going to compete with us on next generation [of DRAM].” (Id. at 1) Tate’s notes indicate that Intel was considering, in the “mid-term,” “whether to do RDRAM or extend/modestly improve sdram[s].” (Id. at 2) Tate’s notes also capture Intel’s belief that the “[R]ambus business model[ — charging royalties, etc. — ] [was] fundamentally at odds with what the dram industry wants” and that it was “never guaranteed [Intel would] use Rambus forever.” (Id. at 3-4) Tate understood that Intel, if it could develop a “best solution without using Rambus IP,” would “use [the non-Rambus solution] and freeze out [R]ambus.” (Id. at 6) Tate noted that, when other DRAM companies found out that Intel was investigating the next generation DRAM without Rambus, it might cause them to not want to work with Rambus on the next generation DRAM which could, in turn, “force [Rambus] to play [its] IP card with the dram companies earlier.” (Id.)
15. Also in April 1998, at Karp’s request, Kroll Ontrack, Inc., performed an on-site information security audit of Rambus. (D.I. 1062 at 1430:7-1431:16, 1436:5-21; RAMTX 99) In a presentation following the audit, Kroll recommended that Rambus “[w]ork [ ] with counsel ... [to] create a written plan covering how long various classes of files should be held.” (RAMTX 99-0026; D.I. 1062 at 1441:20-1446:1)
16. On May 14, 1998, after meeting with group representatives of the four Rambus operating divisions and consulting with Johnson, whom Karp identified as litigation counsel for Rambus, Karp sent an email to Rambus employees announcing that, effective immediately, full system back-up tapes would be saved for only three months and that data to be saved beyond three months must be archived separately.23 (MTX 314; D.I. 1058 at 249:12-250:4) Karp also announced the imminent implementation of a company-wide document retention policy.24 (MTX 314) Also in May, Karp had Johnson [142]*142and others at Fenwick & West consider potential litigation targets, including Micron, Hyundai, and SLDRAM. (MTX 345-0005-0007; D.I. 1058 at 255:13-256:13,257:9-19)
17. On July 22, 1998, Karp emailed a document retention policy to Rambus employees.25 (MTX 337; D.I. 1058 at 266:6-11) Karp and Johnson then made presentations about document retention to Rambus employees.26 (D.I. 1058 at 271:3-23) Karp’s slides from those presentations outline the policy.27 (MTX 343) Johnson’s slides from those presentations characterize the policy28 as a precursor to litigation (MTX 333-0002) and focus on the policy vis-a-vis litigation.29 (MTX 333; D.I. 1058 at 272:7-275:10)
18. In August or September 1998, Ram-bus retained attorney Neil Steinberg as outside counsel.30 (D.I. 1060 at 835:4-7) Stein-berg’s work for Rambus included patent prosecution, licensing, and litigation preparation, and Rambus assigned to him all patent prosecution work.31 (Id. at 835:4-836:15; 837:16-24; MTX 375-0001)
19. On September 3, 1998, Rambus employees participated in an event dubbed “Shred Day,” during which they destroyed documents pursuant to the company’s new document retention policy.32 (D.I. 1058 at 283:14-284:13; D.I. 1059 at 526:15-20; MTX 368) David Rhoades, vice president of the company Rambus hired to shred the documents, estimated that 400 banker’s boxes-worth of documents were destroyed.33 (D.I. 1062 at 1469:2-13; MTX 398-0027-0028)
[143]*14320. On September 15, 1998, Tate sent an email to Rambus executives summarizing his meeting with executives from Samsung, then the world’s largest DRAM manufacturer in terms of revenue. (MTX 371; D.I. 1061 at 1102:16-1103:22) Samsung had committed to devoting nineteen percent of its DRAM production to RDRAM by the fourth quarter of 1999.34 (MTX 371-0001; D.I. 1061 at 1104:3-10) According to Tate, Samsung also had “an aggressive direct rdram roadmap”35 and had “substantially increased [its] resources on direct rambus” in anticipation of increased market demand. (MTX 371-0002; D.I. 1061 at 1105:3-12)
21. On October 16, 1998, Intel announced that it would invest $500 million in Micron36 in support of Micron’s commitment to producing Direct RDRAM.37 (RAMTX 273; D.I. 1061 at 1106:16-1107:13) In meetings held sometime in October or November 1998, Micron executives told Roberts that Micron planned to be in “volume production with Rambus parts ... by the second half of 1999.” (D.I. 1061 at 1108:22-1109:24)
22. Later in October 1998, Karp made another strategy presentation to Rambus executives. (D.I. 1058 at 294:1-4; MTX 375) At that time, Rambus was helping its manufacturing partners in working toward mass production of Direct RDRAM. (D.I. 1061 at 1102:6-11) With mass production hopefully imminent and, with it, the establishment of a “dominant [DRAM] standard,” Karp advised against commencing litigation in 1999, even though he believed Rambus could have sued multiple alleged infringers in Quarters 1 and 2 of 1999. (MTX 375-0004) Karp advised “DO NOT ROCK THE DIRECT BOAT” and pointed out that there was no “rush” to sue infringers and that Rambus should not assert its patents against its Direct RDRAM manufacturing partners until their investment in Direct RDRAM had reached “the point of no return,” which he believed would probably not occur until Quarter 1 of 2000. (MTX 375-0004; D.I. 1058 at 294:22-295:2, 296:22-298:2, 299:10-20). Accordingly, Karp advised that Rambus should continue in “stealth mode” during 1999, strengthening its patent portfolio — which Karp labeled as the “top priority” — and proceeding with reverse engineering efforts.38 (MTX 375-0005; D.I. 1058 at 298:11-300:1)
23. On November 24, 1998, the PTO issued to Rambus United States Patent No. 5,841,580 (“the '580 patent”). (MTX 236)
24. Sometime in late 1998 or early 1999, after a November 30,1998 management staff meeting, Karp drafted and circulated a memorandum outlining what he thought Rambus’ [144]*144strategy should be in the “very unlikely” event that Intel cancelled its RDRAM production and moved instead to a competing technology — what Karp termed a “nuclear winter scenario.”39 (MTX 763-0002; D.I. 1058 at 193:3-6, 196:20-22) Karp identified the major goal in this scenario to be convincing Intel that, because of Rambus’ ability to assert its patents (including the '327, '481, and '580 patents to cover DDR SDRAM), it would be costly to move away from RDRAM. (MTX 763-0002) He also identified potential targets, causes of action, and fora if negotiations with Intel failed. (MTX 763-0003-0006; D.I. 1058 at 303:12-22) The memorandum indicated specifically that Rambus already had claim charts showing that Micron infringed one of the Rambus patents. (MTX 763-0003; see also MTX 802)
25. In April 1999, Karp instructed outside counsel Lester Vincent40 of the Blakely Sok-oloff law firm to begin “clear[ing] out” the Rambus patent files for patents that had issued.41 (MTX 420-0004 (emphasis original); see also MTX 934; MTX 601; MTX 412-0001-0012; MTX 427; D.I. 1058 at 304:8-11; D.I. 1061 at 1320:23-1321:8, 1322:6-13, 1328:22-1329:4) Vincent cleared out the patent files over the course of April, May, June, and July of 1999.42 (D.I. 1322:14-22) Documents cleared out included hard and electronic copies of draft amendments, draft claims, and attorneys’ handwritten notes,43 and correspondence regarding patent prosecution, some of which did not exist in any other form.44 (D.I. 1323:6-1325:11,1347:17-1348:1)
26. On June 22, 1999, the PTO issued to Rambus the '105 patent, the earliest issued of the patents-in-suit. (MTX 428) On or about June 27, 1999, Karp, with Steinberg’s input, prepared goals for Quarter 3, 1999.45 [145]*145(D.I. 1058 at 305:15-306:24, 308:22-309:18; see also MTX 434, MTX 436, MTX 438) Karp’s goals included “[initiating reverse engineering of infringing devices as required for litigation prep;” “[m]onitor industry for potential infringing activity — all semiconductors;” “[djevelop complete licensing strategy;” “[pjresent licensing strategy to exec and gain approval;” “[p]repare licensing positions against 3 manufacturers;” “[p]repare litigation strategy against 1 of the 3 manufacturers;” and be “[r]eady for litigation with 30 days notice.”46 (MTX 438-0001) The goals are in keeping with Tate’s instruction to Karp in a June 24, 1999 email to think about a “target and bottom line terms for licensing [Rambus’] IP for infringing DRAMs;” the identity of the “first target and why;” and what Rambus’ “strategy [would be] for the battle with the first target that we will launch in [0]ctober.” (MTX 429-0003)
27. Consistent with Tate’s instruction, Karp asked Johnson to compare the duration and timing of patent litigation in the United States District Court for the Northern District of California and the United States District Court for the Eastern District of Virginia.47 (MTX 441-0002; D.I. 1058 at 317:13-318:2, 318:23-319:19) On July 8, 1999, Johnson’s associate sent the requested information to Karp. (MTX 441-0001)
28. On August 26, 1999, Rambus had another shred day. (D.I. 1062 at 1470:13-23; MTX 398-0034, 0035; MTX 450) The shredding company’s invoices reflect its estimate that 300 boxes of documents were destroyed. (MTX 398-0034, 0035; see also D.I. 1062 at 1470:24-1471:12) At the time Rambus had its second shred day, the '105 patent had already issued to Rambus, and Rambus had received notices of allowance regarding the '263 and '804 patents. (D.I. 1059 at 363:24-366:11; D.I. 1060 at 756:21-757:24; MTX 428)
29. In September 1999,48 Karp and Stein-berg made a presentation to Rambus management entitled “IS THERE LIFE AT RAMBUS AFTER INTEL?” (MTX 801; D.I. 1058 at 326:7-22) The presentation slides state that, with Rambus’ relationship with Intel waning, Rambus “must increase the industry’s perception of [its] value through aggressive assertion of [its] IP rights ... [i]f Rambus is to have a future.”49 (MTX 801-0001-0003) The slides go on to state that, for the Rambus intellectual property to earn respect, Rambus must substantiate its patent claims by either a lucrative licensing deal with an “industry powerhouse” or “winning in court.” (MTX 801-0004) The slides state further that the “[b]est route to IP credibility is through victory over a major DRAM manufacturer,” it is acknowledged that “[companies like Micron will fight us tooth and nail [146]*146and will never settle.”50 (MTX 801-0004r-0005)
30. After this meeting, on September 29, 1999, Tate sent an email indicating that consensus had been achieved on, inter alia, the “need to sue a dram company to set an example: visit them a few weeks in advance at exec level to explain our IP, position, offer terms.”51 (MTX 4644)001-0002) Tate also stated in the email that Rambus should publicize the patents and the suits to “put all dram/controller companies that use sdram/ ddr/ ... on notice,” with the expectation that “no one will settle with [Rambus] until after initial suit is resolved: 1-2.5 years if jury trial; if pre-trial settlement 0.5-1 year[].” (Id.) (ellipses original)
31. On October 14, 1999, Karp made another presentation to the Board. (MTX 468; D.I. 1059 at 359:9-19) This presentation set out a detailed analysis of why Hitachi was the most desirable litigation target.52 (MTX 468; D.I. 1059 at 360:13-361:22) Thus, Karp recommended to the Board that Rambus approach Hitachi during Quarter 4 of 1999 with its infringement allegations, which included showing claim charts to Hitachi that covered, inter alia, “SDRAM (PC100 and PC133),” “DDR SDRAM,” and “SH-5 (SuperH Rise Engine).” (MTX 468-0008) Karp’s recommendation contemplated next giving Hitachi one month to rebut Rambus’ infringement position, followed by a second meeting at which Rambus would present a business proposal. (Id.) If Hitachi failed to respond to the business proposal within two weeks, Karp recommended that Rambus file suit in Delaware.53 (Id.)
32. On October 22, 1999, Karp sent a letter to Hitachi asserting the '105, '263, and '804 patents.54 (MTX 473-0001) Rambus met with Hitachi once to discuss the Rambus patents, but Hitachi thereafter did not respond to Rambus’ requests to meet. (D.I. 1059 at 462:7 — 463:21) Rambus concluded that Hitachi was “stonewalling” and moved toward litigation. (D.I. 1062 at 1628:16-1629:13; D.I. 1071 at 63:10-64:1)
33. In November or December 1999, Rambus entertained presentations from several law firms seeking to represent Rambus in its contemplated litigation against Hitachi, with Rambus ultimately selecting the Gray Cary law firm. (D.I. 1062 at 1506:9-1507:6) In December 1999, Rambus instituted a litigation hold.55 (D.I. 1060 at 857:11-24, 1002:11-1003:4; D.I. 1061 at 1312:2-13; D.I. 1059 at 539:24-541:4, 551:15-552:10)
34. On January 18, 2000, Rambus sued Hitachi.56 (MTX 493; D.I. 1058 at 298:3-10) On June 22, 2000, Rambus settled its suit with Hitachi and, as part of the settlement, entered into a licensing agreement for non-compatible DRAM products.57 (MTX 533; [147]*147D.I. 1059 at 453:24-454:7) During the first half of 1999, Rambus also entered into licensing agreements with Toshiba, Oki, and NEC for non-compatible DRAM products. (D.I. 1059 at 453:24-454:7)
35. On August 24, 2000, Tate emailed Micron CEO Steve Appleton. (RAMTX 8) On August 28, 2000, Appleton replied to Tate via email and informed him that Micron had filed the instant declaratory judgment action. (RAMTX 8; see also D.I. 1) On Wednesday, August 30, 2000, Tate replied to Appleton and stated that Rambus preferred licensing to litigation and that, if Micron were interested in a negotiated resolution, he would be happy to meet with Appleton. (RAMTX 191)
36. On December 28, 2000, at Rambus’ request, a shredding contractor disposed of as many as 480 boxes of Rambus materials in connection with an office move. (MTX 577; D.I. 1472:8-10, 1472:23-1473:2, 1474:11-16)
37. Rambus continued in litigation against Micron and also became engaged in litigation with Hynix and Infineon. During the course of these litigations, one of Ram-bus’ Rule 30(b)(6) witnesses testified that the document retention policy had not been presented to the Board (D.I. 1071 at 181:1— 182:15) before later changing that testimony to acknowledge that presentations regarding the document retention policy had been given to the Board at least twice. (Id. at 179:5-19) Likewise, Rambus’ Rule 30(b)(6) witnesses testified that Rambus conducted only one shred day (D.I. 1060 at 856:11-19; D.I. 1071 at 173:8-21, 178:16-18) before later changing that testimony to acknowledge that Rambus conducted three shred days. (D.I. 1071 at 170:3-173:24)
38. Rambus employees, in several instances, did not inform outside counsel of the Rambus shred days. (D.I. 1061 at 1292:4-1293:13, 1294:23-1295:4; D.I. 1060 at 961:5-23,962:8-963:14;969:20-970:21,971:2-15) Nor did they otherwise inform outside counsel or Rule 30(b)(6) witnesses of the scope of the document destruction: Rule 30(b)(6) witness Kramer was not aware of the back-up tapes being degaussed until 2005 (D.I. 1071 at 196:1-20); Gray Cary attorney Sean Cunningham, who had been representing Ram-bus since the Hitachi litigation in 2000, was not aware of the back-up tape degaussing until 2007 (D.I. 1060 at 926:11-928:19; D.I. 1061 at 1054:23-1056:6); and, although Micron requested documents concerning the shredding services and back-up tapes in 2001, Rambus did not produce all responsive documents until much later (D.I. 1058 at 189:7-190:12; MTX 683; D.I. 1071 at 193:23-194:11; D.I. 1060 936:15-964:11, 965:13-21, MTX 329).
39. The litigation record at bar is replete with misrepresentations on the part of Ram-bus. As examples:
a. Contrary to Rambus’ assertion that it had not contemplated litigation before implementing its document retention policy (MTX 643), notes from the February 12,1998 meeting between Karp and Leal reveal that Ram-bus had contemplated litigation prior to implementing the policy (MTX 290).
b. Contrary to Rambus’ assertion that its document retention policy “had nothing to do with litigation” (MTX 624), several documents reveal that Rambus considered the policy to be part of its litigation preparation. (MTX 276, MTX 278, MTX 434, MTX 436, MTX 802; see also D.I. 1058 at 212:1-15, 243:1-12)
c. While Rambus’ JEDEC representative Crisp testified that he was minimally involved with Rambus’ patent prosecution (D.I. 1060 at 786:6-13, 792:12-793:21), multiple documents reveal that Crisp played a significant role in Rambus’ prosecution of its divi-sionals. (MTX 133, MTX 848, MTX 47; MTX 54; MTX 65; MTX 77; D.I. 1061 at 1351:5-1353:7)
d. While Tate testified that, upon learning that the Rambus patents might not cover competing DRAMs, he took no action to file [148]*148additional patent claims or have Rambus’ pending patent claims modified (D.I. 1059 at 651:15-652:4, 656:22-657:4, 663:9-18), documents reveal that Tate discussed amending the Rambus patent claims with outside counsel and discussed adding claims to the original patent application with other Rambus employees. (MTX 848; MTX 40; MTX 188; see also MTX 48-0017)
III. CONCLUSIONS OF LAW
A. Standard of Review
40. The imposition of sanctions for spoliation are controlled, in patent cases, by regional circuit law. Monsanto Co. v. Ralph, 382 F.3d 1374,1380 (Fed.Cir.2004).
41. “Spoliation is the destruction or significant alteration of evidence, or the failure to preserve property for another’s use as evidence, in pending or reasonably foreseeable litigation.” Kounelis v. Sherrer, 529 F.Supp.2d 503, 519 (D.N.J.2008) (citing MO-SAID Tech. Inc. v. Samsung Elec. Co., 348 F.Supp.2d 332, 335 (D.N.J.2004)). “Spoliation occurs when a party has intentionally or negligently breached its duty to preserve potentially discoverable evidence,” and cannot occur in the absence of such a duty. Id. at 518-19.
42. A duty to preserve evidence arises when there is knowledge of a potential claim. Winters v. Textron, Inc., 187 F.R.D. 518 (M.D.Pa.1999). A potential claim is generally deemed cognizable in this regard when litigation is pending or imminent, or when there is a reasonable belief that litigation is foreseeable. For instance, a duty to preserve evidence can arise many years before litigation commences; immineney is sufficient to create the duty, but it is not a requirement. See Kronisch v. United States, 150 F.3d 112, 126 (2d Cir.1998); In re Napster, Inc. Copyright Litig., 462 F.Supp.2d 1060, 1068 (N.D.Cal.2006); In re Quintus Corp., 353 B.R. 77, 80-84 (Bankr.D.Del.2006), aff'd, 2007 WL 4233665 (D.Del.2007). As soon as a potential claim is thus identified, a party is under a duty to preserve evidence which it knows, or reasonably should know, is relevant to the future litigation. Nat’l Ass’n of Radiation Survivors v. Turnage, 115 F.R.D. 543, 556-57 (N.D.Cal.1987).
43. If a party has a record retention policy, it is appropriate for the court to determine “whether the policy is reasonable considering the facts and circumstances surrounding the relevant documents,” or whether it was instituted in bad faith. Lewy v. Remington Arms Co., 836 F.2d 1104, 1112 (8th Cir.1988).
Even if the court finds the policy to be reasonable given the nature of the documents subject to the policy, the court may find that under the particular circumstances certain documents should have been retained notwithstanding the policy. For example, if [a party] knew or should have known that the documents would become material at some point in the future then such documents should have been preserved. Thus a [party] cannot blindly destroy documents and expect to be shielded by a seemingly innocuous document retention policy.
Gumbs v. Int’l Harvester, Inc., 718 F.2d 88, 96 (3d Cir.1983). Therefore, “[o]nee a party reasonably anticipates litigation, it must suspend its routine document retention/destruction policy and put in place a ‘litigation hold’ to ensure the preservation of relevant documents.” Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 218 (S.D.N.Y.2003).
44. “Sanctions are appropriate when there is evidence that a party’s spoliation threatens the integrity of [the court].” MO-SAID, 348 F.Supp.2d at 335. Sanctions serve three functions: a remedial function (by restoring the aggrieved party to its original position), a punitive function, and a deterrent function. Id.
45. Potential sanctions for spoliation include dismissal or the entry of judgment, preclusion of evidence, an adverse or spoliation inference, and awarding of costs. See, e.g., Computer Assocs. Int’l v. Am. Fundware, Inc., 133 F.R.D. 166, 170 (D.Colo.1990) (concluding that no sanction short of default judgment would adequately punish the spoliator and deter like-minded litigants); United Med. Supply Co. v. United States, 77 Fed.Cl. 257, 276 (2007) (precluding both eross-exami-[149]*149nation of plaintiffs expert and introduction of defense expert as a sanction for spoliation by defense); Travelers Prop. & Cos. Co. v. Cooper Crouse-Hinds, LLC, No. 05-CV-6399, 2007 WL 2571450, at *12 (E.D.Pa. Aug. 31, 2007) (ordering an adverse inference jury instruction in lieu of judgment for defense as a sanction for plaintiffs spoliation of evidence); Turner v. Hudson Transit Lines, Inc., 142 F.R.D. 68, 77-78 (S.D.N.Y.1991) (awarding costs, including attorney fees, for destruction of evidence).
46. A district court has the inherent power to sanction parties. Schmid v. Milwaukee Elec. Tool Corp., 13 F.3d 76, 78 (3d Cir.1994). The choice of a sanction to impose in a particular case is a matter within the sound discretion of the court. See Gates Rubber Co. v. Bando Chem. Indus. Ltd., 167 F.R.D. 90, 106 (D.Colo.1996). The key considerations in determining whether spoliation sanctions are warranted include:
(1) the degree of fault of the party who altered or destroyed the evidence; (2) the degree of prejudice suffered by the opposing party; and (3) whether there is a lesser sanction that will avoid substantial unfairness to the opposing party and, where the offending party is seriously at fault, will serve to deter such conduct by others in the future.
Schmid, 13 F.3d at 79. See also In re Daim-lerChrysler AG, No. Civ. A. 00-993-JJF, 2003 WL 22951696, at *1 (D.Del. Nov. 25, 2003).
47. Dispositive sanctions aré the most severe that can be imposed, and generally are reserved for egregious offenses against an opposing party or the court. Computer Assocs., 133 F.R.D. at 169. Consistent with the analytical framework identified in Schmid, such sanctions are not warranted in the absence of demonstrated bad faith (i.e., the intentional destruction of evidence) and prejudice. With respect to the latter, the imposition of a dispositive sanction should be confined to those cases where the failure to produce “‘materially affect[s] the substantial rights of the adverse party’ and is ‘prejudicial to the presentation of his ease.’ ” Gates, 167 F.R.D. at 104 (citing Wilson v. Volkswagen of America, Inc., 561 F.2d 494, 503 (4th Cir.1977)).
48. The required burden of proof to establish spoliation is not a matter of settled law in the Third Circuit. On the one hand, in order to prove prejudice “[ujnder Schmid, a party need only ‘come forward with plausible, concrete suggestions as to what [the destroyed] evidence might have been.’ ” In re Wechsler, 121 F.Supp.2d at 423 (citing Schmid, 13 F.3d at 80). See also Gates, 167 F.R.D. at 104 (the aggrieved party must establish “a reasonable possibility, based on concrete evidence,” that the spoliated material would have produced evidence favorable to its cause). On the other hand, because dis-positive sanctions “contravene the strong public policy [that] favors adjudication of cases on their merits,” a higher burden of proof may be appropriate. Id. at 108. As explained by the court in Gates, “[t]he elimination of valued rights should not occur in the absence of a degree of proof [that] reflects the very serious nature of the decision,” that is, proof by clear and convincing evidence. Id.
49. Although the court recognizes that requiring clear and convincing evidence for the imposition of dispositive sanctions for spoliation places an onerous burden on the aggrieved party (where the very proof of intent and prejudice arguably has been destroyed), nevertheless, the court concludes that this higher burden can appropriately operate as the clear and convincing burden operates in the patent arena in proving inequitable conduct. More specifically, once intent and prejudice have been established, the court must determine whether their total weight satisfies the clear and convincing standard of proof. In this regard, the showing of intent (i.e., bad faith) can be proportionally less when balanced against high prejudice. In contrast, the showing of intent must be proportionally greater when balanced against low prejudice. See, e.g., N.V. Akzo v. E.I. DuPont de Nemours, 810 F.2d 1148, 1153 (Fed.Cir.1987).
B. Analysis
50. The record at bar demonstrates that, as early as 1996, Rambus contemplated [150]*150industry-wide adoption of its DRAM technology through an aggressive use of its intellectual property, characterized as its “patent minefield.” In October 1997, Rambus hired Joel Karp to manage its patent portfolio. By March 1998, Karp was assigned the task of developing licensing and litigation strategies; although licensing was the ultimate goal, Rambus was commencing its plans for litigation in order to establish a royalty rate and validate its patents. Specific to the litigation strategy, Karp proposed the implementation of a document retention policy. By July 1998, documents were being destroyed pursuant to the new policy; the first shred day was held in September.
51. In April 1998, Intel had informed Rambus that it was investigating the next generation DRAM without Rambus. Nevertheless, throughout 1998, Rambus and its manufacturing partners (including Micron, with the support of Intel) worked toward the mass production of the Rambus Direct RDRAM. In this regard, however, Karp advised executives in October 1998 that Ram-bus should not assert its patents against its Direct RDRAM manufacturing partners until their investment in Direct RDRAM had reached the point of no return, which he predicted would be Quarter 1 of 2000. Accordingly, Karp further advised that Rambus should continue in “stealth mode” during 1999.
52. By the close of 1998, Karp was preparing strategies to enable Rambus to survive the termination of its relationship with Intel. In this regard, Karp had identified potential litigation targets, causes of action, and fora if negotiations with Intel failed. Rambus already had claim charts asserting infringement against Micron.
58. In April 1999, Karp instructed patent counsel to purge the Rambus patent files. By June 1999, litigation was a written goal of Karp’s for which he was preparing. Rambus held its second shred day in August. By September, there was a consensus among Rambus executives of the need to sue a DRAM company to set an example; by October, Hitachi was chosen as the most desirable target. Hitachi was put on notice in October of the Rambus patents; a litigation hold was instituted in December and Hitachi was sued the following month.
54. It is apparent from the record described above that Rambus, from its inception, was prepared to be an aggressive competitor in a very competitive industry. Its patent portfolio was considered a weapon to be used, as necessary, in its chosen theater of operations, the DRAM market. Under these circumstances, one could safely predict that litigation was inevitable. Nevertheless, the court is not prepared to rely on Rambus’ generally assertive approach to business to answer the question before the court, that is, did Rambus intentionally destroy documents at a time when litigation was reasonably foreseeable and, if so, what sanctions should be imposed to cure the prejudice to Micron.
55. The court concludes that litigation was reasonably foreseeable no later than December 1998, when Karp had articulated a time frame and a motive for implementation of the Rambus litigation strategy. Moreover, because the document retention policy was discussed and adopted within the context of Rambus’ litigation strategy, the court finds that Rambus knew, or should have known, that a general implementation of the policy was inappropriate because the documents destroyed would become material at some point in the future. Therefore, a duty to preserve potentially relevant evidence arose in December 1998 and any documents purged from that time forward are deemed to have been intentionally destroyed, i.e., destroyed in bad faith.
56. In determining the degree of prejudice suffered by Micron as a result of this spoliation of evidence, Micron has carried its burden under Schmid to prove that the documents destroyed58 were discoverable and the type of documents that would be relevant to the instant litigation. More specifically, Micron asserts unenforeeability due to patent misuse and violation of the antitrust and unfair competition laws (based in part on Rambus’ conduct at JEDEC), as well [151]*151as inequitable conduct. These are defenses that are illuminated by evidence of a nonpublic nature, e.g., by internal Rambus documents.59 Because the record demonstrates that there were documents relevant to these defenses,60 the court concludes that Micron has been prejudiced by Rambus’ conduct. That prejudice has been compounded by Rambus’ litigation conduct, which has been obstructive at best, misleading at worst.
57. In determining which of the potential sanctions for spoliation should be imposed, the court is directed to find the least harsh sanction that serves both to avoid substantial unfairness to Micron but deter such conduct in the future. In reviewing the record, the court concludes that the showing of bad faith is so clear and convincing that the showing of prejudice can be proportionally less. The spoliation conduct was extensive, including within its scope the destruction of innumerable documents relating to all aspects of Rambus’ business; when considered in light of Rambus’ litigation conduct, the very integrity of the litigation process has been impugned. Sanctions such as adverse jury instructions and preclusion of evidence are impractical, bordering on meaningless, under these circumstances and in the context of a typical jury trial.61 Therefore, the court concludes that the appropriate sanction for the conduct of record is to declare the patents in suit unenforceable against Micron. An order will issue.
ORDER
At Wilmington this 9th day of January, 2009, consistent with the opinion issued this same date;
IT IS ORDERED that:
1. United States Patent Nos. 5,915,105; 5,953,263; 5,954,804; 5,995,443; 6,032,214; 6,032,215; 6,034,918; 6,038,195; 6,324,120; 6,378,020; 6,426,916; and 6,452,863 are unenforceable against plaintiff and counterclaim defendants.
2. The court shall conduct a telephone conference with the parties on Friday, January 16,2009 at 9:30 a.m., to discuss the status of the case in light of the opinion. Plaintiffs counsel shall initiate the call.