[00:00:01] Speaker 03:
Our next case for argument is 24-2296, Exifer versus Microsoft.

[00:00:07] Speaker 03:
Ms.

[00:00:08] Speaker 03:
Bernard, please proceed.

[00:00:09] Speaker 03:
Thank you, Your Honor.

[00:00:19] Speaker 02:
May it please the court.

[00:00:21] Speaker 02:
This appeal turns on two legal errors at the 702 stage in the district court that wiped out Exifers damages expert opinion in its entirety.

[00:00:31] Speaker 03:
If we agree with you that Mr. Block's testimony should have been allowed, is that the only issue we have to reach in this case?

[00:00:38] Speaker 03:
Yes, that's correct, Your Honor.

[00:00:39] Speaker 02:
If you agree that... Why are you focused on that?

[00:00:42] Speaker 02:
Okay.

[00:00:42] Speaker 02:
So the first legal error that the district court made in excluding Mr. Block's testimony

[00:00:48] Speaker 02:
Excuse me, Your Honor, if I understand, do you want me to focus on the exclusion of Mr. Block slowly?

[00:00:54] Speaker 02:
Thank you.

[00:00:55] Speaker 02:
So the first legal error the district court made in including Mr. Block's testimony is that it applied a categorical rule that revenues from non-accused products can never be included in the royalty base for a reasonable royalty analysis.

[00:01:09] Speaker 02:
Exafer cements that is not consistent with this court's precedent.

[00:01:13] Speaker 02:
In Section 284's command, the damages be adequate to compensate for the infringement.

[00:01:20] Speaker 02:
So from this court's precedent, the foundational principle in a damages analysis

[00:01:25] Speaker 02:
is determining what was taken, the value of the use of the patented invention.

[00:01:32] Speaker 02:
In this case, Microsoft's use of the patented invention was its use of the VFP FastPath and SmartNIC technology.

[00:01:39] Speaker 01:
I want to ask you a question.

[00:01:40] Speaker 01:
I mean, we have some cases like NPLAS that have what sounds like, I suppose, a categorical rule that you cannot base damages on non-infringing products.

[00:01:53] Speaker 01:
And then we have maybe some other cases where there might be, you know, there's different alternative ways to measure damages, like a cost savings approach or something like this.

[00:02:02] Speaker 01:
What would you say the rule is for when someone, a damages expert, can reasonably rely on or base the estimated reasonable royalty on a non-infringing product, sales of a non-infringing product?

[00:02:21] Speaker 02:
The rule that we would advance here is something similar to what this rule set forth in the Brumfield case, where it rejected a categorical rule that foreign conduct could not, that reasonable royalty damages could not be based on foreign conduct.

[00:02:36] Speaker 00:
So here, what we are trying to do is it could, as long as there was an adequate causal connection between the infringement and the gain.

[00:02:48] Speaker 02:
That's correct.

[00:02:49] Speaker 02:
What Bromfield said was the foundational principles, this value of what was taken.

[00:02:54] Speaker 02:
were trying to value the use of the patented invention.

[00:02:58] Speaker 02:
And when you're looking at hypothetical negotiation, you're trying to value how the parties will look at it at that time.

[00:03:03] Speaker 02:
And part of that analysis under Brumfield was taking a look at the relationship between the non-infringing activities and the infringing activities.

[00:03:10] Speaker 02:
And in Brumfield, this court found that when there's a causal relationship, and it gave a great example of what that means that is applicable here.

[00:03:19] Speaker 02:
when that infringing activity enables otherwise unavailable profits from, in that case, the foreign conduct.

[00:03:28] Speaker 01:
So the expert would have to have factual information to reasonably rely on in order to satisfy doubt work, right?

[00:03:37] Speaker 01:
Correct.

[00:03:38] Speaker 01:
So there would be... And there also would have to be a correct methodology, right?

[00:03:41] Speaker 02:
I'm sorry.

[00:03:42] Speaker 01:
A correct methodology.

[00:03:43] Speaker 01:
They would have to have... I think you're talking about methodology, really, maybe.

[00:03:48] Speaker 01:
And I think about Doward, I think about that the expert has to be able to rely on, there has to be some basis factually to support what the expert is saying.

[00:03:57] Speaker 01:
And also they have to have a proper methodology, right?

[00:04:01] Speaker 02:
I agree with you 100%.

[00:04:02] Speaker 02:
In a rule 702, there's methodology and there's sufficient facts of data.

[00:04:06] Speaker 02:
So he would have to have any expert, damage expert, to meet this threshold that we're talking about in Brubfield, would have to have evidence.

[00:04:14] Speaker 02:
Evidence of an economic nexus, this causal relationship between the non-infringing activities and infringement.

[00:04:19] Speaker 03:
And so Mr. Locke was trying to testify, if I understand it right.

[00:04:22] Speaker 03:
that the VFP CPUs, which are the patented and claimed device, resulted in a substantial, I won't give the number because I know that it's confidential, but a substantial increase in VM hour production.

[00:04:36] Speaker 03:
Because basically, the patentee invented something that made stuff work a lot faster.

[00:04:42] Speaker 03:
And so he's figured, how do you value that?

[00:04:44] Speaker 03:
It's not a standalone product, doesn't sell, so he valued it by looking at how many hours were ultimately produced, right?

[00:04:52] Speaker 02:
That is 100% correct, as he put it in his, the but four hours.

[00:04:56] Speaker 02:
So you have the VFP fast path, and you have the SmartNet technology.

[00:05:00] Speaker 02:
And when Microsoft, at the time of the hypothetical negotiation, was implementing this technology into its Azure cloud system.

[00:05:06] Speaker 03:
And it was Microsoft's own documents that established the improvement

[00:05:10] Speaker 03:
that this technology would provide in VM hours, correct?

[00:05:14] Speaker 02:
That is a hundred percent correct, Your Honor.

[00:05:15] Speaker 02:
We've laid that out in the brief, and that was not a... Same time for rebuttal?

[00:05:20] Speaker 02:
Okay.

[00:05:20] Speaker 02:
Thank you, Your Honor.

[00:05:22] Speaker 02:
Oh, wait.

[00:05:22] Speaker 02:
Could I ask you a quick house use instruction?

[00:05:24] Speaker 01:
I'm sorry.

[00:05:25] Speaker 01:
A lot of things in Mr. Block's report are designated as confidential.

[00:05:30] Speaker 01:
Some of them probably are still confidential, but others seem like, why are these being indicated as confidential?

[00:05:36] Speaker 01:
For example, in page A1226, it appears that

[00:05:40] Speaker 01:
a factor from Georgia Pacific is designated as confidential.

[00:05:45] Speaker 01:
Has there been any review of these kinds of things and what's confidential and what's not to try to narrow it down so the court can do its job?

[00:05:52] Speaker 02:
There was, at some point in this briefing, that the court came back and asked Microsoft to designate.

[00:05:58] Speaker 02:
Because Mr. Block's report, obviously, is mostly Microsoft's confidential information.

[00:06:01] Speaker 02:
So we had them take a look at it.

[00:06:03] Speaker 02:
I believe in the appendix, most of it was still marked confidential as it was in the underlying district court litigation.

[00:06:09] Speaker 01:
I mean, it's not what I'm talking about, says in Georgia Pacific, the court held that the determination of reasonable royalties.

[00:06:13] Speaker 01:
Maybe I shouldn't read this.

[00:06:14] Speaker 01:
But it goes on, and it says what the Georgia-Pacific court held.

[00:06:18] Speaker 01:
And so I'm wondering why that's designated as confidential.

[00:06:21] Speaker 01:
So go ahead.

[00:06:21] Speaker 01:
I agree with you.

[00:06:22] Speaker 02:
That should not be confidential.

[00:06:23] Speaker 02:
And we're happy to work with Microsoft and try to get some of that cleaned up for the court so you know, for your opinion, what is.

[00:06:30] Speaker 00:
Don't we still have a rule, maybe even in FRAP itself, that requires the parties to get together to review overconfidentially?

[00:06:41] Speaker 00:
over marking of material as confidential?

[00:06:44] Speaker 00:
And we did.

[00:06:44] Speaker 02:
And that did happen earlier on in this case.

[00:06:47] Speaker 00:
Might be worth a second shot with respect to the actual report.

[00:06:53] Speaker 02:
Certainly, Your Honor.

[00:06:53] Speaker 02:
We will definitely work with Microsoft to take a look more through the appendix and make sure that we designate some of that as a report.

[00:07:02] Speaker 02:
We can have that information for its opinion.

[00:07:03] Speaker 02:
Thank you very much.

[00:07:06] Speaker 03:
Ms.

[00:07:06] Speaker 03:
Punsackert?

[00:07:10] Speaker 04:
Thank you, Your Honor.

[00:07:11] Speaker 04:
May it please the court.

[00:07:11] Speaker 04:
Kelly Hunsaker on behalf of Microsoft.

[00:07:14] Speaker 04:
Exifer lost this case because it made a series of deliberate choices in a singular pursuit of an ultimately invalid damages theory.

[00:07:23] Speaker 04:
In doing so, it abandoned alternative approaches and left the record with no evidence, no methodology, and no roadmap from which a jury could determine a non-speculative, non-zero royalty.

[00:07:40] Speaker 04:
with its only damages theory excluded.

[00:07:42] Speaker 00:
Why don't you just stick at what?

[00:07:45] Speaker 00:
Supporting the first part.

[00:07:46] Speaker 00:
If we disagree with the first part, we don't need to get to the second part.

[00:07:52] Speaker 04:
With respect to the summary?

[00:07:53] Speaker 00:
With what you were about to say, that without anything else, then the judge can end the case.

[00:07:57] Speaker 00:
Forget about that for now, and just talk about why you think the exclusion of his testimony was proper.

[00:08:06] Speaker 04:
Yes, Your Honor.

[00:08:08] Speaker 04:
With respect to categorical rules versus alternate ways of calculating damages, the idea that unpatented products may not be part of a royalty base in a running royalty, a usage-based royalty, goes all the way back to Garrison versus Clark.

[00:08:27] Speaker 00:
So these are method claims, right?

[00:08:30] Speaker 04:
There are method claims and there are system claims.

[00:08:34] Speaker 00:
So what would be an unpatented product in a method claim?

[00:08:39] Speaker 04:
Virtual networks are a separate product.

[00:08:41] Speaker 04:
Azure has 200 different products and services.

[00:08:44] Speaker 04:
Virtual networks are a separate product.

[00:08:47] Speaker 04:
VFP lives entirely in virtual networks.

[00:08:51] Speaker 04:
The source code for VFP is entirely in virtual networks.

[00:08:54] Speaker 00:
When you say virtual networks, you're talking about a name with a capital V and a capital N, not kind of common language words.

[00:09:01] Speaker 04:
It's a category of products.

[00:09:02] Speaker 04:
There's actually four different types of virtual networks.

[00:09:05] Speaker 04:
But it's a category of products.

[00:09:07] Speaker 04:
It's a different line of business.

[00:09:09] Speaker 04:
than the compute group, which is where virtual machines live.

[00:09:12] Speaker 00:
Just to get back to the affirmative, my understanding is Mr. Block said that as a result of the practicing of this method, Microsoft was able to sell many more hours.

[00:09:29] Speaker 00:
And he's trying to just figure.

[00:09:32] Speaker 00:
And then he made an estimate of the ink current

[00:09:37] Speaker 00:
of the additional hours of the thing that actually Microsoft gets money for and then allocate, figured out what the profit level was on that and then allocated that profit between the two parties.

[00:09:53] Speaker 00:
And then kind of sort of at the end said, I'll just translate that into a base and a per unit.

[00:10:02] Speaker 00:
But why is that not reasonable and economically sound?

[00:10:07] Speaker 04:
Because all that Mr. Congdon did, and all that Mr. Block did in his damages series, would show a functional relationship between a computer and a network.

[00:10:17] Speaker 04:
If this isn't the virtual world, we're talking about a computer as the virtual machine, and a whole network as the virtual networks.

[00:10:26] Speaker 03:
And so, well, Mr. Cronin's testimony was not struck.

[00:10:31] Speaker 03:
And his testimony was that the VFP FastPath provided X, because I'm not going to say the number, improvement in VFP CPUs, which is directly correlated to the increase in VM hours.

[00:10:45] Speaker 03:
So if you have a method of manufacturing and it results in you producing, I'll just say a random number, 20% more product in an hour,

[00:10:54] Speaker 03:
Why can't the measure of damages be focused on the additional products that were produced?

[00:11:00] Speaker 04:
Because, Your Honor, unless the entire market value rule applies, unless the patented feature drives customer demand for the whole system that you're drawing the circle around.

[00:11:12] Speaker 03:
Why?

[00:11:12] Speaker 03:
If the Microsoft-owned documents expressly said how many additional VM hours it thought it would achieve by utilizing this exact product.

[00:11:23] Speaker 04:
It actually did not, Your Honor.

[00:11:25] Speaker 04:
What it said was it anticipated a 50% increase.

[00:11:30] Speaker 03:
That's the number you're not supposed to say.

[00:11:32] Speaker 03:
That's the confidential number.

[00:11:33] Speaker 03:
That's why we keep saying X. I apologize.

[00:11:36] Speaker 01:
It's your number, but like.

[00:11:38] Speaker 01:
Do you want to answer the housekeeping question and tell us that nothing in here is confidential?

[00:11:44] Speaker 04:
The vast majority of it is we can de-designate.

[00:11:48] Speaker 04:
redacted very lightly.

[00:11:50] Speaker 01:
It makes it very hard for us to have a conversation with you when you're able to say it out loud, because you know what's confidential and what's not.

[00:11:56] Speaker 01:
Yet we can't.

[00:11:58] Speaker 01:
So could you tell us what is confidential and what's not?

[00:12:01] Speaker 04:
You can assume it's not confidential for purposes of this argument.

[00:12:06] Speaker 01:
OK.

[00:12:07] Speaker 01:
So does that mean nothing's confidential for purposes of this argument?

[00:12:12] Speaker 04:
Your Honor, a lot of this dates back 15 years.

[00:12:18] Speaker 01:
When we went back and did redactions with a conclusive clear answer would be really helpful.

[00:12:25] Speaker 04:
Yes, I believe you can speak to anything in the record.

[00:12:27] Speaker 04:
OK.

[00:12:27] Speaker 03:
All right, well, OK, so let's get back to the merits of the point then.

[00:12:32] Speaker 03:
Microsoft documents suggested a potential fit, or at least Doctor Congdon testified that there would be a 50% improvement in VFP CPUs, which can directly correlate to an increase in the number of VM hours on the Azure platform.

[00:12:48] Speaker 04:
And completely conclusory.

[00:12:49] Speaker 04:
There is no substance.

[00:12:51] Speaker 04:
There is no evidence.

[00:12:52] Speaker 04:
There is no basis for that correlation.

[00:12:55] Speaker 03:
That's not the issue.

[00:12:56] Speaker 03:
The issue here, that expert was not excluded, and you have not challenged on appeal the failure to exclude that testimony.

[00:13:01] Speaker 03:
So that is a factual statement that you have not challenged on appeal.

[00:13:06] Speaker 03:
So why is it wrong for Mr. Block to accept that factual statement, which is not challenged on appeal, as true, and then base his methodology the way he did?

[00:13:15] Speaker 04:
Because these are different products.

[00:13:18] Speaker 04:
One of those products is a network.

[00:13:20] Speaker 04:
One of those products is a computer.

[00:13:22] Speaker 04:
And under the rule of apportionment.

[00:13:24] Speaker 01:
You don't sell the computer.

[00:13:26] Speaker 01:
Or you don't sell one of those things.

[00:13:27] Speaker 01:
I mean, how else would they value their invention?

[00:13:31] Speaker 01:
You're just complaining about the methodology, right?

[00:13:33] Speaker 01:
I mean, you have to agree that there is a factual premise here on the record.

[00:13:39] Speaker 01:
You're just saying even if there is a factual premise for valuing the invention this way,

[00:13:45] Speaker 01:
that it's per se improper to do so?

[00:13:50] Speaker 04:
I believe it is per se improper to use an unpatented, unaccused product as a royalty base in a running royalty.

[00:13:59] Speaker 04:
What the case law says is that you can consider it.

[00:14:02] Speaker 04:
You can consider it in the rate.

[00:14:04] Speaker 04:
You can consider it in the Georgia Pacific factors.

[00:14:07] Speaker 04:
But what you can't do is create a tax, a running royalty, on an unpatented product.

[00:14:15] Speaker 00:
So I guess my recollection, and just correct me, is that Dr. Block kind of ended up phrasing his dollar amount as of the close of the damages period that was subject to discovery, which is 2020.

[00:14:41] Speaker 00:
He did a calculation on overall quantities

[00:14:45] Speaker 00:
and then translated that.

[00:14:47] Speaker 00:
I think their term was expressed it as say a running royalty, but it was a fixed dollar amount.

[00:14:55] Speaker 00:
So why are you don't have a running royalty in front of us?

[00:14:59] Speaker 04:
We do.

[00:15:00] Speaker 04:
His royalty was a two and a half cent

[00:15:05] Speaker 04:
per virtual machine hour running royalty.

[00:15:09] Speaker 00:
Because he just took the total numbers and did a division at the end after justifying the analysis in terms of the total numbers.

[00:15:19] Speaker 04:
He took the total virtual machine revenues.

[00:15:22] Speaker 04:
He divided them by the average selling price of virtual machine hours to arrive at a number of virtual machines

[00:15:29] Speaker 04:
then he assumed that 50% to 100% of those virtual machine hours were attributable solely to the patent in suit.

[00:15:39] Speaker 04:
That was how he arrived at the message, the number.

[00:15:42] Speaker 04:
And that is based on a running royalty for a product that is not accused of infringement.

[00:15:48] Speaker 04:
In every case, you have to separate the patented and unpatented components.

[00:15:52] Speaker 04:
And he did not do that.

[00:15:53] Speaker 04:
Virtual machines don't infringe by using another component.

[00:15:57] Speaker 03:
How do you do that for method claims?

[00:16:00] Speaker 03:
What is the product in a method claim?

[00:16:03] Speaker 04:
Virtual networks.

[00:16:04] Speaker 04:
It's sold by Microsoft.

[00:16:06] Speaker 04:
The only thing that's not sold is the VFP component of virtual networks.

[00:16:11] Speaker 03:
Forget about this case.

[00:16:11] Speaker 03:
A method of manufacturing.

[00:16:14] Speaker 03:
Are you saying that it would be improper?

[00:16:15] Speaker 03:
It's a regular method of manufacturing.

[00:16:17] Speaker 03:
It would be improper to use the product produced by that method in order to assess a royalty.

[00:16:23] Speaker 04:
It's not a product produced by the method.

[00:16:26] Speaker 03:
But I'm asking you as a matter of law.

[00:16:29] Speaker 03:
In a simple case, method of manufacturing results in a single product.

[00:16:33] Speaker 03:
Do you think a damages expert could not use that product for its royalty analysis?

[00:16:41] Speaker 04:
It's not a method of manufacturing.

[00:16:43] Speaker 04:
So the end was.

[00:16:45] Speaker 03:
You're fighting a hypothetical.

[00:16:47] Speaker 03:
Please don't fight a hypothetical.

[00:16:48] Speaker 03:
OK.

[00:16:51] Speaker 03:
May I ask you to clarify the question, please?

[00:16:55] Speaker 03:
How do you?

[00:16:56] Speaker 03:
I'm trying to get to.

[00:16:58] Speaker 03:
What I'm trying to point out to you is there are many instances in which the measure of damages, the royalty base, is on something that is not necessarily the claim thing, because in a method of manufacture, you can't do that.

[00:17:12] Speaker 03:
In a method of manufacture, you have to put the royalty base on the product produced by the method, which may not itself be patented, the product.

[00:17:22] Speaker 03:
So there are instances

[00:17:25] Speaker 03:
when an expert can look to for a royalty base, a product that is the product of a method, and then use the number of products produced by that method as the royalty base.

[00:17:38] Speaker 03:
And I'm kind of wondering, do you think there's something wrong with that theoretically?

[00:17:42] Speaker 03:
Because you want a per se rule, but I don't see how a per se rule applies when there are method claims.

[00:17:52] Speaker 04:
A per se rule applies to apply the rule of apportionment.

[00:17:56] Speaker 04:
You have to tie the award of damages to the infringement.

[00:18:01] Speaker 04:
What infringes is the claims.

[00:18:04] Speaker 04:
Those claims reside in a different product that is sold by the Networking Services Division.

[00:18:10] Speaker 04:
It's not the virtual machines.

[00:18:12] Speaker 04:
It's not the virtual machines that include it.

[00:18:15] Speaker 04:
If you ask for the source code of virtual machines, you will find VFP nowhere.

[00:18:19] Speaker 04:
If you ask for the source code for virtual networks,

[00:18:22] Speaker 04:
That's where it lives.

[00:18:23] Speaker 04:
That product is sold.

[00:18:25] Speaker 04:
It's sold in four different implementations.

[00:18:27] Speaker 04:
Dr. Congdon examined all of those.

[00:18:30] Speaker 04:
And instead of relying on the smallest saleable patent practicing unit, which this court's cases say is required, they didn't look to networking services revenue because it was too small.

[00:18:45] Speaker 01:
Can I ask you a question?

[00:18:47] Speaker 01:
I mean, OK.

[00:18:47] Speaker 01:
So I see you were.

[00:18:51] Speaker 01:
briefing as really emphasizing end class and other cases and advocating for a per se rule that there can never be a royalty award provided based on a non-infringing product, even when it's a method claim.

[00:19:08] Speaker 01:
That's what I see you saying.

[00:19:09] Speaker 01:
I don't see a lot of emphasis on apportionment like that argument you're making right now.

[00:19:15] Speaker 01:
I'm not sure if I agree with you there wasn't apportionment here anyway.

[00:19:20] Speaker 01:
That's a separate issue.

[00:19:21] Speaker 01:
But tell me where you're talking about apportionment in your brief.

[00:19:27] Speaker 04:
In the sense of apportioning a single product between patented and unpatented components, we are not suggesting apportionment in that respect.

[00:19:39] Speaker 04:
With respect to the concept of

[00:19:43] Speaker 04:
not using unpatented products as the royalty base, doctrinally, we believe that that comes from 140 years of case law.

[00:19:56] Speaker 04:
And we believe that that's inherent in section 284.

[00:20:01] Speaker 01:
Now, can I go back to Judge Moore's, Chief Judge Moore's hypothetical?

[00:20:05] Speaker 01:
I don't know if I heard you answer that, but you are aware of cases that say where awards damages have been awarded on method claims when those method claims are based on the product that is produced, right?

[00:20:18] Speaker 04:
If the method claims are being infringed by a user, that user is using the virtual network.

[00:20:25] Speaker 04:
That user isn't using the virtual machine.

[00:20:28] Speaker 01:
So the user... I'm not going to answer the question.

[00:20:31] Speaker 01:
My question was, do you agree that there's cases, or hypothetically, somebody who, a party that manufactures, uses a method to manufacture product, that there could be a royalty based on the sales of the product?

[00:20:49] Speaker 04:
So Your Honor, I do understand the question.

[00:20:53] Speaker 04:
Exifer did not come forward with any cases in that hypothetical.

[00:21:01] Speaker 04:
Demonstrate that so you just don't really answer the question.

[00:21:06] Speaker 04:
I don't know the answer to the question.

[00:21:07] Speaker 04:
It's not a method of manufacture conceptually You're telling me that doctrinally there's a move on okay, okay?

[00:21:24] Speaker 04:
Would if there's no further questions regarding the exclusion order I'd like to turn to the summary judgment

[00:21:32] Speaker 04:
With Exeter's only damages theory excluded, the question is, what was left?

[00:21:43] Speaker 04:
And was there enough evidence in the summary judgment record for a jury to determine a reasonable royalty award without speculation and guesswork?

[00:21:55] Speaker 04:
And in the summary judgment record, there was no way

[00:22:01] Speaker 04:
There was no way for any jury to arrive at a damages number.

[00:22:06] Speaker 04:
There was no structure to a royalty.

[00:22:09] Speaker 04:
There was no methodology, no base, no rate.

[00:22:14] Speaker 04:
Absolutely no way for a jury to formulate a reasonable royalty based on the evidence.

[00:22:22] Speaker 04:
And under a plain application of Rule 56, we think that granting summary judgment

[00:22:28] Speaker 03:
was the only choice that the district court had based on the court's case law.

[00:22:41] Speaker 02:
I sort of doesn't have anything more to add on rebuttal, unless you have additional questions.

[00:22:47] Speaker 00:
You just heard your friend on the other side, or at least I think I heard, say that the claimed method is not used in the manufacture of what Microsoft sells, VM time, if that, I don't

[00:23:05] Speaker 00:
That may not be the correct description, but the thing that it sells, you know, sort of timeshare on their servers for something.

[00:23:13] Speaker 00:
Is that right?

[00:23:15] Speaker 00:
Wrong?

[00:23:15] Speaker 02:
What?

[00:23:16] Speaker 02:
The use of the patented invention by Microsoft is the VFP FastPath and SmartNet technology.

[00:23:23] Speaker 02:
And what that technology does is it adds efficiencies to the networking.

[00:23:27] Speaker 02:
So it's how it processes the data that's going through the network.

[00:23:31] Speaker 02:
And those efficiencies are cost savings in the number of CPUs that are used for networking in a server.

[00:23:39] Speaker 02:
And so that's the direct correlation between the cost savings and then the value of those cost savings per Microsoft's own contemporaneous business documents in 2014.

[00:23:49] Speaker 02:
That value was measured by the additional VMs that you could add per server, leading to VM hours used by the customers and then leading to revenue.

[00:23:58] Speaker 00:
So it's a slightly more

[00:24:01] Speaker 00:
complicated causal connection than saying, I have a method for making ketchup.

[00:24:09] Speaker 00:
Put the tomatoes in first and then the sugar and boy, it goes a lot faster and we can now make lots more ketchup.

[00:24:16] Speaker 02:
Perhaps, but we submit it is really direct economic nexus and causal relationship between

[00:24:22] Speaker 02:
The use of the patented invention, which is what we're trying to get damages for here, nothing more.

[00:24:27] Speaker 02:
And the extra VMs that are put on a server, and the resultant extra VM hours from the customers.

[00:24:36] Speaker 02:
All right.

[00:24:36] Speaker 02:
Thank you, counsel.

[00:24:37] Speaker 02:
This case is taken under submission.