The Tesla-AMD-Global Foundries Deal: What We Know And Don't Know (Yet)

News
By Fritz Nelson, Paul Alcorn
published


Some media outlets are reporting that GlobalFoundries is working with Tesla on AI technology for its cars. This erroneous report stems from a comment GloFlo's CEO Sanjay Jha made on stage on Wednesday at the fab's annual get-together in San Jose. Here's what CNBC reported:

"On Wednesday Sanjay Jha, CEO of AMD spin-off GlobalFoundries, said at the company's technology conference in Santa Clara, California, that the company is working directly with Tesla. GlobalFoundries, which fabricates chips, has a wafer supply agreement in place with AMD through 2020." CNBC in an article published on 9/20/2017

But what Jha actually said—which we can confirm because we were present to hear it firsthand—was that GlobalFoundries is trying to attract companies as business models change:

"As we develop these new technologies, we are also seeing a big shift in the business model and the foundry business. What is happening is that system companies like Google, like Amazon, like Tesla, like Microsoft, are coming directly to foundries. They are working directly with IP companies and system development companies because they want to control the hardware and software."

Global Foundries is not saying that it's working with Tesla--but that's not to say that AMD isn't working with Tesla. Jim Keller, formerly the chief architect for AMD's microprocessors, is now VP of autopilot hardware at Tesla.

Last year, AMD lost what Tesla CEO Elon Musk called a tight race against Nvidia for the auto company's GPU/AI business. Since that time, AMD has continued to show strength across multiple sectors.

The CNBC report said that its sources tied AMD and Tesla together, but neither AMD or Tesla will comment on the situation. The report indicated that Tesla was on a mission to develop its own chip for autonomous cars in order to be more vertically integrated, but that Tesla was potentially relying on building that "on top of AMD intellectual property." That particular wording certainly paints a dotted line to GlobalFoundries.

Further, during a presentation at the Deutsche Bank Technology Conference last week, AMD CFO Devinder Kumar responded to a question from AMD's U.S. semiconductor research chief Ross Seymore about downward pressure on the company's semi-custom business due to the cyclical nature of the gaming console business by saying that although revenue will be down year over year, there are "design wins" coming. He stated:

"The design wins you talked about is coming in 2018. That is a non-game console product and that revenue should be happening in the 2018 timeframe to continue to help the revenue on the semi-custom which sits within the EESC segment."

News of these developments, despite the tentative veracity of the Tesla-AMD linkage, sent AMD's stock price up. We'll keep pressing on the matter, among others.

Fritz Nelson
Fritz Nelson
Fritz Nelson is Editor-at-Large of Tom's Hardware US.
See more CPUs News
9 Comments Comment from the forums
  • jasonelmore
    Pretty dumb not to go with Nvidia on this.. I understand them wanting to vertically integrate but Nvidia is 5 years ahead of everyone else in compute and machine learning hardware.
    Reply
  • AlistairAB
    That's a typical exaggeration, they have a year's head start that's all. And google and apple and huawei all have dedicated inferencing chips already being made.
    Reply
  • Mousemonkey
    20195128 said:
    That's a typical exaggeration, they have a year's head start that's all. And google and apple and huawei all have dedicated inferencing chips already being made.

    Only a year? So how comes there is a video on it from two years ago?
    Reply
  • Fluffy_Hedgehog
    NVidia is not 5 years ahead. Google has been using their own designs for years now, they just did it quietly and only now opened up their hardware to third party users. Their performance for AI is at least equal to the NVidia offers.

    Microsoft is using custom designs too but they are even more secretive about it so it's next to impossible to get solid information on what exactly is going on there and how they perform.

    If you just look at compute (16 and more importantly 8 bit for ai) performance even amd is entering with specialized chips derived from their gpu devision and while their gpu's are less than stellar for gaming, their chip-design is more than a match to NVidia in raw compute, which is where it counts for AI.

    NVidia was the one on the marketing drum the most, pushing products to the open market, but they are not ahead technology wise in any significant way compared to the other players, they are just louder (as usual).
    Reply
  • Mousemonkey
    20195585 said:
    NVidia is not 5 years ahead. Google has been using their own designs for years now, they just did it quietly and only now opened up their hardware to third party users. Their performance for AI is at least equal to the NVidia offers.

    Microsoft is using custom designs too but they are even more secretive about it so it's next to impossible to get solid information on what exactly is going on there and how they perform.

    If you just look at compute (16 and more importantly 8 bit for ai) performance even amd is entering with specialized chips derived from their gpu devision and while their gpu's are less than stellar for gaming, their chip-design is more than a match to NVidia in raw compute, which is where it counts for AI.

    NVidia was the one on the marketing drum the most, pushing products to the open market, but they are not ahead technology wise in any significant way compared to the other players, they are just louder (as usual).

    The company ahead of the curve, however, is Nvidia, which is primarily known for producing chips behind video games. It has shifted its attention to cars, starting in 2007 when its systems powered the integration of Google Earth inside an Audi. Nvidia said its automotive unit posted 85 percent annual growth in sales for the last fiscal year. Several luxury carmakers use its supercomputers, which can take the reams of data coming from up to 12 cameras on a car, plus any Lidar and ultrasonic sensors, and make sense of it.

    https://www.recode.net/2015/10/27/11620026/meet-the-companies-building-self-driving-cars-for-google-and-tesla

    Seems that they may be more than five years ahead.
    Reply
  • xrodney
    The thing has Nvidia had headstart for last 5 years, but just in last 12 months, many companies started to catching up very quickly.

    If you are a pioneer in some technology areas it takes time to develop, however, once new technology/business area is there, others can simply enter it, and if their area of expertise is in similar technology all of your headstart is down to almost nothing.
    Reply
  • mzarate
    Paul,

    The remaining semi-custom design win has been in the making since 2014. I provided sources for this here:

    Given that time frame, do you still think this could be related to to Tesla? 4 years feels like a long time, but perhaps not out of the question since it'd be a new market & product for AMD.

    https://www.reddit.com/r/AMD_Stock/comments/6pjzo9/amd_vaults_10_ceo_su_says_expect_another/dkqk9tn/
    Reply
  • grimfox
    Curious. Is GLOFLO the correct abbreviation? Seems like GLOFO would make more sense. Or is it just a weird anarchism of abbreviations?

    I could see why AMD is reticent to clarify their position with Tesla because it's probably given them a little boost in the market they wouldn't want to squash. Or perhaps they aren't talking because they are under NDA. I seem to recall them being cagey prior to the last console announcements where they had won the custom design.
    Reply
  • ryan.j.matheson
    Loved this article. All the financial news articles ate this news, spun it around, and the stock went nuts. Seeing the actual quote from the conference because Toms was actually there is way different then all the other reports and sheds light on the "real" situation. Nice reporting work Toms.
    Reply