[00:00:00] Speaker 04:
Yes, your honor.

[00:00:05] Speaker 04:
You can begin, counsel.

[00:00:06] Speaker 04:
May it please the court.

[00:00:10] Speaker 04:
Lambert does not disclose the following claim limitation.

[00:00:14] Speaker 02:
Do you agree claim one of the 789 patent is ill ester due?

[00:00:19] Speaker 04:
Yes, your honor.

[00:00:23] Speaker 02:
Does the 789 patent define arbiter and decoder?

[00:00:29] Speaker 02:
And if so, where?

[00:00:38] Speaker 02:
I'm asking because on page eight of the blue brief, you say, the memory interface includes an arbiter for selectively providing access for the first devices and the decoder to the memory.

[00:00:51] Speaker 04:
Right.

[00:00:51] Speaker 04:
The reason why you have a decoder.

[00:00:52] Speaker 02:
No, no.

[00:00:53] Speaker 02:
My question was, does it define them?

[00:00:56] Speaker 04:
I don't know if there's an explicit definition, but I think the purposes of the decoder, as well as an orbiter, are described in the packet.

[00:01:07] Speaker 04:
The purpose of the decoder is you have compressed data.

[00:01:11] Speaker 04:
Whenever you have a movie, there's so much data that it has to be compressed before you put it on a medium or before you transmit it.

[00:01:19] Speaker 04:
The decoder, as you're watching the movie, for example, will take that compressed form

[00:01:25] Speaker 04:
and decompress it so it can be displayed.

[00:01:28] Speaker 02:
Don't waste your time with attorney argument on it.

[00:01:30] Speaker 02:
If it's not there, it's not there.

[00:01:42] Speaker 02:
You say on page 18 of the blue brief, counsel, that the PTAB improperly shifted the burden to you by

[00:01:55] Speaker 02:
quote, by basing its anticipation finding on its belief that there is nothing in reference to figure 21 in its corresponding description suggesting that when PCI bus 120 is placed in multimedia mode, multimedia device 144D is precluded from accessing main memory 110 and operating in real time.

[00:02:18] Speaker 02:
How is that burden shifting?

[00:02:20] Speaker 04:
Sure, it's impossible to understand these things unless you look at the picture.

[00:02:25] Speaker 04:
There's a very good picture on page 25 of the blue brief.

[00:02:29] Speaker 04:
It's an annotated, it's a figure 21 from Lambric that was annotated by the petitioner to make it easier to understand what their anticipation assertion is.

[00:02:41] Speaker 04:
So if you look at item 144D and it's in green and it's identified as the decoder,

[00:02:49] Speaker 04:
That's what they allege is the decoder.

[00:02:52] Speaker 04:
The memory that is shared is in red, and that's item 110.

[00:02:58] Speaker 04:
In order for data to be transferred in real time, so the particular limitation from claim one that we're discussing is a decoder that requires access to the shared memory sufficient to maintain real time operation.

[00:03:16] Speaker 04:
In order for that decoder

[00:03:18] Speaker 04:
to meet this claim limitation.

[00:03:22] Speaker 04:
Each component along that path from the decoder at 144D to the memory at 110 needs to be able to operate in real time, has to be capable of operating in real time.

[00:03:37] Speaker 04:
So what is along that path?

[00:03:39] Speaker 04:
You see the PCI bridge subset, item 106.

[00:03:43] Speaker 04:
You also see another bus, item number 108,

[00:03:47] Speaker 04:
which is between 106 and 110.

[00:03:50] Speaker 04:
So in order to justify a finding of anticipation, what was required by the board was to answer specific questions on each of these components.

[00:04:02] Speaker 04:
And the specific question that they should have asked and then answered is, does the PCI bridge subset 106, which is along the path from the decoder to the shared memory,

[00:04:16] Speaker 04:
Does it operate in real time and access the PCI bus when it is in real time mode?

[00:04:23] Speaker 04:
That was a very important question.

[00:04:25] Speaker 04:
It's something we emphasize in our patented response.

[00:04:28] Speaker 04:
We emphasize that I spent several minutes on that issue at the hearing.

[00:04:34] Speaker 04:
Another very important issue, which is necessary for a finding of anticipation, the memory bus 108.

[00:04:43] Speaker 04:
The question they should have asked is,

[00:04:46] Speaker 04:
Does the memory bus along the path from the decoder to the shared memory operate in real time and access the PCI bus when it is in real time mode?

[00:04:59] Speaker 04:
There's no dispute that each of these components along the path has to operate in real time mode in order to meet this claim limitation.

[00:05:08] Speaker 04:
Think if you're driving a fast car on the track and you're trying to go 150 miles an hour.

[00:05:14] Speaker 04:
Each

[00:05:15] Speaker 04:
segment along the track has to be capable of maintaining that speed.

[00:05:21] Speaker 04:
If you had a segment where they were doing some reconstruction and there was dirt with potholes, then it wouldn't be capable of sustaining that speed.

[00:05:31] Speaker 04:
You'd go into a pothole and your struts would break.

[00:05:33] Speaker 04:
You'd probably kill yourself and wreck the car.

[00:05:36] Speaker 04:
So it's similar here.

[00:05:38] Speaker 04:
It's analogous to the principle... Not the way I drive.

[00:05:42] Speaker 04:
It's analogous to the principle that

[00:05:45] Speaker 04:
Each component along this path between the decoder and the main memory has to operate in real-time mode.

[00:05:54] Speaker 04:
And what the board should have done was to ask that question for each of these components.

[00:06:00] Speaker 04:
The board did not do that.

[00:06:02] Speaker 04:
Instead, this is what the board puts in its final written decision.

[00:06:10] Speaker 02:
I'm quoting.

[00:06:11] Speaker 02:
There is nothing... Council, you're saying that... I think you're saying that...

[00:06:15] Speaker 02:
Failure to answer these questions equates burden shifting.

[00:06:19] Speaker 02:
That's part one.

[00:06:21] Speaker 04:
Part two I'm going to read to you.

[00:06:23] Speaker 01:
What page are you reading from please?

[00:06:31] Speaker 04:
I'm reading from the final written decision.

[00:06:57] Speaker 04:
21.

[00:06:58] Speaker 04:
Thank you.

[00:07:00] Speaker 04:
Okay, and I'll read it to you.

[00:07:05] Speaker 04:
Here's what the board says.

[00:07:06] Speaker 04:
There is nothing in reference to figure 21 and its corresponding description suggesting that when PCI bus 120 is placed in multimedia mode, multimedia device 144D is precluded from accessing main memory 110 and operating in real time.

[00:07:28] Speaker 04:
That is improper burden shifting.

[00:07:30] Speaker 03:
You know, we get these arguments from the petitioners all the time when they're rejecting your positive arguments as insufficient and say that's burden shifting.

[00:07:40] Speaker 03:
That sentence follows positive statements from the petitioner's declarant that says exactly the opposite.

[00:07:47] Speaker 03:
How is that burden shifting when all the board is doing is saying, we believe them, not you?

[00:07:53] Speaker 04:
Because they didn't approach this issue with enough specificity.

[00:07:57] Speaker 03:
Well, that's for us to decide.

[00:07:59] Speaker 03:
That's not burden shifting.

[00:08:01] Speaker 03:
We can look at their conclusion that considering the disclosures in the land draft, and I'm reading from the sentences right before what you cited to us, Dr. Stone testifies that one of ordinary skill and an art would have understood and goes on to get to the exact conclusion you think the board didn't make.

[00:08:21] Speaker 03:
You may disagree with that, but it's not burden shifting.

[00:08:24] Speaker 03:
It's disagreement with their

[00:08:26] Speaker 03:
The petitioner is expert.

[00:08:28] Speaker 04:
Well, Your Honor, I respectfully disagree for the following reason.

[00:08:32] Speaker 04:
It's undisputed that each of these components between what's alleged to be the decoder and the main memory has to be the decoder.

[00:08:39] Speaker 02:
Let's counsel, counsel.

[00:08:40] Speaker 02:
Stop talking.

[00:08:43] Speaker 02:
If the board makes a finding of fact after weighing evidence, do you believe that that always shifts the burden to the other side?

[00:08:53] Speaker 04:
No, I would not say.

[00:08:57] Speaker 02:
That seems to be what you're arguing.

[00:08:59] Speaker 04:
Well, I was focusing on the statement that's in the final written decision.

[00:09:03] Speaker 04:
But my argument isn't that the board's findings are not supported by substantial evidence.

[00:09:10] Speaker 04:
My argument is more specific than that.

[00:09:14] Speaker 04:
My argument is that the board did not make the findings that were necessary in order to determine whether this claim limitation was met by this particular reference, Lambric.

[00:09:24] Speaker 04:
It was necessary.

[00:09:25] Speaker 04:
But it said it did.

[00:09:27] Speaker 04:
No, there's no specificity that it never said that the PCI bridge, it never said that the PCI bridge subset is capable of sustaining real time operation.

[00:09:37] Speaker 04:
It never said the memory bus 108.

[00:09:40] Speaker 03:
It didn't go through piece by piece, but it said in view of Lambert, we think this discloses everything and it is capable of doing it in real time.

[00:09:50] Speaker 03:
Isn't that what those sentences in 21 that, that I decided to you say?

[00:09:59] Speaker 03:
I mean, maybe you would have preferred them to... That glosses over the important issues.

[00:10:05] Speaker 03:
Well, this is a substantial evidence case.

[00:10:07] Speaker 03:
And if the board makes the conclusion that this discloses everything and explains it in a reasonable manner, and it's supported by substantial evidence, which you seem to concede... I may have misspoke.

[00:10:19] Speaker 04:
I don't concede that it's supported by substantial evidence.

[00:10:23] Speaker 04:
You just did.

[00:10:24] Speaker 04:
Well, I may have misspoke.

[00:10:27] Speaker 04:
It's not supported by substantial evidence because they did not dig deep enough to make the proper inquiries to determine whether or not each of these components is capable of sustaining.

[00:10:41] Speaker 02:
Just so you're aware, you're into your rebuttal time, but keep going.

[00:10:45] Speaker 04:
It's up to you.

[00:10:47] Speaker 04:
Well, I have some other arguments with respect to another claim limitation where there's a bus having a sufficient bandwidth

[00:10:56] Speaker 04:
to enable the decoder to access the memory and operate in real time when the first device simultaneously accesses the bus.

[00:11:07] Speaker 04:
In this case, the board did not rely solely on Figure 21.

[00:11:12] Speaker 04:
Rather, it relied on both Figures 6 and 21.

[00:11:16] Speaker 04:
It improperly picked and chose from different unrelated embodiments of Figures 6 and 21 in order to establish that Lambert

[00:11:26] Speaker 04:
disclose this particular limitation.

[00:11:29] Speaker 01:
Can you tell me what's wrong with this understanding of what I took to be going on, which is that the board looks at Figure 21, some of the language, particularly in Column 27 of Figure 21, and reads Dr. Stone and concludes, you know, this looks like Lambrecht is teaching

[00:11:52] Speaker 01:
a use of even in multimedia mode of the main memory.

[00:12:00] Speaker 01:
And then you say, you come back and say, well, one reason that can't be so

[00:12:10] Speaker 01:
is that there's something missing.

[00:12:12] Speaker 01:
Is it the mode logic or something?

[00:12:14] Speaker 01:
Piece?

[00:12:14] Speaker 04:
Interface logic.

[00:12:15] Speaker 01:
The interface logic.

[00:12:16] Speaker 04:
If you look at the figure, the interface logic is on at the code.

[00:12:19] Speaker 01:
Right.

[00:12:19] Speaker 01:
And I thought it was in that context.

[00:12:23] Speaker 01:
And the interface logic is not claimed elements, just something that you say is, as a practical matter, necessary to get the shared access to the main memory at multimedia mode speeds.

[00:12:37] Speaker 01:
And the board says,

[00:12:40] Speaker 01:
Dr. Stone says no.

[00:12:42] Speaker 01:
And what's more, responding to your practical argument about a missing unclaimed component says figure six kind of tells us you don't need what you say is needed.

[00:12:55] Speaker 01:
So it's not combining elements from figure six to figure 21, but using figure six as evidence that your counter argument to their reading of column 27 is wrong.

[00:13:08] Speaker 04:
I understand what you're saying.

[00:13:09] Speaker 04:
Our argument is that Lambric indicates that you need this interface logic in the decoder in order to access the bus when it's operating in real time mode.

[00:13:19] Speaker 04:
That's what's in there.

[00:13:21] Speaker 04:
And that's missing from the PCI bridge subset, for example.

[00:13:25] Speaker 04:
And then they go to figure six and they say, well, in figure six it's not there.

[00:13:29] Speaker 04:
But figure six is a very different embodiment.

[00:13:31] Speaker 04:
Figure six has two hardwired buses.

[00:13:34] Speaker 04:
It doesn't have a bus that can operate either in multimedia mode or real time

[00:13:39] Speaker 04:
So that's why it's not that interface logic is not required in Figure 6, but it is absolutely required in Figure 21.

[00:13:50] Speaker 04:
So they should not have been referring to Figure 6 because Figure 6 has two hardwired buses, so that shows nothing.

[00:13:57] Speaker 04:
The interface logic is absolutely required when you have a bus that has two different modes, one real-time and one not real-time.

[00:14:07] Speaker 04:
You want to reserve the rest of your time,

[00:14:09] Speaker 04:
Sure, the only thing I'll take five seconds.

[00:14:13] Speaker 04:
Take whatever you want, but I'm not restoring any.

[00:14:15] Speaker 04:
I understand.

[00:14:17] Speaker 04:
The board made a decision on the claim construction of real time and then it didn't use that claim construction decision.

[00:14:26] Speaker 04:
It used a limit that was in another reference to say what is real time and what is not.

[00:14:32] Speaker 04:
And we feel that wasn't proper because the board didn't use the claim construction that was proper.

[00:14:37] Speaker 04:
in its anticipation analysis.

[00:14:40] Speaker 04:
Thank you, counsel.

[00:14:50] Speaker 02:
Ready to proceed?

[00:14:52] Speaker 00:
Go ahead.

[00:14:52] Speaker 00:
May it please the court, contrary to the appellant's attempts to characterize what this case is about, the issues before the court are purely factual, more made factual findings, and they're supported by substantial evidence.

[00:15:04] Speaker 00:
I'll begin by drawing the court's attention to Appendix 29,

[00:15:07] Speaker 00:
and near the bottom.

[00:15:11] Speaker 00:
In response to the suggestion that the board did not address each issue specifically enough, at the bottom of 29, the board says that figure 21, its corresponding description in Lambert, together with the knowledge of one ordinary skill in the art, we agree with petitioners that PCI bus 120 is placed in multimedia mode.

[00:15:32] Speaker 00:
It supports data transfers at a rate sufficient to enable

[00:15:35] Speaker 00:
all components coupled there too, specifically calling out PCI-Bit-Bridge-Chip-Set-106 interfaces with the main memory and the memory bus 108 to operate in real time.

[00:15:48] Speaker 00:
The suggestion that the board didn't address the issue specifically enough or make factual findings on that is not correct.

[00:15:54] Speaker 00:
They made all the factual findings that they were supposed to and just happened to find against the opponent in this case.

[00:16:00] Speaker 00:
There's substantial evidence to support that in the testimony of Dr. Stone,

[00:16:05] Speaker 00:
in Lambric itself, a mere disagreement with how this specification is read, according to Inry Garth's side, is not sufficient to say that there isn't substantial evidence.

[00:16:14] Speaker 00:
Disagreement alone isn't enough.

[00:16:15] Speaker 00:
The Supreme Court has said that as well.

[00:16:17] Speaker 00:
So the board made factual findings.

[00:16:20] Speaker 00:
At the end of the day, the appellant made his arguments to the board.

[00:16:24] Speaker 00:
They disagreed with them.

[00:16:25] Speaker 01:
A lot of them had to do with- Can you focus not so much on that high level of generality, but the very specific point about

[00:16:34] Speaker 01:
and I'm going to use the shorthand and hope that you will unpack it fairly, the missing interface circuit in order to make this work.

[00:16:41] Speaker 00:
The missing interface circuit.

[00:16:43] Speaker 01:
Isn't that what it was called, what your colleague there called, as part of his criticism of the invocation of Figure 6?

[00:16:53] Speaker 00:
Correct.

[00:16:53] Speaker 00:
So the argument there stems from the fact that Dr. Stone testified, as you pointed out, that Lambert 21 and the Figure 21 discloses all the claim elements.

[00:17:04] Speaker 00:
And that one of Skill in the Art would understand that the chips at 106 would interface with the PCI bus, 120, when in multimedia mode.

[00:17:12] Speaker 00:
The response that came back from the patent owner was that figure one somehow limited the understanding or the operability of what's going on in figure 21.

[00:17:24] Speaker 00:
And the response from Dr. Stone was, that's not correct.

[00:17:27] Speaker 00:
You can't take what's happening in figure one, which had two separate buses, and 130.

[00:17:32] Speaker 01:
Right.

[00:17:35] Speaker 01:
Let me try to identify my confused state of thinking, maybe confused.

[00:17:41] Speaker 01:
I took it one of their arguments was figure 21 is nothing but smushing together the two things that are in figure one and each mode does only the thing that in figure one is in two separate buses.

[00:17:58] Speaker 01:
And there's a perfectly decent argument for saying

[00:18:02] Speaker 01:
That's, you're over reading the characterization of figure 21 where the spec says this is similar to figure one.

[00:18:11] Speaker 01:
There's more than just putting together two things with each mode perfectly mirroring the two buses.

[00:18:17] Speaker 01:
Then I took it they had a separate argument.

[00:18:20] Speaker 01:
A separate argument that says even if you might otherwise read column 27 and whatnot to

[00:18:28] Speaker 01:
describe figure 21 as showing multimedia mode that gets back to the main memory.

[00:18:36] Speaker 01:
There's a reason you can't do that.

[00:18:38] Speaker 01:
And that's because there's some technical piece that would be required in order for that to work.

[00:18:45] Speaker 01:
And that technical piece is not shown.

[00:18:48] Speaker 01:
I think the interface circuit are the words I'm remembering, but I may be misremembering the words.

[00:18:53] Speaker 01:
Can you focus on that second piece?

[00:18:55] Speaker 00:
Right.

[00:18:55] Speaker 00:
That second piece also stems from the earlier argument about it.

[00:19:00] Speaker 00:
And Dr. Stone disagreed.

[00:19:02] Speaker 00:
He said that one of skill in the art would read that to understand that all the circuitry necessary in order for it to operate are there.

[00:19:09] Speaker 00:
And if you take a look at Appendix 1745, it actually defines chipset 106, which is what we're talking about.

[00:19:17] Speaker 00:
The chipset 106 interfaces with the bus and the memory bus.

[00:19:21] Speaker 00:
And it says it is a Triton chip from Intel or a like one that accommodates real-time communication.

[00:19:29] Speaker 00:
So the specification in the Lambert disclosure itself says that it is capable of the real-time interface.

[00:19:37] Speaker 00:
Dr. Stone said it is.

[00:19:39] Speaker 00:
Their argument, again, goes back to figure one and the limiting that's going on and their interpretation of what's happening with respect to the 106 disclosed in that figure and how it operates

[00:19:49] Speaker 00:
the bus 130 and the bus 120, and what's happening there.

[00:19:53] Speaker 00:
The board discredited Dr. Thornton's testimony on that regard, and credited Dr. Stone's testimony on that regard to say, to one of skill in the art, they would understand that that is, in fact, in and disclosed in Lamerick, Figure 21.

[00:20:11] Speaker 00:
So the board, as Judge Ronto pointed out, the board made specific findings with regard to Figure 21.

[00:20:19] Speaker 00:
and the fact that Dr. Stone said that all elements are there, and then added an additional element, which is that Figure 6 supported his understanding of what was already disclosed in Figure 21.

[00:20:30] Speaker 00:
So it is not putting in elements that were missing or supplying something that wasn't there.

[00:20:35] Speaker 00:
Everything is disclosed in Figure 21.

[00:20:37] Speaker 00:
The board instituted on Figure 21 and had a decision on Figure 21.

[00:20:43] Speaker 00:
With respect, Judge Wallach, to your question about whether

[00:20:47] Speaker 00:
The 789 defines decoder and arbiter.

[00:20:50] Speaker 00:
You'll find at appendix 55, there's a discussion of the decoder.

[00:20:56] Speaker 00:
Same with appendix 53.

[00:20:58] Speaker 00:
I don't believe that there's a specific definition of arbiter, but there is a discussion of arbiter at appendix 57 as well.

[00:21:05] Speaker 01:
With regard to real time.

[00:21:06] Speaker 01:
And that's just so interesting.

[00:21:07] Speaker 01:
I mean, arbiter is just a device that says, various people want access to me, and I have to decide who goes first.

[00:21:12] Speaker 01:
Exactly.

[00:21:14] Speaker 00:
With respect to the real time,

[00:21:17] Speaker 00:
There is no different claim construction.

[00:21:19] Speaker 00:
It's the same claim construction.

[00:21:21] Speaker 00:
What the board did is he put in the context of the claim limitation itself, which is the decoder must operate in real time, but the bandwidth on the shared bus must be sufficient for it to do so.

[00:21:33] Speaker 00:
And in determining whether or not it's sufficient, the board looked to the disclosure to determine what that bandwidth would be required in order to get to real time communication or real time operation of the decoder.

[00:21:44] Speaker 00:
Unless there are any further questions.

[00:21:47] Speaker 00:
Nope.

[00:21:48] Speaker 00:
We ask the court to affirm.

[00:21:50] Speaker 00:
Thank you, counsel.

[00:21:53] Speaker 02:
You have 28 seconds, counsel.

[00:21:57] Speaker 04:
I'll be very brief.

[00:21:58] Speaker 04:
Figure 21, it shows the decoder.

[00:22:01] Speaker 04:
It shows that the interface logic is there.

[00:22:04] Speaker 04:
So it's needed to access the bus when it's operating in real-time mode.

[00:22:09] Speaker 04:
It's not present in the PCI chipset.

[00:22:14] Speaker 04:
And that is a component along the path

[00:22:16] Speaker 04:
which must operate in real time in order for the decoder to access the memory in real time.

[00:22:22] Speaker 04:
Thank you.