FTC Announce Settlement With Intel

[god_43]

The Federal Trade Commission approved a settlement with Intel Corp. that resolves charges the company illegally stifled competition in the market for computer chips. Intel has agreed to provisions that will open the door to renewed competition and prevent Intel from suppressing competition in the future.

The settlement goes beyond the terms applied to Intel in previous actions against the company and will help restore competition that was lost as a result of Intel’s alleged past anticompetitive tactics. At the same time, the settlement will leave the company room to innovate and offer competitive pricing.

http://ftc.gov/opa/2010/08/intel.shtm


and here is the link to the agreement,

http://ftc.gov/os/adjpro/d9341/100804inteldo.pdf

[570091D]

a webcast for a settlement announcement? somehow, i doubt the world is holding it's breath....

[god_43]

a webcast for a settlement announcement? somehow, i doubt the world is holding it's breath....

lol maybe it should...ftc just opened the door for nvidia x86!

[rogueagent6]

lol maybe it should...ftc just opened the door for nvidia x86!

Maybe I'm missing something, but no where in that settlement do I see the FTC requiring Intel to license x86 to nvidia. The only reference to nvidia I saw was regarding chipsets.

[god_43]

Maybe I'm missing something, but no where in that settlement do I see the FTC requiring Intel to license x86 to nvidia. The only reference to nvidia I saw was regarding chipsets.

ah my young friend, you skipped over the part about x86 companies allowed to change control, and their licenses do not become void. they are also allowed to form joint ventures and mergers as well.

intel must create a new license with VIA.....who owns a 16-20% share in VIA (ill give you a hint...it starts with N)? still, i wish the arrangement went further.

[rogueagent6]

ah my young friend, you skipped over the part about x86 companies allowed to change control, and their licenses do not become void. they are also allowed to form joint ventures and mergers as well.

B. In the event the Designated Intel Competitor undergoes a “change of control” (as defined in the relevant Designated Intel Competitor Patent Agreement) that is publicly announced or a Designated Intel Competitor otherwise notifies Respondent that it has undergone a change of control within five (5) days of such change of control:

1. for a period of thirty (30) days from the date of the change of control, Respondent shall not initiate patent litigation against the party acquiring the Designated Intel Competitor (“Acquiring Entity”) with respect to products previously manufactured by or acquired from the Designated Intel Competitor, unless the Designated Intel Competitor and/or the Acquiring Entity or another entity controlled by one of them has first filed any suit against Respondent;

2. within ten (10) days from the date of the change of control, Respondent shall offer to enter into a written, reciprocal Standstill Agreement with the Acquiring Entity, such Standstill Agreement to comprise the following terms:

a. Respondent and the Acquiring Entity shall enter into good faith negotiations regarding the future patent relationship, if any, between them;

b. to facilitate those good faith negotiations, for a period of one year from the change of control, neither Respondent nor the Acquiring Entity (or any Affiliate of either of them) shall initiate patent litigation against the other or any Affiliate thereof;

c. the Standstill Agreement shall not act as a license or provide any patent or other intellectual property rights or defenses to any person or party, either expressly or by implication, estoppel, exhaustion, license, waiver, laches or otherwise; and

d. Respondent shall afford the Acquiring Entity not less than ten (10) days from receipt of Respondent’s written offer to accept in writing the offered Standstill Agreement.

e. For purposes of this Section, “Affiliate” means any entity that is directly or indirectly controlled by, under common control with, or that controls the subject entity.

I saw that part, but that is just a Standstill Agreement, it's not forcing Intel to offer the new company a license, only that they have negotiate a new agreement... the negotiations could fail knowing the arrogance of these companies.

That's how I read it at least.

[Russian]

Imo, Intel got off easily. This just opens up the door for competition if anything.

[Eastcoasthandle]

I saw that part, but that is just a Standstill Agreement, it's not forcing Intel to offer the new company a license, only that they have negotiate a new agreement... the negotiations could fail knowing the arrogance of these companies.

That's how I read it at least.

Yeah I see what you mean, good point.

[Humminn55]

lol maybe it should...ftc just opened the door for nvidia x86!

If you thought Intel was having a difficult time developing Larrabee, even with IP from Nvidia, cannot imagine how costly and intensive it'd be for Nvidia to produce a truly competitive x86 cpu despite having access to Via's IP (if the VIA buyout by Nvidia is what's being referred to), which is licensed from Intel.

Trying to do that might just spell the doom of Nvidia.

[Manicdan]

define competitive,
they can start small with a hybrid chip built for HTPCs where its under 100W and able to do anything. i wouldnt expect them to have something competing in top end or perf/watt battles. but it could easily work for 80% of desktop needs and get them some market share

[Eastcoasthandle]

So there is a 6 year requirement for Intel's cpus/chipset to support the PCI bus. This is the only real thing I see for nv more so then AMD. But I honestly don't think it will take 6 years before we see Fusion and Intel's version (whatever they call it) to hit the street in full force.

[Humminn55]

define competitive,
they can start small with a hybrid chip built for HTPCs where its under 100W and able to do anything. i wouldnt expect them to have something competing in top end or perf/watt battles. but it could easily work for 80% of desktop needs and get them some market share

Even if Nvidia bought out VIA, got a licensing agreement signed with Intel, and started really researching a hybrid cpu/gpu, do you think they'd have anything remotely ready to sell within the next 4-5 years? It's not like Nvidia has any experience with cpus, and VIA's are nothing like what's being designed and built today by AMD and Intel.

So, any competitive arena. To try to do so would simply bankrupt Nvidia. Despite their cash reserves, it'd take more than Nvidia can come up with to scratch design a cpu/gpu hybrid chip.....you can just about throw VIA's crap out the window.

[god_43]

If you thought Intel was having a difficult time developing Larrabee, even with IP from Nvidia, cannot imagine how costly and intensive it'd be for Nvidia to produce a truly competitive x86 cpu despite having access to Via's IP (if the VIA buyout by Nvidia is what's being referred to), which is licensed from Intel.

Trying to do that might just spell the doom of Nvidia.

yes i am referring to a nvidia/via merger, i am fully aware of how much the whole thing might cost them. 4-5bln to buy via, and another 4-5bln to develop something. you are assuming that nvidia has no experience with cpus. they hired transmeta employees a few years back, they have good engineers, they have a lot of experience with chipsets. they probably wont be able to compete in the high end for at least 4-5 years, but they can compete just fine in the mainstream market.

I saw that part, but that is just a Standstill Agreement, it's not forcing Intel to offer the new company a license, only that they have negotiate a new agreement... the negotiations could fail knowing the arrogance of these companies.

That's how I read it at least.

yes if they were taken over; i.e hostile take over is usually what a standstill agreement refers to. i believe a merger would not warrant this? but either way, via is now able to use gpus from other vendors.

Respondent shall, within thirty (30) days after the date this Order becomes final,
offer to Via to sign written amendments to the Via Patent Agreements to:

1. extend the “Capture Period” in Sections 1.4 of the Litigation Settlement
Agreement between Via Technologies, Inc. and Intel Corporation dated
April 7, 2003, 1.3 of the Via Patent Cross License Addendum of the same
date, and 1.1 of the Via Microprocessor Addendum of the same date, to
provide that the Capture Periods end on the fifteenth yearly anniversary of
the Effective Date of those agreements;

2. confirm that under the Via Patent Agreements, Via is permitted to make,
use, sell or import Via Microprocessors that are compatible with the x86
instruction set but not pin compatible or bus compatible with Intel
Microprocessors, including such Via Microprocessors with graphics
technology designed by and supplied to Via by a third party, so long as
Via does not exceed the scope of the licenses expressly granted under or
otherwise breach the terms of those Agreements; and

3. provide that Respondent shall, upon the request of Via, publicly state that
Via is permitted to make, use, sell or import Via microprocessors that are
compatible with the x86 instruction set but not pin compatible or bus
compatible with Intel microprocessors, including such Via
microprocessors with graphics technology designed by and supplied to
Via by a third party, so long as Via does not exceed the scope of the
licenses expressly granted under or otherwise breach the terms of those
Agreements.

Imo, Intel got off easily. This just opens up the door for competition if anything.

i agree, and i do not believe this goes far enough to even foster competition; much more so than already.

[rogueagent6]

yes if they were taken over; i.e hostile take over is usually what a standstill agreement refers to. i believe a merger would not warrant this? but either way, via is now able to use gpus from other vendors.

I think in that case, the standstill would only protect via from the party looking to take them over, i.e. hostile takeover by nvidia.

I think the FTC is referring to this:

Another type of standstill agreement is an agreement whereby two or more parties agree not to deal with other parties in a particular matter for a period of time. For example, in negotiations for a merger or acquisition, the target and the purchaser may enter into an agreement where they each agree not to solicit or embark on acquisitions from or of other parties. This allows the parties to invest more heavily into the negotiation, due diligence, and details of a potential acquisition.

Standstill Agreement

Yeah, it's wiki, but it seems to fit what the FTC is talking about.

[mstp2009]

Intel got away EASY on this one. REALLY EASY.


I don't forsee much competition coming from this simply b/c it costs so much in R&D to develop a good x86 CPU. After NVIDIA's recent market share price crash (look over the last 2 years), they no longer have the funds to really push development on x86 AND GPUs. They have to pick one, if you ask me.

[god_43]

I think in that case, the standstill would only protect via from the party looking to take them over, i.e. hostile takeover by nvidia.

I think the FTC is referring to this:



Standstill Agreement

Yeah, it's wiki, but it seems to fit what the FTC is talking about.

hmm i think you are right.....wow wtf, ftc couldn't even make it easy for nvidia to obtain a license...this settlement is a bunch of fail! damn i had high hopes, that the market would finally be corrected....ftc chairmen are probably sporting new expensive cars at intels expense...

[rogueagent6]

hmm i think you are right.....wow wtf, ftc couldn't even make it easy for nvidia to obtain a license...this settlement is a bunch of fail! damn i had high hopes, that the market would finally be corrected....ftc chairmen are probably sporting new expensive cars at intels expense...

There is a flip side to that though, if someone were to buy out via, and intel didn't extend the x86 license to the purchasing party for any reason, it would look bad for them. They may have the cash, but I think they are walking on eggshells at the moment and know it. This is all speculation, but it will be interesting to see how it plays out. I'm with you though, I'd like to see nvidia buy out the rest of via and start producing an x86 chip, if only to force AMD and Intel to continue to innovate.

[god_43]

There is a flip side to that though, if someone were to buy out via, and intel didn't extend the x86 license to the purchasing party for any reason, it would look bad for them. They may have the cash, but I think they are walking on eggshells at the moment and know it. This is all speculation, but it will be interesting to see how it plays out. I'm with you though, I'd like to see nvidia buy out the rest of via and start producing an x86 chip, if only to force AMD and Intel to continue to innovate.

honestly...i know we make fun of the guy but; i think nvidia ceo is the only one with balls right now because nvidia is not going to settle (at least that's what they say now). nvidia is going head first to court with Intel.

GRAAAH HUANG SMASH! EAT INTEL FOR BREAKFAST! lol i didn't know he had a tattoo....

[STEvil]

He looks both worried and constipated at the same time...

[Face]

Some more on the Via talk:

Intel: Will FTC Deal Trigger Acquisition Of Taiwan-Based Via?

One of the basic facts of the PC microprocessor market is that there simply isn’t much real competition; there’s Intel (INTC), Advanced Micro Devices (AMD), and not much else. Often overlook in the discussion is a small Taiwan-based chip company with a license to make x86 chips called Via Technologies. But Via could turn out to be a more important player in the wake of Intel’s settlement with the Federal Trade Commission this week of antitrust charges.

Bernstein Research analyst Stacey Rasgon notes in a research report that the FTC in its analysis of the proposed consent order in the case has “explicitly called out that they are leaving the door open for a new competitor to enter the x86 space, if one were to genuinely desire to do , through an acquisition of Via.” The FTC states that the modification of the change-of-control provisions of Intel’s license with Via, as required by the proposed agreement, “opens the door to a potential joint venture or acquisition of Via and its x86 license by a strong and well-financed entrant to the x86 market.”

Via would be a way to jump stat any potential new Intel rival; the company has a market cap of just $850 million. Of course, as Rasgon points out, to truly be competitive with Intel “would require significant innovation and investment beyond simply possessing a license.”

http://blogs.barrons.com/techtraderd...d=yahoobarrons

[rogueagent6]

Some more on the Via talk:

the FTC in its analysis of the proposed consent order in the case has “explicitly called out that they are leaving the door open for a new competitor to enter the x86 space, if one were to genuinely desire to do , through an acquisition of Via.”

http://blogs.barrons.com/techtraderd...d=yahoobarrons

So it was their intent to leave that door open, it's going to be interesting to see how it all plays out.

[informal]

Agner Fog's take on the settlement:

Out of court settlement with FTC
Yesterday, the Federal Trade Commission (FTC) announced that they are going for an out of court settlement with Intel. See their press release and proposed decision.

I will comment only on the part of the settlement that deals with Intel's compilers and function libraries. The FTC orders that Intel must inform its software customers about the CPU dispatch mechanism that leads to suboptimal performance on non-Intel CPUs. It also recognizes that certain published benchmarks were misleading. The advantage of an out-of-court settlement is that it is faster. A court battle could take so long time that the issues were obsolete before a decision was made. In a comment, the FTC explains that the purpose of the order is not punitive but remedial.

The settlement with FTC is less far-reaching than the settlement with AMD. The AMD settlement requires that Intel remove any "Artificial Performance Impairment", while the FTC settlement requires only that Intel inform their customers of what they do. This will not solve the problem, only make it more visible. The wording of the settlement is also somewhat ambiguous as to which clauses apply to both the compiler and the function libraries, and which clauses apply to the compiler only. This is unfortunate since many software developers are using the Intel function libraries without using the Intel compiler.

The FTC have asked me to testify in court about the CPU dispatching in Intel's compilers and function libraries. I will not have to do this now, of course, but I will continue to publish my findings here on my blog. I am currently doing a survey of software that is affected by the biased CPU dispatching and I am going to publish the results here soon.

Since Intel have not removed the biased CPU dispatching from their MKL library despite the settlement with AMD, and since the settlement with FTC does not require them to do so, we can expect that the problem will persist.

It is interesting that the FTC in their comment suggests that software developers can override the code dispatch mechanisms in Intel compilers and libraries. This is a technique that I have developed and described in my C++ manual. However, I doubt that commercial software developers will be happy to use such hacking techniques that rely on undocumented features.

The response of the software community will probably be to avoid Intel software products entirely. In my test of the optimizing performance of C++ compilers, the Intel compiler and the Gnu compiler for Linux shared the first place. Unfortunately, the Gnu compiler for Windows is not up to date so we still need a good replacement for the Intel compiler for Windows. It is not a profitable business to make a well optimized math function library. If we cannot use Intel's libraries then we probably have to rely on the open source community for making such libraries. The Gnu function libraries (glibc) are not very well optimized, so there is still a lot of work to do. The work of optimizing the Gnu function libraries is going very slowly and is done mainly by an Intel guy. Why don't AMD and independent programmers contribute to this work to make sure the software performs well on non-Intel processors as well?

After all, the FTC settlement leaves the software community with more problems than we could expect after the AMD settlement. Maybe this reflects the limited power of the FTC?
Reply To This Message

In his previous blog post he writes about the fact that intel's Math Kernel Library (v. 10.3) has the CPU dispatch bias(still uses an inferior code path for non-Intel processors).He is doing a survey at the moment,using an application for changing CPUID that he developed for Via's x86 Nano chip(that has this capablity).He said he will publish the results of the survey soon.

New version is still crippling Intel's competitors
Author: Agner Fog Date: 2010-06-29 04:16
Intel have released a new version of their Math Kernel Library (v. 10.3) in beta test.

I have tested the new libraries and found that the CPU dispatching works basically the same way as before. The standard math library, vector math library, short vector math library and the 64-bit version of other math kernel library functions still use an inferior code path for non-Intel processors.

I have found the following differences from previous versions:

* Many functions now have a branch for the forthcoming AVX instruction set, but still only for Intel processors. This will increase the difference in performance between Intel and AMD processors on these functions. Both Intel and AMD are planning to support AVX in 2011.

* The CPU dispatcher for the vector math library has a new branch for non-Intel processors with SSE2. Unlike the generic branch, the new non-Intel SSE2 branch is used only on non-Intel processors, and it is inferior in many cases to the branch used by Intel processors with the same instruction set. The non-Intel SSE2 branch is implemented in the 32-bit Windows version and the 32-bit Linux version, but not in the 64-bit versions of the library.

* A new Summary Statistics library uses the same CPU dispatcher as the vector math library.

Obviously, I haven't tested all functions in the library. There may be more differences that I haven't discovered. But it is clear that many functions in the new version of the library still cripples performance on non-Intel processors. I don't understand how they can do this without violating the legal settlement with AMD.

[Manicdan]

so the settlement with AMD should force intel to remove the inferior code for non intel cpus, so why would the FTC have them do anything different? does one settlement override the other?

[informal]

Nope,it's just that FTC doesn't have the power that court has(or the contract AMD and intel made in their settlement).However,if intel does the dirty stuff again, FTC can pursue the thing to court as they wanted this time too. And if AMD wishes to,they can use the biased CPU dispatching in latest intel software as a poof of settlement breach and go to court again.

[terrace215]

so the settlement with AMD should force intel to remove the inferior code for non intel cpus

No.

That settlement basically says Intel cannot do:

If (AMD_processor) waste_time;

But it is free to do:

If (Intel_processor & special_instruction_set) go_faster;

And it is under no obligation to spend its own time or money implementing and then ensuring that

If (AMD_processor & special_instruction_set) go_faster;

would actually work in all cases, never have a bug, etc.

http://download.intel.com/pressroom/..._agreement.pdf

See section 2.3.

---------

And the FTC agreement basically says this behavior is fine, as long as you agree to notify compiler customers that you take no responsibility for optimizing for AMD products even on compatible instruction sets.

In other words, if you don't like it, AMD, Agner, etc., go spend your own time & money developing, testing, and marketing your OWN compiler for any & all x86 processors.