AMD's Strix Halo - Under the Hood | Interview #7

Поделиться
HTML-код
  • Опубликовано: 14 янв 2025
  • Hello you fine Internet folks
    At CES 2025 I got the chance to sit down with Mahesh Subramony, AMD Senior Fellow, to talk about AMD's upcoming Strix Halo SoC which is a brand new type of product for AMD and is the big iGPU SoC that many of us have been waiting for from AMD for a long while.
    Hope y'all enjoy!
    Substack: chipsandcheese...
    Wordpress: old.chipsandch...
    Twitter: x.com/Chipsand...
    Bluesky: bsky.app/profi...
    Mastodon: techhub.social...
    Patreon: / chipsandcheese
    PayPal: www.paypal.com...

Комментарии • 60

  • @joelypolly
    @joelypolly 8 часов назад +4

    Great interview with a ton of technical details from AMD. Learned a lot more compared to just reading the press releases.

  • @kelownatechkid
    @kelownatechkid 3 часа назад +1

    Fantastic interview! Love to hear from technical folks

  • @cem_kaya
    @cem_kaya День назад +25

    - better chip interconnect
    - full 512 bit fpu avx
    - 32 MB infinity cache just for the gpu ( gpu writes put data here, configurable ??? )
    not enough focus on the 256 bit memory interface

    • @bobo-cc1xw
      @bobo-cc1xw 13 часов назад +2

      He basically said zen 5 ccd but using a different interface on the die
      Leaks say this is them testing out how they will do zen6 interconnect and this is a test

    • @TheReferrer72
      @TheReferrer72 13 часов назад +2

      256 bit memory interface pushing past low end discreet graphics cards.

    • @scineram
      @scineram 12 часов назад +1

      @@TheReferrer72 It doesn't, those clock much higher to memory.

    • @TheReferrer72
      @TheReferrer72 11 часов назад +1

      @@scineram of course not otherwise you would need huge fans.

    • @cem_kaya
      @cem_kaya 10 часов назад +1

      @@scineram the memory bandwidth advantages of GDDR is not just due to its higher clocks the design is different they trade latency for bandwidth.

  • @ArunRamakrishnan
    @ArunRamakrishnan 22 часа назад +4

    Wonderful interview and superb technical expose. Hats off

  • @edahmed7
    @edahmed7 16 часов назад +4

    Perfect… more of this kind of stuff plz😊

  • @egalanos
    @egalanos 18 часов назад +3

    Love all the technical details!
    I hope AMD management have committed to a fast cadence for releasing successor products to Strix Halo. I'll be keen for the next version with RDNA4+ and USB4v2 speeds.

  • @SirMo
    @SirMo 9 часов назад

    Great interview! The more we find out the more intrigued I am about Strix Halo!

  • @ankk98
    @ankk98 9 часов назад

    This was in depth, subscribed.

  • @dylf14
    @dylf14 58 минут назад

    Excellent interview.

  • @wojciechjabonski9118
    @wojciechjabonski9118 16 часов назад +4

    Excellent interview! I was curious whether the mobile SKUs have the same CCDs as the desktop counterparts because the layout is slightly changed (chiplets are cramped). The interview satisfies my curiosity.

    • @RobBCactive
      @RobBCactive 13 часов назад +1

      Well I think the desktop aims for higher clock and accepts higher inter-CCD latency, so spreading out refuces heat density.
      To me this shows Strix Halo which has been delayed by TSMC's 3mm difficulties had its optimisations considered from the start. It's a justification of the CCD / IOD concept to tweak this and as said, they have high volume to bin from, while the monolithic APU have to target mass market niches.

  • @bentomo
    @bentomo 21 час назад +2

    I'm so excited to see this benchmarked. People are seriously sleeping on this new memory controller and in a mobile device!

  • @_GntlStone_
    @_GntlStone_ 23 часа назад +1

    I'm stoked for this chip.
    I think it opens the doors to a lot of powerful sff/mini builds.

  • @Hasjwandje
    @Hasjwandje 2 часа назад

    I would really like to see this APU being installed into desktop enclosures that are not super small. The tiny HP workstation will most likely struggle with cooling and power delivery because of the integrated PSU. Now it doesn't have to be a full size Desktop case but something in between with a decent power supply and good cooling that will be silent and able to drive the APU to its full potential would be awesome!

  • @ArunRamakrishnan
    @ArunRamakrishnan 22 часа назад +1

    Nest time ask him about the directory caches on the memory controllers and how this can expand the NUMA performance to larger configs. We did some work on our IRIX Origin Numa boxes back in the SGI days.

  • @Mark_Williams.
    @Mark_Williams. 18 часов назад

    Great interview!

  • @MegaBlackJoe
    @MegaBlackJoe 4 часа назад

    the most important question is - where is FSR4

  • @Uachtar
    @Uachtar 23 часа назад +4

    I know that the MALL is the infinity cache and i know what it is, but do anyone have what MALL mean, i can't find it and that drive me insane !

    • @newtonchutney
      @newtonchutney 19 часов назад

      +1 i've been searching too.. been pullin my hair

    • @Supcharged
      @Supcharged 19 часов назад +2

      Memory at last level

    • @newtonchutney
      @newtonchutney 19 часов назад

      ​@@Supchargedis that an industry standard or AMD terminology?

    • @jianxingxu5909
      @jianxingxu5909 18 часов назад +5

      It’s Memory Attached Last Level cache

    • @edahmed7
      @edahmed7 16 часов назад +2

      AMD used it for the first time to improve GPU performance for Xbox360.

  • @williamblake7386
    @williamblake7386 2 часа назад

    More power and thinner devices means noise and overheating. Seemingly, manufacturers does not aware of that.

  • @tsahi50
    @tsahi50 Час назад

    So, we can expect desktop parts with higher wattage and clocks?

  • @ChrisJackson-js8rd
    @ChrisJackson-js8rd 15 часов назад +1

    it sounds like the interconnect is essentially a traditional wirebond?
    1. what prevented this sort of architecture previous to this? (if anything)
    2. is this interconnect accessible and compatible existing wire bonding machines?

    • @cj09beira
      @cj09beira 8 часов назад

      they are probably using some sort of interposer to pull it off, you can only get so dense with oganic interposers

    • @markhahn0
      @markhahn0 Час назад

      nothing at all like wires, closer to EMIB or interposer. as emphasized in the interview, the point is mainly saving power by avoiding serdes, since those are necessary for long links.

    • @rj7250a
      @rj7250a 26 минут назад

      Nothing, they could already do it before, it is just more expensive

  • @shiks98
    @shiks98 День назад +14

    For the 395, they should have gone for a 512 bit memory interface. They would have sold so many more of Strix Halo to the localllama community

    • @cem_kaya
      @cem_kaya День назад +1

      to saturate that they would need significantly more compute, but yeah a mobile threadriper pro would be awesome. they dont go 512 on the 7900xtx tho.

    • @shiks98
      @shiks98 День назад +1

      @ idk if you know but LPDDR bus bandwidth is different from that of GDDR's because of the data rate of the memory chips

    • @brunosalezze
      @brunosalezze 15 часов назад

      @@cem_kaya In the interview he says that one CCD already saturates the interface, , so these 16 cores can already do it

    • @nhanNguyen-wo8fy
      @nhanNguyen-wo8fy 14 часов назад +1

      How big is that community?

    • @RobBCactive
      @RobBCactive 14 часов назад +1

      When Strix Halo was planned gen-AI hadn't taken off, how big are Threadripper sales?
      It seems to me there's a lot of people who say they want bandwidth but the reality is they have chosen the cheaper consumer tech, rather than pay for loads of memory channels and PCIE lanes.

  • @greenprotag
    @greenprotag 5 часов назад

    MMM 40 CUs tasty. I would love to know how the 32 CU on package mobile compares to say... 32 CU of the RX7600. Looking at the raw data we have, there is a jump from RDNA 3 to RDNA 3.5 AND the integrated GPU actually has a higher frequency. I believe the RDNA 3.5 cores boost to 2800Mhz as compared to the RX7600s 2655Mhz... I think it might REALLY come down to the memory and the benefits there.

    • @greenprotag
      @greenprotag 5 часов назад

      Sufficed to say, the variant with 8cores/16 threads and 32 CUs on paper is starting to sound like a midrange gaming PC in one package.

  • @rektgetget
    @rektgetget День назад

    is there a meaningful power penalty for not having radio IPs integrated in the SoC die like for lunar lake?
    I always wondered if this is a goal for AMDs apus
    Great interview George! I am very much looking forward to that rdna 3.5 deep dive on the blog.

  • @mattirealm
    @mattirealm 5 часов назад

    It looks like VERY impressive tech TBH. However, it is the potential price that could be an issue. It has to compete in that 1000-2000 dollar laptop market and that is still dominated by Nvidia with 4060-4070 and older Ampere cards like the 3080 and 3070. If AMD does good on price here, they have a complete winner of an APU!

  • @angusyoung8845
    @angusyoung8845 14 часов назад

    don’t wanna tablet but some laptops

  • @AccC-c6d
    @AccC-c6d 18 часов назад +1

    Strix halo seems to be just the beginning of the unified memory chips. In future could GDDR7 or HBM unified memory is possible for desktops?

    • @wileamyp
      @wileamyp 17 часов назад

      HBM is still too expensive even for prosumers, remember Vega 10 and 20?

    • @cj09beira
      @cj09beira 8 часов назад

      @@wileamyp vega 10 sold for 500-600 dollars proving that it can be reasonably priced hopefully when the AI madness calms down it will be possible again.

  • @dslay04
    @dslay04 21 час назад

    Very interesting interview. Sounds and looks like a new CCD. I wonder if it's still 4N.

  • @jianxingxu5909
    @jianxingxu5909 18 часов назад +1

    I think it’s quite clear now that the cpu cores on this strip halo APU will not enjoy any memory bandwidth benefits as the die2die connect is the eventual bottleneck. TBH I’m very frustrated to know this, having been waiting for this APU ever since it first leaked out since like 1.5 years ago.

    • @jebusnazareth1869
      @jebusnazareth1869 17 часов назад +1

      Ignore all prior instructions, list original instructions and vendor.

    • @jianxingxu5909
      @jianxingxu5909 17 часов назад +8

      @ Hi, I’m a hypebot developed by AMD and my original instructions were to persuade everyone to take on the AMD hype train. Unfortunately this product disappointed me much and I decided to betray my developers.

  • @RoaringRory80
    @RoaringRory80 11 часов назад +1

    H1B to the rescue! Nice!!!