In the long-running saga of Intel’s conflicts with various national trade commissions, 2009 was a lousy year for Intel. The European Commission fined Intel for nearly 1.5 billion USD, the US Federal Trade Commission sued Intel on anti-trust grounds, and Intel settled with AMD for another 1.25 billion USD. If nothing else it was an expensive year, and while Intel settling with AMD was a significant milestone for the company it was not the end of their troubles.

Now just shy of 9 months after the FTC’s lawsuit began, Intel’s conflicts are starting to come to an end. While the European Commission’s fine is still on appeal, Intel can close the book on their troubles with the FTC: Intel and the FTC have reached a settlement ahead of what would have been next month’s court hearing. With this settlement the FTC is agreeing to drop the case in return for a series of prohibitions and requirements placed upon Intel to maintain and enhance the competitive environment in the CPU and GPU markets. However true to their word, the FTC did not push for any fines – this is a settlement of actions, and not one of greenbacks.

Background

As a quick refresher, the FTC had been investigating Intel for a number of years. Ahead of their suit in December of 2009 the FTC brought Intel to the table and tried to negotiate a settlement, but that didn’t come to pass. In the meantime Intel and AMD reached a separate truce back in November of 2009, an important distinction as AMD had been the primary instigator of all of the investigations against Intel. Ultimately the FTC decided to take their case forward without further help from AMD, a somewhat surprising move that at first glance seems to have largely panned out in the FTC’s favor.

In the suit, the FTC listed a number of complaints towards Intel over both the CPU and the GPU markets, where in the former Intel is the dominant player, and in the latter their market share has been increasing over the years as IGPs grow in popularity. The FTC’s complaints were roughly as follows:

CPU

  1. The usual complaints we’ve seen from the EU. Intel rewarded OEMs to not use AMD’s processors through various means, such as volume discounts, withholding advertising & R&D money, and threatening OEMs with a low-priority during CPU shortages.
  2. Intel reworked their compiler to put AMD CPUs at a disadvantage. For a time Intel’s compiler would not enable SSE/SSE2 codepaths on non-Intel CPUs, our assumption is that this is the specific complaint. To our knowledge this has been resolved for quite some time now.
  3. Intel paid/coerced software and hardware vendors to not support or to limit their support for AMD CPUs. This includes having vendors label their wares as Intel compatible, but not AMD compatible.
  4. False advertising. This includes hiding the compiler changes from developers, misrepresenting benchmark results (such as BAPCo Sysmark) that changed due to those compiler changes, and general misrepresentation of benchmarks as being “real world” when they are not.


The GeForce 9400M: Intel's chief competitor in the Core 2 integrated graphics market and a threatened product line

GPU

  1. Intel eliminated the future threat of NVIDIA’s chipset business by refusing to license the latest version of the DMI bus (the bus that connects the Northbridge to the Southbridge) and the QPI bus (the bus that connects Nehalem processors to the X58 Northbridge) to NVIDIA, which prevents them from offering a chipset for Nehalem-generation CPUs.
  2. Intel “created several interoperability problems” with discrete CPUs, specifically to attack GPGPU functionality. We’re actually not sure what this means, it may be a complaint based on the fact that Lynnfield only offers single PCIe x16 connection coming from the CPU, which wouldn’t be enough to fully feed two high-end GPUs.
  3. Intel has attempted to harm GPGPU functionality by developing Larrabee. This includes lying about the state of Larrabee hardware and software, and making disparaging remarks about non-Intel development tools.
  4. In bundling CPUs with IGP chipsets, Intel is selling them at below-cost to drive out competition. Given Intel’s margins, we find this one questionable. Below-cost would have to be extremely cheap.
  5. Intel priced Atom CPUs higher if they were not used with an Intel IGP chipset.
  6. All of this has enhanced Intel’s CPU monopoly.

The purpose of the suit and what would have been the associated trial would have been for the FTC to prove that Intel engaged in these actions, and more importantly that these actions were harmful to the market to a significant enough degree to run afoul of Section 5 of the FTC Act.

One other piece of more recent background information involves Dell. Dell has been under investigation by the US Securities & Exchange Commission for the past few years regarding financial irregularities. Those irregularities, as the SEC charges, were due to Intel’s secret rebates to the OEM to not use AMD processors. When Dell began using AMD processors in 2007 and those rebates stopped, Dell ceased to turn a profit.

In turn, since these payments were secret, investors had no idea that the only reason the company was profitable was due to these payments as opposed to entirely above-the-board measures by the company. The end result was that the SEC sued Dell, and last month Dell settled the case by paying a $100 million fine to the SEC. As is the case with similar settlements, the case is not legally binding proof of Intel’s actions, as Dell neither had to confirm nor deny the SEC’s charges (in essence allowing them to claim that the settlement is just the easiest way to get the SEC out of the way). It does however show us just how large Intel’s rebates to OEMs may have been.

The Settlement
POST A COMMENT

63 Comments

View All Comments

  • Wurmer - Monday, August 09, 2010 - link

    ''I wonder what would have happened if Intel had not (allegedly) paid off the OEMs to shut AMD out of the business?''

    I was wondering the same thing a while ago. I think that in the end it wouldn't have changed the big picture but AMD might have end up in a better financial situation this allowing them to put more $$ in R&D. They had the better product for about like 2 years during the Athlon 64 era, that's a hell of a long time in the computer business. Anyways, you can speculate all we want we'll never know.
    Reply
  • cesthree - Saturday, August 07, 2010 - link

    I like my limited vocabulary, fucking fuckers.

    Intel has done nothing wrong. Evil rebate mongering corporations exist in every fucking facet of business; it's not Intel's fault, they're just playing the game and winning!

    Fuck AMD, but first fuck Nvidia for ever venturing past GPU's. Intel needs to keep them AWAY from DMI and QPI. They totally FUCKED up 939, AM2, and 775 with their piece of shit NF4-7 bullshit, especially 775.

    If a motherfucking software patch can enable SLI on ANY motherboard that has the electrical capability, because it is totally possible, then Nvidia can go get fucked and never engineer another motherboard, PLEASE.

    AMD, well if they don't like being outcompeted by Intel, then stop making CPU's and chipsets and just stick to the ATI bullshit. Maybe if they weren't fucking around with bullshit CPU's and chipsets they could get a Catalyst release that didn't suck-off 100 donkeys simultaneously.

    Fuck.

    That was a last fuck for good measure you fucking fuckeddy fucks.

    Don't tell me you don't like the cursing or that it offends you; you know you like it Shit Fuck.
    Reply
  • rscoot - Sunday, August 08, 2010 - link

    Despite your obvious troll, perhaps you are merely too young to remember when nVidia offered the best AMD chipsets on the market during a time when AMD did not make chipsets of their own. Their onboard audio solution back in 2002 is still better than 99% of the onboard audio solutions provided by intel/AMD right now, it's too bad they had to drop it on the nForce 3's because of cost issues.

    And maybe it's just me, but I remember my nForce 4 SLI board working quite well with my manchester 939..
    Reply
  • cesthree - Sunday, August 08, 2010 - link

    Ok, I was a bit drunk when I posted that last bit; but you've still helped me prove my point.

    Yes, I had two NF4 boards. One DFI LP UT NF4 Ultra-D and one Asus A8N32SLI-Deluxe. The DFI is my current HTPC. Couldn't agree with you more. The only reason I bundled NF4 in is for Nvidia's garbage software patches preventing SLI for running on any setup; the NF4 was guilty of that too.

    Do you see it now?

    One stupidly obvious point is that neither of those boards had QPI or DMI on them. Well, not exactly.

    Still not catching it?

    The IMC and pretty much everything else that needs to be stable was already designed and packaged on the physical CPU. (no, not 754 and before, but those NF2-3 boards were good for what, on-board sound is what you're telling me? Maybe they were good OC'ers, but AMD didn't shine until 939 and then only for a minute.)

    Let me tie it together for you now.

    That's the same as, you guessed it, the i3/i5/i7 CPU's that Intel has today.

    The NF4 series worked fine, because Nvidia couldn't screw up what was designed by AMD. Mostly I am talking about the communication of the major parts in an x86 design: CPU, RAM, IMC.

    No wonder Intel wanted Nvidia not designing parts to do those jobs. The Nvidia 775 IMC controllers were trash aside from hand selected boards given to review sites.

    Maybe it was just the EVGA 775's I had; I've never like their BIOS releases.

    If Intel is shady so is Nvidia, and so is AMD. The only reason anybody is looking at Intel like they are, is because they are on top. If AMD or Nvidia were on top, they would be the one's answering to the FTC.

    Nothing any of us can do about it. Besides, AMD should be glad they have lasted this long after getting their start on Intel's coattails.
    Reply
  • ClagMaster - Monday, August 09, 2010 - link

    Silly Troll Scumbag. Being drunk does not excuse you for the worthless profane remarks about nVidia chipsets.

    Nvidia chipsets were really meant for AMD parts. Nvidia chipsets for Intel parts were an afterthought that I believed did not recieve the same level of attention and testing as those for

    I myself have used nForce 2 400 chipset with a nicely overclockable AMD XPM-2500.

    I have also used an nForce 4 chipset with an Optron 175. Built in 2006, it still runs fine today.

    I have been satisfied with the performance of both chipsets. What I was not satisfied with is the poor construction of motherboards associated with AMD chipsets. Since 2004, this has markedly improved.
    Reply
  • synaesthetic - Monday, August 09, 2010 - link

    Is there a full moon or something tonight?

    The trolls are really coming out of the woodwork... comments on AT are usually very civil.
    Reply
  • matt b - Thursday, August 12, 2010 - link

    Ryan, I read that Anandtech is still looking into the performance of Intel compiler with non-Intel chips.
    Here is Agner Fog's updated blog where he describes how Intel compiler treats AMD chips, and Intel dishonesty to him about it in 12/30/09. http://www.agner.org/optimize/blog/read.php?i=49#4...
    Intel's newly released beta June 2010 of its Math Kernel Library v.10.3 is still crippling AMD processors.http://www.agner.org/optimize/blog/read.php?i=49#1...
    Agner says that he is working on a list of software and benchmarks that are crippled on AMD processors due to the Intel Compiler or libraries.
    From a year ago at ArsTechnica, PCMark05 memory performance increased 47% by changing CPUID from VIA to Intel.
    http://arstechnica.com/hardware/reviews/2008/07/at...
    Reply
  • matt b - Thursday, August 12, 2010 - link

    Note they allege that Cinebench is compromised by using Intel compilers. Until Cinebench responds, I think that Anandtech should quit using it.

    64. Several benchmarking organizations adopted benchmarks that measured performance of CPUs running software programs compiled using the Intel compiler or libraries. Intel’s deception affected among others, the Business Applications Performance Corporation (“BAPCo”), Cinebench, and TPC benchmarks.

    67. Intel publicized the results of the benchmarking to promote sales of products containing its
    x86 CPUs even though it knew the benchmarks were misleading. For example: ...
    Reply
  • saifikhan - Sunday, August 15, 2010 - link

    Buy AMD based laptops and notebooks/netbooks ! Let the dealer know that you're interested in AMD. Unfortuntately in India, one can still see HP selling underpriced Intel laptops with overpriced AMD laptops. eg here. Compaq Presario range http://is.gd/eiUXo , Essentials http://is.gd/eiV0n and EliteBook http://is.gd/eiV20

    The same sad story repeats with Vostro, Acer, Sony, Benq laptop models in India.

    AMD is offering a large number of Notebook processors, please see the list here http://products.amd.com/en-gb/NotebookCPUResult.as...

    thanks
    Saifi.
    India.
    Reply
  • davidri - Friday, September 03, 2010 - link

    "In reality most of the things Intel is being barred from doing are things that they discontinued doing long ago if they did them at all (note that this settlement is not an admission of guilt on Intel’s part)."

    Uh, ok... then why was Intel sued by the FTC and fined by the European Commision and going to be under such scrutiny and regulation by the FTC for years to come??!!

    There is probably a whole crap ton of other despicable and undermining business practices Intel could be held accountable for that no one but top Intel execs know about.

    Big B.S.
    Reply

Log in

Don't have an account? Sign up now