[00:00:00] Speaker 02:
The first case is Netlist versus Samsung et al.

[00:00:06] Speaker 02:
2024-1521, Mr. Crudo.

[00:00:10] Speaker 00:
Thank you, Your Honor, and may it please the Court.

[00:00:13] Speaker 00:
Our appeal is based on a simple premise, and that is it was Samsung's verdict to explain in its petition how the obviousness combination here was supposed to work.

[00:00:23] Speaker 00:
It didn't do that, and the board let Samsung off the hook, which led to a multi-tiered APA violation.

[00:00:29] Speaker 00:
The petition here argued that one component in the hearing issue reference

[00:00:34] Speaker 00:
would somehow control another unrelated component to do something it was never designed to do.

[00:00:40] Speaker 00:
Samsung didn't explain how that would be done, nor did it ever suggest that Hiroshi would be modified so that it could be done.

[00:00:47] Speaker 04:
Yeah, but the reason the board disagreed with you is it said that

[00:00:53] Speaker 04:
they merely treated what was the example cited in the petition as an example.

[00:00:59] Speaker 04:
And they're allowed to evaluate obviousness in light of but explicit disclosures and a person skilled in the art, which is exactly what the petitioner pointed to.

[00:01:11] Speaker 04:
So why is that problematic?

[00:01:13] Speaker 00:
Because, Your Honor, the board didn't make that statement until its decision on rehearing.

[00:01:17] Speaker 00:
And even then, Samsung, even though it used the phrase EG, which I think Your Honor is referring to as an exemplary example of how its combination would work, it never proposed any other example of how the implementation would be performed.

[00:01:34] Speaker 00:
And we see that at appendix 1333, which is Samsung's expert declaration,

[00:01:40] Speaker 00:
There he proposed that Hiraishi's re-leveling circuit would determine the amount of delay to apply using Butt's throat-centering algorithm.

[00:01:50] Speaker 00:
And then that re-leveling circuit would somehow control the delay circuit to apply that delay.

[00:01:56] Speaker 00:
He never suggested borrowing circuitry from Butt.

[00:01:59] Speaker 00:
And he never suggested changing Hiraishi's circuitry.

[00:02:02] Speaker 00:
But are we talking about paragraph 63 in Butt?

[00:02:06] Speaker 04:
Because that's what the briefs were discussing.

[00:02:10] Speaker 04:
And you allege that was new in the hearing.

[00:02:13] Speaker 04:
And that's not correct, is it?

[00:02:14] Speaker 00:
Well, that's the first time that the board had ever relied on paragraph 63 to address this particular argument.

[00:02:22] Speaker 00:
Did they do it?

[00:02:23] Speaker 00:
The board at appendix 18 was addressing a separate argument that we allege below that is no longer relevant on appeal.

[00:02:34] Speaker 00:
That is that Butt requires a computer, that Harashi lacks.

[00:02:38] Speaker 00:
The board rejected that argument based on Butt's paragraph 63.

[00:02:43] Speaker 00:
But importantly,

[00:02:44] Speaker 00:
Samsung's expert never even cited paragraph 63 of the butt reference.

[00:02:50] Speaker 00:
And so Samsung never thought that that paragraph was relevant to its disclosure.

[00:02:55] Speaker 03:
I think the problem here is- Do you agree that the board is not bound to only cite exactly what Samsung cited?

[00:03:02] Speaker 00:
That's true, Your Honor.

[00:03:03] Speaker 00:
But the board is bound to the obviousness argument that the petitioner raises.

[00:03:08] Speaker 00:
And here, Samsung was very clear in its opening declaration and petition that it wasn't relying on a theory that required modification of Hiraishi.

[00:03:18] Speaker 00:
Their expert acknowledged at his deposition that he wasn't relying on modification.

[00:03:23] Speaker 00:
So we then came back and showed that the read leveling circuit had no way to control the delay circuit without modification.

[00:03:30] Speaker 00:
At that point, Samsung was at a crossroad.

[00:03:33] Speaker 00:
It could have proffered supplemental expert declaration testimony saying why we were wrong or shifting to a modification theory, but it didn't do that.

[00:03:43] Speaker 03:
What appendix page can you point us to for where you said the expert acknowledged that it wasn't relying on?

[00:03:49] Speaker 00:
Yes, Your Honor.

[00:03:50] Speaker 00:
At appendix 7-7-9-2, beginning in 2014, this is the deposition of Sam Sung's expert, and he's discussing modifying the Hiraishi reference.

[00:04:06] Speaker 00:
He says, for the first time at his deposition, he suggested, well, if Hiraishi doesn't have the circuitry,

[00:04:11] Speaker 00:
you could borrow it from but.

[00:04:13] Speaker 00:
And he says, well, this is not in my report.

[00:04:16] Speaker 00:
I would be generating new information.

[00:04:19] Speaker 00:
I would be generating new opinions.

[00:04:21] Speaker 00:
But I mean, you would have to do whatever is necessary in order to implement but.

[00:04:26] Speaker 00:
So he acknowledged that this was never set forth in his opening declaration.

[00:04:31] Speaker 00:
And critically, Samsung never proffered a supplemental declaration.

[00:04:35] Speaker 00:
So we reasonably believed that the modification theory was not at play.

[00:04:39] Speaker 00:
It wasn't until the decision on rehearing that the board first mentioned a theory whereby he or she could be modified.

[00:04:50] Speaker 00:
We had no opportunity to respond to that position, and that's the quintessential APA violation under Magnum Oil.

[00:04:57] Speaker 00:
And I think it's telling that on appeal, Samsung says at page 48 of its red brief that the petition

[00:05:04] Speaker 00:
never argued that hirishi would not be modified.

[00:05:08] Speaker 00:
But that improperly flips the burden.

[00:05:10] Speaker 00:
It's Samsung that bore the burden to establish reasonable expectation of success.

[00:05:15] Speaker 00:
And the absence of evidence or argument in the petition doesn't satisfy that burden.

[00:05:20] Speaker 03:
In the pages you point us to, counsel, it talks about, or the answer talks about, I felt it's already there.

[00:05:27] Speaker 03:
Or something similar said on the next page, appendix page 7793.

[00:05:31] Speaker 03:
What is being referred to by the witness there?

[00:05:34] Speaker 00:
He's referring to circuitry and hiriyoshi for implementing the combination.

[00:05:40] Speaker 00:
But to your honor's point, it's a little difficult to tell precisely because he didn't discuss what circuitry would be needed.

[00:05:47] Speaker 00:
He didn't discuss what modifications would be required, and neither did the board.

[00:05:52] Speaker 00:
His position was you would do whatever is necessary to implement but.

[00:05:57] Speaker 00:
And this court's cases, such as Personal Web and, in fact, the Samsung versus Elm 3DA case, say that that's not enough, especially when the technology is complicated.

[00:06:08] Speaker 00:
The board has to provide a clear, evidence-backed analysis of how the combination is supposed to work.

[00:06:15] Speaker 00:
And the board didn't do that, and neither did Samsung.

[00:06:19] Speaker 00:
There is a third problem with how the way this case unfolded.

[00:06:23] Speaker 00:
And that's the board didn't resolve factual disputes related to the arguments that Samsung actually proposed.

[00:06:31] Speaker 00:
And we know that because Samsung defends the board's ruling on appeal by invoking fact findings that the board never even made.

[00:06:40] Speaker 00:
And there are three in particular.

[00:06:42] Speaker 00:
The first is that we showed that Hiraishi's delay circuit is designed to apply only a fixed delay, not a variable one.

[00:06:50] Speaker 00:
And that's important because our claims require a variable delay based on prior operations.

[00:06:57] Speaker 00:
So the circuit would somehow have to be redesigned.

[00:07:00] Speaker 00:
Now, Samsung disputed that below and on appeal, but the board never resolved that issue.

[00:07:04] Speaker 00:
It simply didn't address the argument.

[00:07:07] Speaker 00:
That's the first point.

[00:07:08] Speaker 00:
The second is that we showed Kiriishi's delay circuit and read leveling circuit for separate components that serve unrelated functions.

[00:07:18] Speaker 00:
The read leveling circuit simply controls the input buffers that allow access into the buffer.

[00:07:24] Speaker 00:
The delay circuit delays strobe signals.

[00:07:27] Speaker 00:
They're not related at all.

[00:07:29] Speaker 00:
Again, Samsung disputed that below and disputes it on appeal, but the board never addressed that argument at all.

[00:07:36] Speaker 00:
And then finally, we showed that the combination would require not only connecting the two components, but additional feedback mechanism

[00:07:47] Speaker 00:
Here, their expert actually agreed with us at appendix 7792 to 7793.

[00:07:53] Speaker 00:
He agreed that some additional programming and logic would be required, none of which was discussed in the petition.

[00:08:00] Speaker 00:
The board didn't address this issue either.

[00:08:03] Speaker 00:
Now, I think this all goes to show that the board sidestepped these critical factual disputes by changing the obviousness combination.

[00:08:11] Speaker 00:
And for that reason, it's sort of a three-tiered APA violation, as I noted.

[00:08:20] Speaker 00:
I do want to, if there are no further questions about reasonable expectation of success, I do want to say just a few words about motivation to combine.

[00:08:30] Speaker 00:
The parties seem to dispute whether the board committed legal error in applying KSR, but the court doesn't even have to reach that issue because substantial evidence doesn't support the fact finding that served as the linchpin in the board's analysis.

[00:08:45] Speaker 00:
And in fact, the board flatly misinterpreted the evidence.

[00:08:49] Speaker 00:
At the bottom of Appendix 14,

[00:08:51] Speaker 00:
The board found that Kiriishi's data and strobe signals would be misaligned by up to 28 degrees, simply because the prevailing industry standard at the time, the JEDEX standard, says the data signals can be misaligned by 28 degrees.

[00:09:08] Speaker 00:
And based on that premise, inciting our experts' testimony,

[00:09:12] Speaker 00:
The board found that a skilled artisan would correct that misalignment, the 28 degree misalignment, which would lead the skilled artisan to butt strobe centering technique.

[00:09:24] Speaker 00:
But the board got it exactly backwards.

[00:09:26] Speaker 00:
The evidence that it cited doesn't say that the JEDEC 28 degree misalignment is problematic.

[00:09:33] Speaker 00:
It says that it isn't problematic, such that there's nothing to realign.

[00:09:38] Speaker 00:
And we can look at the evidence.

[00:09:40] Speaker 00:
that the board cited.

[00:09:41] Speaker 00:
It cited our expert's testimony at appendix 7559 to 7561.

[00:09:48] Speaker 00:
But there he says that the 28 degrees view ensures that Hirayoshi's buffer operates correctly, such that there is no need to perform additional strobe centering operation.

[00:10:00] Speaker 00:
He also said the same thing at appendix 6052, which the board also cited.

[00:10:05] Speaker 00:
He said that the 28 degree skew is the maximum tolerable skew, meaning that Kiriishi works just fine without any sort of realignment.

[00:10:16] Speaker 00:
So the board erred in relying on that evidence to draw a conclusion that is exactly contrary to our expert's testimony.

[00:10:24] Speaker 00:
The last point I'll make on this is that not even Samsung endorses the board's reasoning here.

[00:10:28] Speaker 00:
On appeal, Samsung doesn't contend that a 28 degree skew or misalignment alone is sufficient to motivate the combination.

[00:10:37] Speaker 00:
Rather, it argues that that skew or misalignment would get worse as the signals propagate through the buffer.

[00:10:45] Speaker 00:
But once again, the board didn't make that finding.

[00:10:48] Speaker 00:
It never reached that issue.

[00:10:50] Speaker 00:
And this court, of course, cannot affirm the board based on findings that were never made.

[00:10:54] Speaker 00:
So we think that the court can and should at least vacate and remand either on reasonable expectations of success or motivation to combine.

[00:11:03] Speaker 03:
Now, there was competing expert testimony from the other side's expert, though, that would have supported up to 28 degrees.

[00:11:09] Speaker 00:
There was competing expert testimony, Your Honor, but the board didn't cite it.

[00:11:12] Speaker 00:
It relied on its interpretation of our expert's testimony.

[00:11:16] Speaker 00:
Nor did the board make any credibility findings anywhere in the decision.

[00:11:21] Speaker 00:
The board never said that it weighed the evidence and credited their expert over ours.

[00:11:26] Speaker 00:
Instead, it interpreted our evidence to say exactly opposite of what it said.

[00:11:34] Speaker 00:
If there are no further questions, Your Honors, I will save the balance of my time.

[00:11:40] Speaker 02:
We will serve it for you, Mr. Horace.

[00:11:50] Speaker 01:
Thank you, Your Honor.

[00:11:50] Speaker 01:
May it please the Court?

[00:11:52] Speaker 01:
This is an appeal that involves substantial evidence.

[00:11:55] Speaker 01:
The factual issues of a reasonable expectation of success and a motivation to combine are all that are raised.

[00:12:02] Speaker 01:
Really what's important to note here is the ordinary skill in the art, because what is ignored in all the briefs we see from that list is that the board adopted exactly the ordinary skill in the art that was raised by Samsung, and that that was undisputed

[00:12:19] Speaker 01:
Most importantly, that ordinary spell in the yard included, and I will point you to the appendix at page 158, and the board adopted appendix six, the ability to structure and operate circuitry, including ASICs, to control a memory circuit.

[00:12:37] Speaker 01:
So it was in front of the board, and if we look at the petition, we see that Samsung relied on it, and that is in the petition, and the citation there is 198,

[00:12:48] Speaker 01:
through 201, that but, when you consider the person or skill in the art, that the functionality but teaches can be used with an ASIC, that's exactly the person or skill in the art has the knowledge to implement those circuits with ASICs.

[00:13:02] Speaker 01:
So the evidence is lined up exactly to do exactly what the board did.

[00:13:07] Speaker 01:
Take the functionality of but and implement it in her rational.

[00:13:11] Speaker 04:
Well, what do you say?

[00:13:12] Speaker 04:
I mean, your friend develops a theme here, or a couple themes.

[00:13:15] Speaker 04:
And one of the themes is APA.

[00:13:18] Speaker 04:
And he mentioned a few examples of where the board failed to address

[00:13:24] Speaker 04:
certain make-fine certain factual findings with respect to the arguments made.

[00:13:29] Speaker 04:
Do you have any?

[00:13:30] Speaker 04:
I don't want you to take every single one in detail, but do you have a global kind of response to that?

[00:13:35] Speaker 01:
Sure.

[00:13:36] Speaker 01:
I mean, the APA requires the board to make findings that support its ultimate conclusions.

[00:13:41] Speaker 01:
It does not require the board to, with regard to every finding, consider every piece of evidence.

[00:13:46] Speaker 01:
And what this court has done historically is look at the finding and look at the record and say, is there evidence to support it?

[00:13:52] Speaker 04:
Well, you used the wrong word, I think.

[00:13:53] Speaker 04:
I mean, maybe I wouldn't say the board isn't required to consider everything.

[00:13:57] Speaker 04:
The question is whether the board is required to articulate a response and an analysis of everything raised.

[00:14:03] Speaker 01:
Right.

[00:14:04] Speaker 01:
In the final written decision, the board

[00:14:06] Speaker 01:
identifies some findings, it doesn't have to consider and discuss every piece of evidence.

[00:14:12] Speaker 03:
As long as the findings... And I will full sign on what Judge Proves is saying.

[00:14:14] Speaker 03:
I will take out the word consider.

[00:14:16] Speaker 01:
Consider.

[00:14:16] Speaker 01:
OK.

[00:14:16] Speaker 03:
I think you're making it in terms of the argument.

[00:14:18] Speaker 03:
But your point, I think, is that the board doesn't have to write in its final written decision about every piece of evidence.

[00:14:23] Speaker 03:
Is that a fair statement?

[00:14:24] Speaker 01:
Certainly true.

[00:14:25] Speaker 01:
And this court has routinely found evidence that supported findings that's in the record and is substantial.

[00:14:32] Speaker 01:
And it said, the board made the finding.

[00:14:35] Speaker 01:
The record supports it.

[00:14:36] Speaker 01:
And that's without the board identifying or discussing any particular piece of evidence.

[00:14:41] Speaker 01:
And I think that's true with every one of the issues raised by the appellant with regard to these particular intricate issues.

[00:14:48] Speaker 01:
I would also point out that the appellant here is creating a standard that is nowhere reflected in this court's law.

[00:14:53] Speaker 01:
They require that, well, you need to show this one particular wire between these two boxes, KSR, Intel VPAC-XPP,

[00:15:01] Speaker 01:
All of those cases, say, have electronic or this type of technology.

[00:15:08] Speaker 01:
But if the skill in the art is enough, you don't need to draw the wires.

[00:15:11] Speaker 01:
You don't need to say, well, here's exactly how the circuit would be done.

[00:15:15] Speaker 01:
And the decision was done.

[00:15:16] Speaker 03:
So what is your response to opposing counsel's arguments?

[00:15:20] Speaker 03:
I was talking back and forth with him about certain expert testimony that he said was interpreted exactly the opposite of what the expert actually said.

[00:15:28] Speaker 03:
And he said that was referring to his expert

[00:15:31] Speaker 03:
and what was written in the founding decision.

[00:15:33] Speaker 03:
Do you have a response to that as well?

[00:15:35] Speaker 01:
Certainly.

[00:15:35] Speaker 01:
First of all, I think the board's interpretation is subject to deference and is a reasonable interpretation.

[00:15:41] Speaker 01:
But it doesn't really matter, because this court can rely on our expert's discussion of those very misalignments.

[00:15:48] Speaker 01:
And under Intel v. PACT XPP, you don't even need to get there.

[00:15:52] Speaker 01:
It's a known issue that there's a timing misalignment.

[00:15:56] Speaker 01:
And that's something that can be addressed by this functions in but.

[00:16:00] Speaker 01:
The person of ordinary skill in the art has the knowledge to implement those functions in various types of circuitry, including, as the board called out, an ASIC, specifically described in BUT, specifically included in the undisputed level of ordinary skill in the art.

[00:16:15] Speaker 01:
So there's no issue that person of skill didn't know how to take a signal that reflects the predetermined delay, provide it to the delay circuit, and have the delay circuit delay.

[00:16:27] Speaker 01:
That's all that's at issue here.

[00:16:29] Speaker 01:
The only claim language that was disputed is claim language of configure two, and then gives three functions.

[00:16:37] Speaker 01:
No specific circuitry, no specific requirement for structure.

[00:16:41] Speaker 01:
This is exactly the kind of language that in KSR, the Supreme Court said, look, if the person of ordinary skill in the art has the skill to do this type of work, and there's a motivation, a known reason, that's it.

[00:16:55] Speaker 01:
And the board found that here.

[00:16:57] Speaker 01:
So we don't have any of these issues of how far the board needed to drill down.

[00:17:01] Speaker 01:
The board didn't.

[00:17:02] Speaker 01:
The board needed to have substantial evidence for the findings it made.

[00:17:06] Speaker 01:
And the findings it made are reflected in that substantial evidence.

[00:17:09] Speaker 01:
And two findings I'd like to just point out for you.

[00:17:12] Speaker 01:
And really, those are the key findings here at all are the board's ability to say that the petitioner's contentions are persuasive.

[00:17:20] Speaker 01:
And specifically, and this is on page 18,

[00:17:28] Speaker 01:
just a little bit below the middle of the page, we find that a person of ordinary skill in the art would have had a reasonable expectation of success in combining Butt's retraining techniques with Hiroshi's memory teachings to achieve the claimed invention.

[00:17:42] Speaker 01:
And then on page 19, we also find, as discussed above, that a person of ordinary skill in the art would have combined the teachings of Hiroshi and Butt and would have had a reasonable expectation of success in arriving at the claimed invention.

[00:17:55] Speaker 01:
the board made the required findings.

[00:17:58] Speaker 01:
Now he complains, he says, oh, the petition, the petition was all about this one connection between these two.

[00:18:03] Speaker 01:
Let's look at the petition, because it lines up with the board's findings.

[00:18:07] Speaker 01:
Specifically, nine times the reply brief in this case cites to a particular sentence on appendix page 202.

[00:18:14] Speaker 01:
That's the sentence that has the EG.

[00:18:17] Speaker 01:
It gives that one example of a particular structure.

[00:18:20] Speaker 01:
But let's look at the four pages before that sentence.

[00:18:23] Speaker 01:
Starting on page 198, the petition says, to the extent one might argue Hiroshi alone does not disclose the limitation, it would also be obvious in view of Butt's disclosure of Act.

[00:18:35] Speaker 01:
And then it sets out the claim language and includes brackets for every part of Butt teaching those particular functional requirements.

[00:18:43] Speaker 01:
Given the level of ordinary skill that was undisputed and the board adopted, given the board's findings, given the petition,

[00:18:51] Speaker 01:
There is no basis for a substantial evidence challenge to this decision.

[00:18:57] Speaker 01:
And this court should affirm.

[00:18:59] Speaker 01:
Unless there are questions, I'll seek my time.

[00:19:02] Speaker 02:
Thank you, counsel.

[00:19:03] Speaker 02:
Mr. Crudo has some rebuttal time.

[00:19:13] Speaker 00:
Thank you, your honor.

[00:19:14] Speaker 00:
Just a few points.

[00:19:15] Speaker 00:
Samsung is trying to hide behind the level of skill in the art.

[00:19:19] Speaker 00:
But that doesn't obviate the burden to establish reasonable expectation of success in the petition.

[00:19:24] Speaker 00:
And we're not arguing for a heightened standard.

[00:19:27] Speaker 00:
The standard is set forth in personal web, which I think is exactly on point.

[00:19:31] Speaker 00:
It says the level of analysis that is required to show motivation to combine in reasonable expectation of success has to be commensurate with the complexity of the technology and the complexity of the combination.

[00:19:45] Speaker 00:
Here, it was Samsung is the master of its own petition

[00:19:49] Speaker 00:
that proposed having one physical component in Hiraishi control another physical component to do something it was never designed to do.

[00:19:57] Speaker 00:
And it relied on but only for the functional purpose of determining how much of a delay to apply.

[00:20:05] Speaker 00:
It didn't rely on but for anything else.

[00:20:09] Speaker 00:
Last point, Your Honors, is while it is true that the board doesn't need

[00:20:12] Speaker 00:
to expressly write out an analysis for every single argument in a case, it does have to address the critical arguments and disputes that are relevant to the combination as presented.

[00:20:26] Speaker 00:
The three disputes that I referenced earlier were critical to resolving this case, and the board sidestepped those disputes by changing the modification, the obviousness combination.

[00:20:37] Speaker 00:
That is not permissible, and that is an APA violation.

[00:20:42] Speaker 00:
If there are no further questions, I will cede.

[00:20:45] Speaker 00:
Thank you, Your Honors.

[00:20:46] Speaker 02:
Thank you.

[00:20:46] Speaker 02:
Thank you to both counsels.

[00:20:48] Speaker 02:
The case is submitted.