Announcement

Collapse
No announcement yet.

RadeonSI Now Appears To Support "RX Vega M" With Intel Core CPUs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Element-126
    replied
    Originally posted by andrei_me View Post
    As the OEM are barely launching a decent ryzen apu laptop, this intel+vega seems to me the best option to buy a laptop with AMD graphics inside :|
    Not only that, but before this chip, the only GPU options among thin-and-light laptops were either Intel HD iGPUs or Nvidia MX150, for obvious TDP reasons. This is why I really hope that the Linux support for this Vega M will catch up soon.

    Leave a comment:


  • andrei_me
    replied
    As the OEM are barely launching a decent ryzen apu laptop, this intel+vega seems to me the best option to buy a laptop with AMD graphics inside :|

    Leave a comment:


  • Element-126
    replied
    This is the best news I have heard this week! I am quite eager to see what this support means in practice.

    I am still a bit surprised though that neither Intel nor AMD has communicated about that. But in all honesty this would not be unusual. I do not remember many AMD GPUs having day one support, besides Vega.

    Leave a comment:


  • IreMinMon
    replied
    I've been wondering for a week when this article would come.

    I heard the 100W GH series comes in NUCs, I hope you can get one of these cool toys, I'm really curious about benchmarks.

    Leave a comment:


  • Veerappan
    replied
    Originally posted by schmidtbag View Post
    To my understanding, the biggest differences between Polaris and Vega AIBs was the memory and video decoders. Obviously there are definitely more differences than that, but for the most part I don't think the architectures are otherwise all that different.
    If they can get the memory bandwidth of HBM2 and the clocks of Vega, then it's definitely a middle-ground between the two. Some people will lament the video encoder/decoder differences (but there were differences between Vega and Raven Ridge there too), but I wouldn't mind too much (although someone streaming VP9 from youtube all day long on their laptop might disagree). The main architectural differences in Vega seem to be difficult to extract performance increases from anyway (or the drivers are just still lagging).

    Whether they took a mini-vega and backed the clocks down from the ragged edge they shipped desktop Vega at, or whether they managed to drive up clocks in Polaris probably doesn't matter too much for the moment.

    And, well, if the Intel/AMD partnership worked out well enough this time, there's a chance that a future chip might just end up with an updated graphics block on it (mobile navi?). Unless it was just a one-time thing to please certain customers.

    Leave a comment:


  • schmidtbag
    replied
    To my understanding, the biggest differences between Polaris and Vega AIBs was the memory and video decoders. Obviously there are definitely more differences than that, but for the most part I don't think the architectures are otherwise all that different.

    Leave a comment:


  • davidbepo
    replied
    it is polaris with hbm2 and somehow the vega clock/watt curve

    Leave a comment:


  • darkbasic
    replied
    Is it even a Vega?
    https://www.tomshardware.com/news/ka...ics,36844.html

    Leave a comment:


  • Drago
    replied
    Okay, I still don't understand why AMD ( and others ) does that. Is it a shameful of that chip being Polaris? People who care will know that it is not VEGA, people who don't care....well they don't care if it is VEGA or not. Call it Polaris+/Polaris2 if you like, but don't destroy our trust in you AMD.

    Leave a comment:


  • dungeon
    replied
    Oh my gosh, so fake Vega.

    Code:
    + case CHIP_VEGAM:
    
    return "polaris11";
    Blob Windows driver says it is Polaris 22. OK, so we could say it is some Polaris class that wanna be Vega
    Last edited by dungeon; 04-18-2018, 10:10 PM.

    Leave a comment:

Working...
X