[00:00:00] Speaker 00:
Your case is number 20, 1744, Intel Corporation against VLSI technology.

[00:00:07] Speaker 00:
Mr. Fleming.

[00:00:09] Speaker 01:
Good morning and may it please the court, Mark Fleming on behalf of Intel.

[00:00:13] Speaker 01:
The board failed to address one of Intel's key arguments which affects all the claims.

[00:00:18] Speaker 01:
We argued in our petitions that Takahashi discloses determining whether to power down a portion of a cache

[00:00:24] Speaker 01:
and could be combined with who's comparison of estimated power gain and power loss.

[00:00:29] Speaker 01:
The board acknowledged that in its institution decisions, but then in the final written decisions, the board refused to consider the argument.

[00:00:36] Speaker 01:
That was error, and VLSI has no meaningful response.

[00:00:40] Speaker 01:
The board's decisions on institution make clear that one aspect of our argument was that a skilled artisan could, quote, modify Takahashi to explicitly calculate and compare estimated power loss

[00:00:51] Speaker 01:
and estimated power gain as taught in Hu."

[00:00:54] Speaker 01:
Close quote.

[00:00:55] Speaker 01:
That's reading from 4159 of the appendix.

[00:00:58] Speaker 01:
The same language appears on 5154.

[00:01:00] Speaker 01:
And that's the point.

[00:01:01] Speaker 01:
You can take the determining from Takahashi, and to the extent needed, modify it based on the comparison of loss and gain in Hu.

[00:01:09] Speaker 01:
So for the board to refuse to address an argument on the merits that we made and that the board acknowledged we made is error, as this court has held in cases like Viscor and Shinfu and Bernetics.

[00:01:19] Speaker 01:
VLSI cites no authority to the contrary.

[00:01:22] Speaker 01:
Rather, it's urging affirmance on various alternative grounds that the board didn't reach.

[00:01:27] Speaker 01:
That's not proper in an agency case like this one under Chenery and this court's own case law.

[00:01:32] Speaker 01:
So we would submit that at the very least, the decisions need to be vacated and remanded so the board can consider the merits of the argument in the first instance.

[00:01:41] Speaker 01:
I would say a word, if I may, on the second point of our appeal which has to do with the disclosure of who

[00:01:48] Speaker 01:
And that's the context in which the board did address the determining whether to power down limitation.

[00:01:53] Speaker 01:
It did so only in the context of who and it found that who didn't teach that limitation and that finding we would submit is clearly mistaken.

[00:02:01] Speaker 01:
The board itself.

[00:02:01] Speaker 03:
Can I ask you a question about that because I find this somewhat confusing.

[00:02:07] Speaker 03:
It seems to me clear on the face of it that who does teach determining to power down based on the decay interval.

[00:02:16] Speaker 03:
In that sense, it does satisfy a determining step.

[00:02:20] Speaker 03:
What I thought the board was saying about who is that it doesn't power down based on the relationship between the L2 access leak ratio.

[00:02:34] Speaker 03:
You contend that that affects the decay interval determination and so there is a relationship.

[00:02:40] Speaker 03:
So, it really isn't a question, is it, as to whether

[00:02:44] Speaker 03:
Who teaches the determining step?

[00:02:46] Speaker 03:
The real question is whether it teaches a determining step based on a relationship to the L2 access leak ratio.

[00:02:55] Speaker 01:
I think in fairness to the board, the board did purport at least to decide both issues.

[00:03:01] Speaker 01:
I think that's certainly VLSI's position.

[00:03:04] Speaker 01:
But I'm pleased to take the question as you've asked it, Your Honor, which is let's assume, which I think is plainly correct, that who does describe determining whether to power down.

[00:03:13] Speaker 01:
It does so based on the decay interval.

[00:03:15] Speaker 01:
To answer the question, who does teach that the proper decay interval is chosen based on what it calls the L2 access to leak ratio, and I think that's plainest on pages 1131 and 1132 of the appendix, which is discussing figure nine and its accompanying text.

[00:03:33] Speaker 01:
And if one looks on 1131 on the last full paragraph on that page, who says beginning in the middle of that paragraph,

[00:03:41] Speaker 01:
When the L2 access to leak ratio equals 100, then small decay intervals, less than 8,000 cycles, are detrimental to both performance and power.

[00:03:51] Speaker 01:
This is because short decay intervals may induce extra cash misses by turning off cash lines prematurely.

[00:03:57] Speaker 01:
This effect is particularly bad when L2 access to leak is 100, because high ratios mean that the added energy cost of additional L2 misses is quite high.

[00:04:06] Speaker 03:
And so this is clearly saying that's because... Help me to understand what you think

[00:04:12] Speaker 03:
the commission is saying about simulation because they seem to be saying that there's no relationship between L2 access leak ratio and the decay interval because somehow it's just a simulation.

[00:04:27] Speaker 03:
Is that what you understand what we're saying?

[00:04:30] Speaker 01:
I must confess, Your Honor, I have a hard time understanding what the board makes of the fact that it's a simulation.

[00:04:36] Speaker 01:
I mean, just because who is benchmarking results for four different

[00:04:41] Speaker 01:
L2 access to leak ratio values, that doesn't mean that the values aren't being used to determine whether to power down a cash.

[00:04:47] Speaker 01:
I mean, if one looks at figure nine, one can clearly see what who is doing, which is they've taken the four different curves, which represent the four different ratios, which are at the top of the chart, and they've plotted them to show the energy saved compared to not having a power down policy at all.

[00:05:05] Speaker 01:
based on the various decay intervals.

[00:05:07] Speaker 01:
And it shows that there clearly is a difference as to the optimal decay interval, which is the minimum of the curve, depending on which ratio you choose.

[00:05:14] Speaker 01:
So the entire purpose of this figure, we submit, is to figure out the best decay interval depending on what ratio your particular system has, and who says that the ratio is going to vary depending on how you make the chip, how it's designed.

[00:05:27] Speaker 01:
So it uses these four as benchmarks.

[00:05:29] Speaker 01:
And then based on what your actual chip specs are, you can figure out which decay interval is best for your particular system.

[00:05:35] Speaker 01:
The idea is that this can be used as a model for an actual system based on actual estimates of L2 access to leak.

[00:05:41] Speaker 01:
And in fact, who has a review of reported data in the literature earlier in the reference on 1123 and 1124 is something DLSI admits and put no date of its brief.

[00:05:51] Speaker 01:
So I don't understand really why the board was hung up on the fact that who caused this simulation.

[00:05:56] Speaker 01:
data, it's still recommending a decay interval for almost all the systems of around 8,000, but when the ratio is really high, say 100, then any decay interval shorter than 8,000 will mean that the dissipated energy that occurs due to increased cash misses will quickly use up more power than you would if you didn't power down at all.

[00:06:17] Speaker 01:
Our expert, Dr. Raghunathan, explained this in detail.

[00:06:21] Speaker 01:
So I think this discussion reflects exactly what WHO says it's about in the language that the board quoted on page 820, that a basic principle of the evaluations is to measure the static power saved by turning off portions of the cache, which is the denominator of the LTO access to leak ratio, and then comparing it to the extra dynamic power dissipated by turning off cache lines.

[00:06:43] Speaker 01:
That's the numerator.

[00:06:44] Speaker 01:
And that's what they're plotting in figure 9, and it shows that WHO is determining whether to power down

[00:06:51] Speaker 01:
based on the decay interval, which in turn varies with and is based on the L2 access to leak ratio.

[00:07:00] Speaker 03:
Could you turn to maybe the third issue about whether Takahashi teaches estimated power gain?

[00:07:08] Speaker 01:
Absolutely, Your Honor.

[00:07:09] Speaker 01:
So this comes up in the context of our obviousness argument over Takahashi alone.

[00:07:15] Speaker 01:
The board indicated, I mean, there's a dispute as to whether it actually made a finding, but I think it came very close to saying that Takahashi's use of the cash miss rate would qualify as representative of the estimated power loss.

[00:07:30] Speaker 01:
But it said that by comparing it to a predetermined value, Takahashi was not comparing it to a proxy for estimated power gain.

[00:07:37] Speaker 01:
And we think that's incorrect because Takahashi does exactly what the patent is doing.

[00:07:42] Speaker 01:
Instead of coming up with a separate ratio, like who does,

[00:07:44] Speaker 01:
Takahashi's comparing the current cash miss rate to a predetermined value.

[00:07:48] Speaker 01:
And the fact that Takahashi doesn't call it estimated power gain, that doesn't matter.

[00:07:51] Speaker 01:
The prior art doesn't have to use the same terminology as the patent.

[00:07:55] Speaker 01:
And the 014 patent does exactly the same thing that Takahashi does.

[00:07:59] Speaker 01:
It doesn't measure estimated power, which would be measured in nanojoules per second, or watts, or another scientific measure of power.

[00:08:07] Speaker 01:
Instead, it uses what it calls the amount of valid dirty information.

[00:08:11] Speaker 01:
And then it compares that to a predetermined threshold that it calls TH.

[00:08:15] Speaker 01:
And as the board noted on page 18 of the appendix, the patent doesn't disclose the value of the threshold.

[00:08:21] Speaker 01:
It doesn't even disclose how it's determined.

[00:08:23] Speaker 01:
But it's clear that they're not measures of power.

[00:08:26] Speaker 01:
On page 12 of the red brief, VLSI itself refers to TH as simply a threshold number of dirty bits.

[00:08:35] Speaker 01:
It's not a measurement of power.

[00:08:37] Speaker 01:
And in one embodiment of the patent in column 6,

[00:08:39] Speaker 01:
The patent simply says TH is, quote, representative of the estimated power gain.

[00:08:44] Speaker 01:
Well, if the predetermined threshold amount of dirty information can be a proxy for power gain, as it clearly is in the patent, then Takahashi's predetermined threshold, which is based on a cash miss rate, can too.

[00:08:56] Speaker 01:
Both Takahashi and the patent are simply taking one parameter, in the patent's case dirty information, in Takahashi's case a cash miss rate, and setting a threshold.

[00:09:05] Speaker 01:
where if the actual measured value is higher than the threshold, then powering down would dissipate too much dynamic power so you don't do it.

[00:09:13] Speaker 01:
But if it's below the threshold, then saving the leakage power is worth it.

[00:09:16] Speaker 01:
So you do do it.

[00:09:18] Speaker 01:
So we would say that because Takahashi discloses an indistinguishable implementation from the 014 patent, Takahashi also discloses the claimed estimated power gain and the board clearly erred and was unsupported by substantial evidence in concluding otherwise.

[00:09:33] Speaker 01:
So for that reason and also for the reasons regarding the Takahashi-hu combination that I was discussing earlier, we respectfully submit that the decision should be vacated and the case remanded.

[00:09:43] Speaker 01:
If the court has questions, I'm happy to answer them.

[00:09:45] Speaker 01:
Otherwise, I would appreciate reserving the balance of my time.

[00:09:49] Speaker 00:
Okay.

[00:09:49] Speaker 00:
Any questions for Mr. Fleming at this point?

[00:09:52] Speaker 01:
No.

[00:09:53] Speaker 01:
No.

[00:09:54] Speaker 00:
All right.

[00:09:55] Speaker 00:
Thank you.

[00:09:55] Speaker 00:
Then we'll hear from Mr. Vickili.

[00:10:00] Speaker 02:
Thank you, your honor.

[00:10:00] Speaker 02:
May it please the court, Kevin Zichile on behalf of DLSI Technology.

[00:10:05] Speaker 03:
The 014 pad.

[00:10:09] Speaker 03:
As to this first point about whether Takahashi discloses the determining limitation and the suggestion that that wasn't argued, that's not right, right?

[00:10:19] Speaker 03:
They were incorrect about that.

[00:10:23] Speaker 02:
Well, we certainly disputed that it teaches the determining limitation and we both have disputed whether this was properly raised if that's your honor's question.

[00:10:32] Speaker 03:
The question of whether it teaches it or not is not before us because the board just refused to consider the issue because it said it wasn't raised.

[00:10:40] Speaker 03:
But what I'm suggesting to you is that's clearly wrong, right?

[00:10:43] Speaker 03:
They did raise it in the petition.

[00:10:47] Speaker 02:
So it is our position that what was raised was that Takahashi taught the entire element, the determining element, which is determining whether to power down based on an estimated power gain and an estimated power loss.

[00:11:00] Speaker 02:
That was the first ground applied against the independent claims.

[00:11:05] Speaker 03:
That's 4034.

[00:11:05] Speaker 03:
They argued Takahashi discloses power management circuitry that determines whether to power down at least a portion of the component during a low power mode.

[00:11:15] Speaker 03:
How clear does it have to be?

[00:11:16] Speaker 03:
That's pretty clear.

[00:11:19] Speaker 02:
Your Honor, yes.

[00:11:20] Speaker 02:
In one of the petitions, they separated the determining element, I believe, based on the power management circuitry, because Takahashi describes circuitry as opposed to simulation and who.

[00:11:30] Speaker 02:
But if you look at the substance of the argument, in both cases, in both petitions, petitioner argues that Takahashi teaches the entire determining element, and as an alternative, that who teaches the entire determining element.

[00:11:43] Speaker 02:
There is no suggestion that Takahashi's evaluation of cash miss rate could be popped out and whose L2 access leak ratio popped in.

[00:11:54] Speaker 02:
And that is the new argument raised for the first time in the hearing upon which appellant seeks to rely in this proceeding.

[00:12:00] Speaker 03:
OK.

[00:12:00] Speaker 03:
Well, why don't you go on to the second point, which is what the board is saying about HOOP, which I find very confusing.

[00:12:09] Speaker 03:
I mean, you certainly agree.

[00:12:13] Speaker 03:
that in who the decay interval is determining whether to power down, right?

[00:12:21] Speaker 02:
Yes.

[00:12:21] Speaker 02:
The basis of powering down is a period of time elapsing into which is the decay interval.

[00:12:25] Speaker 02:
That's correct.

[00:12:26] Speaker 02:
OK.

[00:12:27] Speaker 03:
So the real question here, it seems to me that who does teach them determining whether to power down?

[00:12:33] Speaker 03:
The question is whether the L2

[00:12:40] Speaker 03:
ratio is affecting the decay interval calculation, right?

[00:12:46] Speaker 02:
Yes, and in its petitions, Appellant focused on this figure 9 of Who, which appears on page 1132 of the appendix.

[00:12:56] Speaker 02:
And what figure 9 shows is the result of the simulation of a quantity called the normalized leakage energy as a function of two independent variables.

[00:13:05] Speaker 02:
One is the decay interval on the x

[00:13:07] Speaker 03:
I don't understand the commission simulation point.

[00:13:11] Speaker 03:
What does that mean?

[00:13:11] Speaker 03:
I mean, you agree that two in figure nine shows that the decay interval is being determined based on this L2 ratio, right?

[00:13:27] Speaker 02:
No, we don't, Your Honor.

[00:13:28] Speaker 02:
And the reason is because the normalized leakage energy is calculated across a range of decay intervals.

[00:13:34] Speaker 02:
as you see there.

[00:13:35] Speaker 02:
And there are four values of the L2-axis leak ratio that are also applied as parameters.

[00:13:41] Speaker 02:
Who does not say is how to select the decay interval from these curves.

[00:13:46] Speaker 02:
It doesn't say that the minimum should be selected.

[00:13:48] Speaker 02:
It doesn't say that there is a monotonic trend.

[00:13:51] Speaker 02:
It says nothing.

[00:13:52] Speaker 02:
It simply calculates this dependent variable based on two independent variables.

[00:13:58] Speaker 02:
And what Appellant suggested is that there should therefore be a dependence between the two independent variables.

[00:14:02] Speaker 02:
But that is not tied into.

[00:14:04] Speaker 03:
It seems to be a different point, different argument that the commission was addressing.

[00:14:09] Speaker 03:
They said this doesn't show that there's a relationship because it's a simulation.

[00:14:15] Speaker 03:
I really don't understand what that means.

[00:14:19] Speaker 03:
Well, there is a separate point about shows the relationship.

[00:14:23] Speaker 03:
What difference does it make that it's a simulation in terms of what the teaching you get from it?

[00:14:30] Speaker 02:
Well, Your Honor, for one thing, I believe that the Board did acknowledge that WHO does not teach the dependence of the decay interval on L2 access leak.

[00:14:41] Speaker 02:
That was one of their points.

[00:14:43] Speaker 02:
They made a separate point that the L2 access leak ratio takes assumed values.

[00:14:49] Speaker 02:
And this is on page 24 of the final written decision.

[00:14:52] Speaker 02:
They say that WHO uses assumed values as inputs to generate simulation data.

[00:14:56] Speaker 02:
And I think the significance of that is that the claim requires

[00:14:59] Speaker 02:
an estimated power gain and an estimated power loss resulting from powering down as two estimated quantities, whereas WHO merely provides one quantity taking assumed values, 5, 10, 20, and 100, not estimated values.

[00:15:15] Speaker 02:
These are simply made up numbers that the authors chose to see how their simulations would behave.

[00:15:21] Speaker 02:
And so that is a separate basis for a firmness because it fails to show the estimation feature.

[00:15:27] Speaker 03:
But surely it suggests that

[00:15:30] Speaker 03:
These assumed values, as you call them, could be determined in the real world to determine what the actual value is.

[00:15:39] Speaker 02:
Well, it actually teaches against that.

[00:15:41] Speaker 02:
It talks about the difficulty of making estimates of power gain and power loss, which is why they focused on a ratio.

[00:15:49] Speaker 02:
L2 access leak is only a ratio, whereas the claim requires an estimated power gain and a separate estimated power loss.

[00:15:56] Speaker 02:
So they actually describe the difficulty of making the estimates described in the CLANS, which the 014 inventors were the first to divide the methods for.

[00:16:06] Speaker 03:
Well, but that seems to be, again, a different point that the commission was making.

[00:16:10] Speaker 03:
Maybe that's an argument the commission needs to address, but I don't see that that's what the commission decided, that there were difficulty in making the estimates, and therefore it didn't teach making

[00:16:23] Speaker 02:
That's fair, Your Honor.

[00:16:24] Speaker 02:
Yes.

[00:16:25] Speaker 02:
What they did find, however, is that the values upon which the appellant relied in his petitions, and that is, again, the 5, 10, 20, and 100, were not estimated values.

[00:16:34] Speaker 02:
They were just made up values.

[00:16:35] Speaker 02:
The board says, who uses these assumed values of the old practice leak ratio of 5, 10, 20, and 100 as inputs?

[00:16:43] Speaker 02:
to generate simulation data.

[00:16:45] Speaker 02:
Again, not estimated values used to make a determination of whether to power down, but simply made up values used to see how a simulation behaves.

[00:16:58] Speaker 03:
You want to address the third point?

[00:17:01] Speaker 02:
Yes, Your Honor.

[00:17:02] Speaker 02:
So the dispute below with regards to predetermined value centered on whether the predetermined value was nearly a prior measurement of the cache miss rate

[00:17:13] Speaker 02:
or some special, as counsel argues, threshold value representing a proxy for estimated power gain.

[00:17:21] Speaker 02:
And they couch their argument by referring to the 014 patent specification, a very clear example of client site, arguing by analogy that because both are predetermined and both are vague in how they're described, they must therefore be the same.

[00:17:36] Speaker 02:
But Takahashi is very clear that the predetermined value

[00:17:39] Speaker 02:
is really just the last measured value of the cash miss rate, which is compared to the current measured value to determine the change of the cash miss rate.

[00:17:48] Speaker 02:
And that change of the cash miss rate is the foundation of the principle of operation in Takahashi.

[00:17:53] Speaker 02:
What Takahashi does is it powers down a portion of the cash, checks if the cash miss rate increased, and if it did, it powers it back up.

[00:18:01] Speaker 02:
Because it doesn't want to sacrifice performance, and it describes the cash miss rate as sort of a metric of performance.

[00:18:07] Speaker 02:
So this notion that it is some special threshold that is somehow related to an estimated power gain is absent in Takahashi.

[00:18:14] Speaker 02:
It is merely the last measured value of cache miss rate that is compared with the current value to see if there's been a change.

[00:18:26] Speaker 00:
OK, anything else you need to tell us, Mr. Bakili?

[00:18:32] Speaker 02:
One more point I'd like to make about who

[00:18:35] Speaker 02:
There's a number of additional portions of who have been cited describing how a completely new method for calculating, allegedly calculating the decay interval based on the ultra-axis leak ratio or some vague statements about how one might reason about how decay intervals could depend on cost considerations.

[00:18:57] Speaker 02:
We maintain that these are improper being newly cited on appeal.

[00:19:03] Speaker 02:
The before the board was this figure nine.

[00:19:06] Speaker 02:
And even if they were considered, we don't think it would necessarily change the outcome.

[00:19:10] Speaker 02:
But figure nine itself, again, shows two independent variables and the effect on a single dependent variable.

[00:19:18] Speaker 02:
It does not describe how to select particular values of either independent variable based on the other.

[00:19:24] Speaker 02:
And that was before the board.

[00:19:25] Speaker 02:
The board's decision was based on substantial evidence on the record below.

[00:19:32] Speaker 02:
I'm happy to answer any other questions Your Honors may have.

[00:19:35] Speaker 00:
Any more questions for Mr. Akely?

[00:19:37] Speaker 02:
No.

[00:19:38] Speaker 02:
No.

[00:19:39] Speaker 00:
Okay.

[00:19:40] Speaker 00:
Thank you.

[00:19:40] Speaker 00:
Then we'll hear rebuttal from Mr. Fleming.

[00:19:43] Speaker 01:
Thank you, Your Honor.

[00:19:45] Speaker 01:
Only three points in response.

[00:19:46] Speaker 01:
First, on this issue of a new argument in the petition, I direct the court to page 4054 of the appendix.

[00:19:54] Speaker 01:
And at the top of that page, this is our petition, the same language appears on 5049.

[00:19:59] Speaker 01:
We say a person of ordinary skill and the art would have been motivated to specifically combine who's disclosure of estimating power gain and power losses with Takahashi's disclosure of powering down a portion of the cash only when a cash misrate does not exceed a predetermined value threshold.

[00:20:14] Speaker 01:
This wasn't a new argument.

[00:20:15] Speaker 01:
It was in the petition.

[00:20:16] Speaker 01:
The board cited this language specifically in its decisions on institution at 4159 and 5154.

[00:20:23] Speaker 01:
Second point regarding who and the decay interval.

[00:20:26] Speaker 01:
Our point, and we think who discloses this clearly, is that the decay interval varies as the ratio changes.

[00:20:33] Speaker 01:
The text that I read in the colloquy with Judge Stike during the opening presentation on 1131 says that when the L2 access to leak ratio is high, the decay interval should also be higher because if you turn it off too soon,

[00:20:47] Speaker 01:
then there will be additional cash miss rates that use up a lot more power and will quickly get rid of or even exceed the amount of energy that you would say by not even having a policy in place.

[00:20:58] Speaker 01:
Saying that who doesn't teach a relationship between decay interval and the L2 access to leak ratio is just contrary to the plain disclosure.

[00:21:06] Speaker 01:
In fact, if Mr. Vacuoli were right on this, who would not have occasion to say on 1131,

[00:21:13] Speaker 01:
that a decay interval of less than 8,000 cycles works when you have lower ratios but doesn't work when you have a higher ratio like 100.

[00:21:20] Speaker 01:
That is a relationship between the L2 access to leak ratio and the decay interval.

[00:21:25] Speaker 01:
The third point, Mr. Vacuoli said it was that who suggested it was somehow difficult to estimate L2 access or leak figures.

[00:21:34] Speaker 01:
That's not actually right.

[00:21:35] Speaker 01:
What who says,

[00:21:37] Speaker 01:
And this is on 1122, I believe, is, quote, since both leakage and dynamic power values vary heavily with different designs and fabrication processes, it is difficult to nail down specific values for evaluation purposes, close quote.

[00:21:53] Speaker 01:
That's why instead of picking just one system to evaluate, they focus on ratios that they can use as benchmarks.

[00:22:00] Speaker 01:
But who does not suggest that a particular ratio couldn't be determined for a particular system based on actual estimates of LTO access and leak figures.

[00:22:08] Speaker 01:
In fact, who actually sets forth calculations like that from a review of the literature on a 11, appendix 1123 to 24.

[00:22:17] Speaker 01:
And DLSI acknowledges that in the red brief in footnote 8 on page 54.

[00:22:22] Speaker 01:
And those are the basis for the theoretical analysis that WHO does, which the experimental results on 1131 and 1132 later reinforce.

[00:22:30] Speaker 01:
And I would note just as a final point, the claims of the 014 patent do exactly the same thing.

[00:22:35] Speaker 01:
They don't measure power either.

[00:22:38] Speaker 01:
They use proxies.

[00:22:39] Speaker 01:
In the case of the 014 patent, they use the amount of dirty bits compared to a threshold that the board itself on page 18 says is not explained what it is or how it's determined.

[00:22:51] Speaker 01:
So attempting to distinguish the prior art because they supposedly don't measure power is not an effective distinction whatsoever.

[00:22:57] Speaker 01:
The patent is exactly the same.

[00:22:59] Speaker 01:
Unless the court has further questions for me, we would respectfully submit that the final written decisions of the board should be vacated and the matter is remanded.

[00:23:06] Speaker 00:
Any more questions for Mr. Fleming?

[00:23:09] Speaker 01:
No.

[00:23:10] Speaker 00:
No.

[00:23:11] Speaker 00:
OK.

[00:23:12] Speaker 00:
Thank you both.

[00:23:12] Speaker 00:
The case is taken under submission.