We just got off the phone with Nick Knupffer of Intel, who confirmed something that has long been speculated upon: the fate of Larrabee. As of today, the first Larrabee chip’s retail release has been canceled. This means that Intel will not be releasing a Larrabee video card or a Larrabee HPC/GPGPU compute part.

The Larrabee project itself has not been canceled however, and Intel is still hard at work developing their first entirely in-house discrete GPU. The first Larrabee chip (which for lack of an official name, we’re going to be calling Larrabee Prime) will be used for the R&D of future Larrabee chips in the form of development kits for internal and external use.

The big question of course is “why?” Officially, the reason why Larrabee Prime was scrubbed was that both the hardware and the software were behind schedule. Intel has left the finer details up to speculation in true Intel fashion, but it has been widely rumored in the last few months that Larrabee Prime has not been performing as well as Intel had been expecting it to, which is consistent with the chip being behind schedule.

Bear in mind that Larrabee Prime’s launch was originally scheduled to be in the 2009-2010 timeframe, so Intel has already missed the first year of their launch window. Even with TSMC’s 40nm problems, Intel would have been launching after NVIDIA’s Fermi and AMD’s Cypress, if not after Cypress’ 2010 successor too. If the chip was underperforming, then the time element would only make things worse for Intel, as they would be setting up Larrabee Prime against successively more powerful products from NVIDIA and AMD.

The software side leaves us a bit more curious, as Intel normally has a strong track record here. Their x86 compiler technology is second to none, and as Larrabee Prime is x86 based, this would have left them in a good starting position for software development. What we’re left wondering is whether the software setback was for overall HPC/GPGPU use, or if it was for graphics. Certainly the harder part of Larrabee Prime’s software development would be the need to write graphics drivers from scratch that were capable of harnessing the chip as a video card, taking in to consideration the need to support older APIs such as DX9 that make implicit assumptions about the layout of the hardware. Could it be that Intel couldn’t get Larrabee Prime working as a video card? That’s going to be a big question that’s going to hang over Intel’s heads right up to the day that they finally launch a Larrabee video card.

Ultimately when we took our first look at Larrabee Prime’s architecture, there were 3 things that we believed could go wrong: manufacturing/yield problems, performance problems, and driver problems. Based on what Intel has said, we can’t write off any of those scenarios. Larrabee Prime is certainly suffering from something that can be classified as driver problems, and it may very well be suffering from both manufacturing and performance problems too.

To Intel’s credit, even if Larrabee Prime will never see the light of day as a retail product, it has been turning in some impressive numbers at trade shows. At SC09 last month, Intel demonstrated Larrabee Prime running the SGEMM HPC benchmark at 1 TeraFLOP, a notable accomplishment as the actual performance of any GPU is usually a fraction of its theoretical performance. 1TF is close to the theoretical performance of NVIDIA’s GT200 and AMD’s RV770 chips, so Larrabee was no slouch. But then again its competition would not be GT220 and RV770, it’s Fermi and Cypress.

Next, this brings us to the future of Larrabee. Larrabee Prime may be canceled, but the Larrabee project is not. As Intel puts it, Larrabee is a “complex multi-year project” and development will be continuing. Intel still wants a piece of the HPC/GPGPU pie (least NVIDIA and AMD get it all to themselves) and they still want in to the video card space given the collision between those markets. For Intel, their plans have just been delayed.


The Larrabee architecture lives on

For the immediate future, as we mentioned earlier Larrabee Prime is still going to be used by Intel for R&D purposes, as a software development platform. This is a very good use of the hardware (however troubled it may be) as it allows Intel to bootstrap the software side of Larrabee so that developers can get started programming for real hardware while Intel works on the next iteration of Larrabee. Much like how NVIDIA and AMD sample their video cards months ahead of time to game developers, we expect that Larrabee Prime SDKs would be limited to Intel’s closest software partners, so don’t expect to see much if anything leak about Larrabee Prime once chips start leaving Intel’s hands, or to see extensive software development initially. Widespread Larrabee software development will still not start until Intel ships the next iteration of Larrabee, if this is the case.

We should know more about the Larrabee situation next year, as Intel is already planning on an announcement at some point in 2010. Our best guess is that Intel will announce the next Larrabee chip at that time, with a product release in 2011 or 2012. Much of this will depend on what the hardware problem was and what process node Intel wants to use. If Intel just needs the ability to pack more cores on to a Larrabee chip then 2011 is a reasonable target, otherwise if there’s a more fundamental issue then 2012 is more likely. This lines up with the process nodes for those years: if they go for 2011 they hit the 2nd year of their 32nm process, otherwise if they launched in 2012 they would be able to launch it as one of the first products on the 22nm process.

For that matter, Since the Larrabee project was not killed, it’s a safe assumption that any future Larrabee chips are going to be based on the same architectural design. The vibe from Intel is that the problem is Larrabee Prime and not the Larrabee architecture itself. The idea of an x86 many-cores GPU is still alive and well.


On-Chip GMA-based GPUs: Still On Schedule For 2010

Finally, there’s the matter of Intel’s competition. For AMD and NVIDIA, this is just about the best possible announcement they could hope for. On the video card front it means they won’t be facing any new competitors through 2010 and most of 2011. That doesn’t mean that Intel isn’t going to be a challenge for them – Intel is still launching Carkdale and Arrandale with on-chip GPUs next year – but they won’t be facing competition at the high-end too. For NVIDIA in particular, this means that Fermi has a clear shot at the HPC/GPGPU space without competition from Intel, which is exactly the kind of break NVIDIA needed since Fermi is running late.

POST A COMMENT

71 Comments

View All Comments

  • jconan - Sunday, December 06, 2009 - link

    but Intel owns REAL3D and they still haven't improved on the graphics Reply
  • StevoLincolnite - Friday, December 04, 2009 - link

    I'm aware of that, but it was still badged as an Intel Graphics part and thus there first discreet Graphics card. - Glad you made the correction to the article however. Reply
  • - Friday, December 04, 2009 - link

    Larabee is a huge undertaking in x86 speak -to balance speed, low wattage, and tuned software was simply too much to do- if possible at all. Now Intel will have to leave it to the experts...does this mean Intel will be buying IP licenses from AMD? Or does this eye-opening occurrence bring the industry closer together on things like standards?
    As I have said before, AMD has the blueprints and ideas for their next-gen products, and they would be dumb not to share an idea or two if it were to help set de facto standards for next-gen HPC/GPGPU’s ;that spirit could possibly be evident in the (light) 1.5 billion they agreed on. We'll see what happens
    asH
    Reply
  • AnandThenMan - Friday, December 04, 2009 - link

    Intel already has access to at least some of ATI's patents due to the new agreement between the two companies. But having access to patents is hardly a gateway into making a great GPU, you still have to make build the thing, and more importantly you have to support it on the software side. Both Nvidia and ATI have been doing drivers for so long, it's a well oiled machine. Jumping into the fray from scratch is incredibly challenging.

    As for Intel making a discreet part before, the i740 was too slow when it came out, a big reason it never gained a foothold. Sadly, some of the tech still lives on in Intel's GMA, the worst thing ever to happen to graphics.
    Reply
  • mutarasector - Saturday, December 12, 2009 - link

    "Intel already has access to at least some of ATI's patents due to the new agreement between the two companies"

    It's my understanding that the cross licensing agreements between AMD/ATI and Intel do not cover each other's respective GPGPU technologies, only the x86 side of things and extensions.
    Reply
  • dagamer34 - Friday, December 04, 2009 - link

    Why would cross patent licensing between AMD and Intel include any ATI intellectual property? Disputes between the two companies have always been about CPUs, not GPUs. Reply
  • AnandThenMan - Saturday, December 05, 2009 - link

    Because it does? According to Dirk Meyer's comments, the new agreement as part of Intel's cash payout to AMD includes renewed cross licensing that includes ATI GPU tech.

    What we don't know is the actual scope of what is included, but at least some is.
    Reply
  • qcmadness - Friday, December 04, 2009 - link

    The crappiest is that T&L support started from GMA X3000 (G965 variants).

    Virtually software shaders from virtually all GMA IGPs.
    Reply
  • StevoLincolnite - Friday, December 04, 2009 - link

    No, the crappiest was the S3 Savage 2000, Intels TnL implementation actually works, S3's wouldn't function, and if you could get it to function you would get massive amounts of graphic anomalies.

    However S3's IGP's are far superior to Intels these days as they are based on the S3 Chrome chip.

    Intels main issue is the drivers, they plainly suck, they should take a page out of nVidia and ATI's driver development work and implement a similar strategy.

    The Intel x3xxx series had massively varied performance, Direct X 9 performs poorly, Direct X 10 even though the X3100 supports it, will never run with acceptable image quality and thus performance. (If the game ever decides to work), and Direct X 7 and 8 based games, even if they are over 15 years old still perform poorly.

    Hence the term "Intel Decelerators". - Personally I would prefer a GMA 950 over the x3000/x3100/x3500 simply because it's faster, and can be overclocked with GMA Booster, despite having a lacking feature set.

    Until Intel get there act together concerning there IGP's and drivers for them, I will never take them seriously for a discreet part. - As -every- single piece of Graphics hardware to come from that company has simply sucked.
    Reply
  • rs1 - Friday, December 04, 2009 - link

    I've always been fairly skeptical of Intel's ability to successfully jump in to the GPU market. There's a reason why AMD purchased ATI instead of trying to roll its own GPU's, and there's a reason why nvidia doesn't make CPU's. A world of difference exists between the two, and just because a company has a solid CPU engineering pipeline, it doesn't mean that the same resources can be leveraged to make a solid GPU.

    In any case, I think this is a big win for AMD's Fusion architecture. With the failure of Larrabee it now seems very likely that they will be the first/only company to be offering an integrated CPU/GPU combo, or at least the first company to be offering a *decent* one. Purchasing ATI has already proven to be a good move for AMD, and the demise of Larrabee makes it an even better one.
    Reply

Log in

Don't have an account? Sign up now