`
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`SAMSUNG ELECTRONICS CO., LTD.,
`
`Petitioner,
`
`v.
`
`NETLIST, INC.,
`
`Patent Owner,
`
`_____________________________
`
`Case No. IPR2022-00615
`
`Patent No. 7,619,912
`
`_____________________________
`
` REMOTE VIDEOTAPED DEPOSITION BY VIRTUAL ZOOM OF
`
`ANDREW WOLFE, PH.D.
`
`WEDNESDAY, JANUARY 4, 2023
`
`Reported by:
`
`Ashala Tylor, CSR #2436, CLR, CRR, RPR
`
`JOB NO. 5621734
`
`PAGES 1 - 233
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 1
`
`1
`
`2
`
`3 4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
`
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
` SAMSUNG ELECTRONICS CO., LTD.,
`
` Petitioner,
`
` v.
`
` NETLIST, INC.,
`
` Patent Owner,
`
` _____________________________
`
` Case No. IPR2022-00615
`
` Patent No. 7,619,912
`
` _____________________________
`
` Videotaped deposition of ANDREW WOLFE, PH.D.,
`
`taken via virtual Zoom, commencing at 9:04 a.m. and
`
`ending at 6:38 p.m., on Wednesday, January 4, 2023,
`
`before Ashala Tylor, CSR No. 2436, RPR, CRR, CLR.
`
`1
`
`2
`
`3 4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 2
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`APPEARANCES OF COUNSEL:
`
`FOR THE PETITIONER, SAMSUNG ELECTRONICS CO., LTD.:
`
` BAKER BOTTS LLP
`
` BY: TED CHANDLER, ESQ.
`
` 101 California Street
`
` Suite 3600
`
` San Francisco, California 94111
`
` ted.chandler@bakerbotts.com
`
` - and -
`
` BAKER BOTTS LLP
`
` BY: FERENC PAZMANDI, ESQ.
`
` 1001 Page Mill Road
`
` Suite 200
`
` Palo Alto, California 94304
`
` ferenc.pazmandi@bakerbotts.com
`
`///
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 3
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`A P P E A R A N C E S ( c o n t i n u e d )
`
`F O R T H E P A T E N T O W N E R , N E T L I S T , I N C . :
`
` I R E L L & M A N E L L A L L P
`
` B Y : H . A N N I T A Z H O N G , E S Q .
`
` M I C H A E L T E Z Y A N , E S Q .
`
` 1 8 0 0 A v e n u e o f t h e S t a r s
`
` S u i t e 9 0 0
`
` L o s A n g e l e s , C a l i f o r n i a 9 0 0 6 7
`
` ( 3 1 0 ) 2 0 3 - 7 9 3 6
`
` h z h o n g @ i r e l l . c o m
`
` m t e z y a n @ i r e l l . c o m
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`1 0
`
`1 1
`
`1 2
`
`1 3
`
`1 4
`
`1 5
`
`1 6
`
`1 7
`
`1 8
`
`1 9
`
`2 0
`
`2 1
`
`2 2
`
`2 3
`
`2 4
`
`2 5
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 4
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
` I N D E X
`
`WITNESS EXAMINATION BY PAGE
`
`ANDREW WOLFE, PH.D.
`
` Ms. Zhong 8, 107, 222
`
` Mr. Chandler 215
`
` E X H I B I T S
`
`NO. DESCRIPTION PAGE
`
`Exhibit 1001 USP 7,619,912 (Bhakta, et al.,) 150
`
`Exhibit 1003 Expert Declaration of Dr. Andrew 37
`
` Wolfe
`
`Exhibit 1005 Provisional Application for 107
`
` Patent Cover Sheet, 60/588,244
`
`Exhibit 1006 Provisional Application for 127
`
` Patent Cover Sheet, 60/550,668
`
`Exhibit 1007 Provisional Application for
`
` Patent Cover Sheet, 60/575,595
`
`Exhibit 1029 JEDEC Standard, DDR2 SDRAM 8
`
` Specification, JESD79-2
`
`Exhibit 1030 JEDEC Standard, Double Data 8
`
` Rate (DDR) SDRAM Specification,
`
` JESD79
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 5
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6 7 8 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`E X H I B I T S (continued)
`
`NO. DESCRIPTION PAGE
`
`Exhibit 1032 PC2100 and PC1600 DDR SDRAM 9
`
` Registered DIMM Design
`
` Specification, Revision 1.3,
`
` January 2002
`
`Exhibit 1033 Memory Systems - Cache, DRAM, 192
`
` Disk
`
`Exhibit 1034 Synchronous DRAM Architectures, 75
`
` Organizations, and Alternative
`
` Technologies by Prof. Bruce L.
`
` Jacob, 12-10-02
`
`Exhibit 1035 USP 7,363,422 (Perego, et al.) 24
`
`Exhibit 1036 US Application Publication 171
`
` No. 2006/0117152
`
` (Amidi, et al.,)
`
`Exhibit 1037 US Patent Application 155
`
` Publication No. 2006/0277355
`
`Exhibit 2100 MR16R1624(8/G)EG0 Change History 14
`
` Data Sheet
`
`Exhibit 2101 Provisional Application for 131
`
` Patent Cover Sheet
`
`Exhibit 2102 Screen Capture, Sheet 15 of 18 169
`
`///
`
`///
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 6
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`E X H I B I T S (continued)
`
`Previously Marked Exhibits:
`
`NO. DESCRIPTION PAGE
`
`Exhibit 1009 USP 7,286,436 119
`
` QUESTIONS INSTRUCTED NOT TO ANSWER
`
` Page 222, Line 7
`
` Page 222, Line 17
`
` Page 222, Line 24
`
` Page 223, Line 7
`
`1
`
`2
`
`3
`
`4
`
`5 6 7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 7
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
` Wednesday, January 4, 2023
`
` 9:04 a.m.
`
` --o0o--
`
` ANDREW WOLFE, PH.D.
`
` being first duly sworn or affirmed to testify
`
` to the truth, the whole truth, and nothing but
`
` the truth, was examined and testified as follows:
`
` EXAMINATION
`
`BY MS. ZHONG: 09:04
`
` Q. Dr. Wolfe, can you pull up Exhibit 1032, 09:04
`
`Exhibit 1029, Exhibit 1030. You have them on your 09:05
`
`local computer and -- 09:05
`
` THE REPORTER: I'm sorry, Ms. Zhong, I'm 09:05
`
`having a difficult time hearing you. 09:05
`
`BY MS. ZHONG: 09:05
`
` Q. Okay. Dr. Wolfe, can you pull up 09:05
`
`Exhibits 1032, 1030 and 1029. 09:05
`
` (Exhibit 1029 was marked for 09:05
`
` identification and attached 09:05
`
` hereto.) 09:05
`
` (Exhibit 1030 was marked for 09:05
`
` identification and attached 09:05
`
` hereto.) 09:05
`
`1
`
`2
`
`3
`
`4 5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 8
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` (Exhibit 1032 was marked for 09:05
`
` identification and attached 09:05
`
` hereto.) 09:05
`
` THE WITNESS: All three of them you want 09:05
`
`me to pull up? 09:05
`
`BY MS. ZHONG: 09:05
`
` Q. Yes. You have them on your local 09:05
`
`computer, correct? 09:05
`
` A. I do. 09:05
`
` Q. Okay. Let me know when you have them. 09:05
`
` A. I have them. 09:05
`
` Q. What is Exhibit 1032? 09:05
`
` A. It is the PC2100 and PC1600, DDR SDRAM 09:05
`
`Registered DIMM, D-I-M-M, Design Specification, 09:06
`
`Revision 1.3 from January 2002. And we often refer 09:06
`
`to it as a version of JEDEC Standard 21-C. 09:06
`
` Q. How does JEDEC Standard 21-C differ from 09:06
`
`Exhibit 1030? 09:06
`
` A. They are different documents describing 09:06
`
`different things. 09:06
`
` Q. What does Exhibit 1030 describe? 09:06
`
` A. That is the JESD79 standard or 09:06
`
`specification from June 2000. And it describes 09:06
`
`various aspects of double-data-rate DRAM chips 09:06
`
`primarily. 09:07
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 9
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` Q. Okay. So Exhibit 1030 is directed to the 09:07
`
`SDRAM's memory device itself, correct? 09:07
`
` A. Primarily, yes. 09:07
`
` Q. Okay. And Exhibit 1032 describes the 09:07
`
`memory module level, correct? 09:07
`
` A. For one particular type of memory module, 09:07
`
`that is correct. 09:07
`
` Q. Okay. So Exhibit 1032 describes in 09:07
`
`particular the registered DIMM memory module, 09:07
`
`correct? 09:07
`
` A. A particular type or a particular set of 09:07
`
`types of registered DIMM memory modules. There's 09:07
`
`other information in there as well, but that's its 09:07
`
`primary purpose. 09:07
`
` Q. Okay. And the operation of DDR SDRAM 09:08
`
`device itself you refer to Exhibit 1030; is that 09:08
`
`right? 09:08
`
` A. Exhibit 1030 includes a multitude of 09:08
`
`specific characteristic specifications for a 09:08
`
`packaged double-data-rate SDRAM chip at that time. 09:08
`
` Q. Okay. Why don't we go to Exhibit 1032, 09:08
`
`page 16. 09:08
`
` A. Okay. 09:08
`
` Q. On page 16, lower left corner, do you see 09:08
`
`a bunch of signals that's going into an element 09:08
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 10
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`called a register? Do you see that diagram there? 09:08
`
` A. I do see that diagram. 09:08
`
` Q. Okay. And what is a register shown on 09:09
`
`page 16 of Exhibit 1032? 09:09
`
` A. Well, what a register is always depends on 09:09
`
`context. But in this particular context, that would 09:09
`
`be a component on this module that can temporarily 09:09
`
`store signal values. 09:09
`
` So it would store the input values based 09:09
`
`on certain values of the PCK, which stands for 09:09
`
`clock, and PCK bar signals, and make those available 09:09
`
`at the outputs on the right. 09:09
`
` Q. Okay. So the signals on the left of the 09:09
`
`register are input signals that -- (indiscernible.) 09:09
`
` THE REPORTER: I'm sorry, Counsel. I'm 09:10
`
`having a difficult time hearing you. One more time, 09:10
`
`please. 09:10
`
`BY MS. ZHONG: 09:10
`
` Q. So the signals to the left of the element 09:10
`
`of register are input signals received by the DIMM 09:10
`
`from the computer system; is that correct? 09:10
`
` A. In this diagram, the signals that are on 09:10
`
`the left of the box marked "Register" would come 09:10
`
`from -- they are input signals to the DIMM, and in 09:10
`
`most cases they would come from a computer system. 09:10
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 11
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` Q. Okay. And the signals that's to the right 09:10
`
`of the box labeled "Register," those are output 09:10
`
`signals; is that correct? 09:10
`
` A. From the perspective of this particular 09:10
`
`register they are output signals. They would be 09:11
`
`input signals to some other device. 09:11
`
` Q. Okay. And so the signals to the right of 09:11
`
`the register are output signals from the perspective 09:11
`
`of the register and -- but they are input signals to 09:11
`
`the SDRAM devices; is that correct? 09:11
`
` A. They -- each of those would be an input 09:11
`
`signal to one or more SDRAM devices. 09:11
`
` Q. So the answer to my question is "yes"; is 09:11
`
`that correct? 09:11
`
` A. I don't know whether or not we're saying 09:11
`
`the same thing. 09:11
`
` The register will output each of the 09:11
`
`signals shown to the right of the register. Those 09:11
`
`are then provided -- but some are provided to some 09:11
`
`SDRAM devices. Some are provided to others. Some 09:11
`
`are provided to all of them. 09:11
`
` Q. Okay. And there are different types of 09:11
`
`DIMMs; is that correct?
`
` A. In the world, that is true, there are 09:12
`
`different type of DIMMs in the world. 09:12
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 12
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` Q. Okay. Are you familiar with a type of 09:12
`
`DIMM called RIMM, which I believe is specific to the 09:12
`
`Rambus-designed memory modules? 09:12
`
` A. I don't recall it by that name, but there 09:12
`
`were a number of Rambus memory modules in the 09:12
`
`2000-2005 time frame typically. 09:12
`
` Q. Okay. Do you recall what kind of input 09:12
`
`signals do those RIMMs receive? 09:12
`
` A. Not in detail. The RIMM -- the Rambus 09:12
`
`DRAM memory modules used a packetized communication 09:13
`
`scheme that used a narrower communication channel 09:13
`
`than the most common other DIMMs at the time. 09:13
`
` Q. Okay. So, for example, the Rambus memory 09:13
`
`module -- let me rephrase that. 09:13
`
` So the Rambus DRAM memory module do not 09:13
`
`receive, for example, chip select signal, band 09:13
`
`address signal, RAS bar, CAS bar, WE bar, and 09:13
`
`address this as shown on page 16 of Exhibit 1032; is 09:13
`
`that correct? 09:14
`
` MR. CHANDLER: Objection. Form. 09:14
`
` THE WITNESS: Again, I would need to look 09:14
`
`at that specification to recall all the details. 09:14
`
`It's been 20-plus years since I've read it. But my 09:14
`
`recollection is that the same information was 09:14
`
`conveyed but not in the exact same physical form. 09:14
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 13
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`BY MS. ZHONG: 09:14
`
` Q. Okay. Let me see -- let me introduce a 09:14
`
`document and see if this will jog your memory. 09:14
`
` (Exhibit 2100 was marked for 09:15
`
` identification and attached 09:15
`
` hereto.) 09:15
`
`BY MS. ZHONG: 09:15
`
` Q. So, Dr. Wolfe, can you please refresh your 09:15
`
`Exhibit Share link? 09:15
`
` A. Yes. 09:15
`
` Q. And do you see a newly marked 09:15
`
`Exhibit 2001 -- 2100, EX2100? 09:15
`
` A. EX2100. 09:15
`
` Q. Yes. Can you please open it or download 09:15
`
`it. 09:15
`
` A. Okay. 09:15
`
` Q. Exhibit 21 -- okay. Do you have it? 09:15
`
` A. I do. 09:15
`
` Q. So Exhibit 2100 is a Samsung data sheet 09:15
`
`for RIMM, R-I-M-M, module based on 256 megabits or
`
`288 megabits, 32 bank, 16K/32ms Rambus DRAMs.
`
` Can you go to pages 2, 3, 4, and take a 09:16
`
`look. 09:16
`
` MR. CHANDLER: And, Annita, let me just 09:16
`
`lodge an objection to the exhibit. It's -- appears 09:16
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 14
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`to be outside the scope of his declaration. Also 09:16
`
`doesn't appear to be -- it doesn't appear to be 09:16
`
`evidence that I can -- appears to be hearsay. 09:16
`
`BY MS. ZHONG: 09:16
`
` Q. Dr. Wolfe, can you please take a look at 09:16
`
`pages 2 through 4. 09:16
`
` A. I see them. 09:16
`
` Q. Okay. The signals listed on pages 2 09:16
`
`through 4, the input signals to the DIMM, for this 09:16
`
`particular DIMM. 09:16
`
` A. I'm sorry, I didn't understand the 09:17
`
`question. 09:17
`
` Q. So on page 2 through 4 there are a number 09:17
`
`of pins listed on those pages, correct? 09:17
`
` A. They are called pads in this case. But 09:17
`
`there are a number of connections to a module that 09:17
`
`are described on these pages. 09:17
`
` Q. Okay. So the pads that's listed on those 09:17
`
`pages are the -- how do you say? -- the input and 09:17
`
`output pads for the -- for the memory module; is 09:17
`
`that correct? 09:17
`
` A. Some of them probably are. To get into 09:17
`
`the details, I think, requires reviewing other 09:17
`
`documents. 09:17
`
` Q. Okay. Can you tell me which pin would be 09:17
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 15
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`check select signal coming in from? 09:18
`
` A. Not from a cursory review of this document 09:18
`
`I can't, no. There are other documents, the ones 09:18
`
`intended to review together with this, that describe 09:18
`
`how these signals are used in more detail. 09:18
`
` Q. Okay. Do you see any reference to chip 09:18
`
`select signal here in this document? 09:18
`
` A. There may or may not be. Again, I 09:18
`
`don't -- I don't know the definition of the 09:18
`
`functionality of every pin from this document alone. 09:18
`
` There are other documents that were 09:18
`
`supplied at the time by Rambus that would describe 09:18
`
`how each pin was used. And for this particular kind 09:18
`
`of a DIMM, my recollection was that the DIMMs -- 09:18
`
`that pins or pads had multiple functions. 09:18
`
` Q. Okay. Are the signals -- you mentioned 09:19
`
`that for Rambus memory module the address and the 09:19
`
`control signals are packetized; is that correct? 09:19
`
` THE REPORTER: I'm sorry, are what? I'm 09:19
`
`sorry, Counsel, the last part of your question? 09:19
`
`BY MS. ZHONG: 09:19
`
` Q. You mentioned that the Rambus memory 09:19
`
`module -- let me withdraw and reask. 09:19
`
` So in the Rambus memory module, the 09:19
`
`control and address information is packetized; is 09:19
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 16
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`that correct? 09:19
`
` MR. CHANDLER: Objection. Form. 09:19
`
` THE WITNESS: That was my recollection as 09:19
`
`to certain Rambus memory modules that I was familiar 09:19
`
`with in the early 2000s. I don't know, from what 09:19
`
`I've seen here right now, whether or not this 09:20
`
`particular module is consistent with the ones that I 09:20
`
`previously studied. 09:20
`
`BY MS. ZHONG: 09:20
`
` Q. And how does a packetized control and 09:20
`
`address information differ from the signaling 09:20
`
`protocol that we see on page 16 of Exhibit 1032? 09:20
`
` A. There are similarities and there are 09:20
`
`differences, and it would depend on the specifics of 09:20
`
`the specific situation. 09:20
`
` But I think that the biggest difference is 09:20
`
`that in the Rambus protocols that I was familiar 09:20
`
`with at the time, there was -- there were more pins 09:20
`
`that had multiple functions. 09:20
`
` Q. So is there a discrete chip select signal 09:21
`
`as in Exhibit 1032? 09:21
`
` A. Does what have a discrete chip select 09:21
`
`signal? 09:21
`
` Q. Let me put it this way. When the control 09:21
`
`and address information is set by the computer 09:21
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 17
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`system -- let me reask. 09:21
`
` What do you mean by the control and 09:21
`
`address information is packetized? 09:21
`
` A. I'm not sure what you're referring to. 09:21
`
` Q. Well, you said in the Rambus memory module 09:21
`
`the control and address information is packetized, 09:21
`
`correct? 09:22
`
` A. I don't recall if that's exactly what I 09:22
`
`said. But in the Rambus memory modules that I was 09:22
`
`familiar with at one point in time, which again are 09:22
`
`only a subset of the Rambus products, the control 09:22
`
`and address -- as best as I can recall, I don't have 09:22
`
`that protocol in front of me -- the control and 09:22
`
`address data would be broken up into chunks, I 09:22
`
`believe, 8 bits at a time, and sent as a sequence 09:22
`
`over a limited number of wires and then 09:22
`
`reconstituted on the module itself into individual 09:22
`
`signals. 09:22
`
` Q. Okay. So in the Rambus memory module that 09:22
`
`you are familiar with from the early 2000s, the 09:22
`
`control and address information is set in bits and 09:22
`
`reconstituted by a larger component on the memory 09:23
`
`module; is that correct? 09:23
`
` MR. CHANDLER: Object to form. 09:23
`
` THE WITNESS: Now, first understanding 09:23
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 18
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`clearly that this is -- that the module in 09:23
`
`Exhibit 2100 is not one that I'm familiar with. 09:23
`
` But my recollection is that, as I stated 09:23
`
`before, that in the Rambus protocol for the devices 09:23
`
`that I was familiar with, control and address 09:23
`
`information would be sent along a relatively narrow 09:23
`
`channel as a sequence of packets and then 09:23
`
`reconstituted into individual signals by the logic 09:23
`
`functions on the -- within the Rambus chips. 09:24
`
`BY MS. ZHONG: 09:24
`
` Q. Okay. 09:24
`
` A. But again, I haven't seen those documents 09:24
`
`in a long time, so I really can't tell you any of 09:24
`
`the details about it without reviewing them again. 09:24
`
` Q. When you say "reconstituted into 09:24
`
`individual signals by the logic function within the 09:24
`
`Rambus chips," what do you mean by "reconstituted"? 09:24
`
` A. Again, it's been a long time since I 09:24
`
`reviewed this protocol. But my best recollection is 09:24
`
`that the individual selection signals to select 09:24
`
`columns, rows, banks, and die would be reconstituted 09:25
`
`within some logic somewhere on the memory module. 09:25
`
` Q. So the form in which the logic function 09:25
`
`receives the information from the memory system -- 09:25
`
`let me withdraw it and reask. 09:25
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 19
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` So the input -- inputs that the logic 09:25
`
`function receives from the computer system differs 09:25
`
`from the output of that logic function; is that 09:25
`
`correct? 09:25
`
` MR. CHANDLER: Object to form. 09:25
`
` THE WITNESS: I don't understand your 09:25
`
`question. 09:25
`
`BY MS. ZHONG: 09:25
`
` Q. Which part don't you understand? 09:25
`
` A. I don't understand what the question is. 09:25
`
` Q. So the logic function on the Rambus DIMMs 09:26
`
`that you are familiar with receives packetized 09:26
`
`control and address information; is that correct? 09:26
`
` A. In a very general sense. Again, I 09:26
`
`haven't -- I haven't reviewed that protocol in 09:26
`
`decades. 09:26
`
` Q. Okay. And that logic function then 09:26
`
`reconstitute those that receive the packetized 09:26
`
`information and generates signals for output to the 09:26
`
`DRAMs; is that correct? 09:26
`
` MR. CHANDLER: Objection. Form. 09:26
`
` THE WITNESS: Yeah, it depends how you're 09:26
`
`defining some of those terms. I'm not sure what -- 09:26
`
`what -- I mean I answered the question using 09:26
`
`language I was comfortable with, but I don't know 09:26
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 20
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`how -- what -- how you're defining some of the words 09:26
`
`in your question. 09:26
`
`BY MS. ZHONG: 09:27
`
` Q. Which words don't you understand? 09:27
`
` A. Can you go through your question again? 09:27
`
`In particular, I don't understand what you mean when 09:27
`
`you say the signals go to the DRAM. 09:27
`
` Q. And the logic function on the Rambus 09:27
`
`memory module you're familiar with then 09:27
`
`reconstitutes the control and address information 09:27
`
`that it receives and it generates output signals; is 09:27
`
`that correct? 09:27
`
` A. It depends what you mean by output 09:27
`
`signals. But in the very broadest sense where there 09:27
`
`is at least some logic gate that outputs that 09:27
`
`signal, that would be true based on my recollection 09:27
`
`without reviewing the protocol. 09:27
`
` Q. Okay. And the signals that are output by 09:27
`
`the logic function, where do they go? 09:27
`
` A. I don't remember. Again, I haven't seen 09:28
`
`this design in decades. I'd have to see the actual 09:28
`
`design to give you any more details. 09:28
`
` Q. Do they go to the memory devices on the 09:28
`
`module? 09:28
`
` A. Like I said, I don't -- I don't recall. 09:28
`
`Veritext Legal Solutions
`866 299-5127
`
`Page 21
`
`Netlist Ex 2056
`Samsung v Netlist
`IPR2022-00996
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
` Q. And is the -- and is the information 09:28
`
`that's output by the logic function still 09:28
`
`packetized? 09:28
`
` MR. CHANDLER: Objection. Form. 09:28
`
` THE WITNESS: Where and when? 09:28
`
`BY MS. ZHONG: 09:28
`
` Q. Pardon? 09:28
`
` A. Where and when? 09:28
`
` Q. You mention that the logic function would 09:28
`
`take the packetized control and address information 09:29
`
`it receives and reconstitute to generate the output 09:29
`
`signal. So is that output signal still packetized? 09:29
`
` MR. CHANDLER: Objection. Form. 09:29
`
` THE WITNESS: Again, you know, at some 09:29
`
`point the details become important and we have to 09:29
`
`look at an actual circuit and look at an exact spot 09:29
`
`in the circuit. 09:29
`
` But eventually those -- that information 09:29
`
`about control and addresses is not in the same 09:29
`
`packetized form that it is transmitted over the 09:29
`
`Rambus channel. But -- but, you know, it still -- 09:29
`
`it's sequential. But at some point we would