AMD Vega Hardware Reviews

Discussion in 'Architecture and Products' started by ArkeoTP, Jun 30, 2017.

  1. fellix

    fellix Hey, You!
    Veteran

    Joined:
    Dec 4, 2004
    Messages:
    3,506
    Likes Received:
    424
    Location:
    Varna, Bulgaria
    So, Vega is pretty much an optical shrink of Fiji with overstretched pipeline and minor ISA additions.
     
  2. eastmen

    Legend Subscriber

    Joined:
    Mar 17, 2008
    Messages:
    11,172
    Likes Received:
    2,233
    http://www.anandtech.com/show/11717/the-amd-radeon-rx-vega-64-and-56-review/8

    It looks like the vega 56 soundly beats the 1070 and sometimes the 1080 fe the few times the 1070 wins its at only 1080p . The one exception is GTA 5 where the 1070 craps all over even the vega 64 .

    Vega 64 performs similar to the 1080fe where sometimes it can stretch to come close to a 1080ti with it typicly staying at about the 1080fe speed.


    One interesting thing is that in some tests its within in a frame or two of the 56 edtion . Whats going on here that in Dawn of war 3 its 1-3 fps diffrent ? Thats the margin of error .

    Anyway it seems like the 56 is a good card to pick up. Its priced the same as the 1070 and performs mostly better than it.

    For me the power usage would drop down from my 290 and performance would sky rocket . So if prices don't sky rocket for mining and I can get it for $400 on amazon with my discount knocking it to $360 i think its a solid choice for me to get. But again i will see how hard they are to get
     
    #782 eastmen, Aug 14, 2017
    Last edited: Aug 14, 2017
    Lightman likes this.
  3. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    11,087
    Likes Received:
    5,633
    Tomshardware did.
    31MHash/s, clearly not worth the miners' money.
    Though anandtech did discuss it a bit and suggested these are still unoptimized results and miners are bound to explore even bios hacks to increase the mining performance.
     
    Lightman likes this.
  4. Picao84

    Veteran Regular

    Joined:
    Feb 15, 2010
    Messages:
    1,840
    Likes Received:
    901
    Consumer Pascal was pretty much the same over Maxwell. The exception was that Pascal had a much better base to grow on.
     
  5. Mize

    Mize 3dfx Fan
    Moderator Legend Veteran

    Joined:
    Feb 6, 2002
    Messages:
    5,079
    Likes Received:
    1,149
    Location:
    Cincinnati, Ohio USA
    From PC Gamer:

    "The Vega 56 manages around 31MH/s for Ethereum, and the Vega 64 does 33MH/s. Overclocking the VRAM helps boost both cards closer to 40MH/s right now. So at launch, that's not super promising."

    There you have it.
     
    Lightman likes this.
  6. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,289
    Likes Received:
    3,550
  7. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,137
    Likes Received:
    3,036
    Location:
    Finland
    Komplett.no and .se just listed their prices, they're actually what they should be (4999 norwegian kronor / 5090 swedish kronor cheapest, which is about 533 euros, prices include VAT)
     
  8. Anarchist4000

    Veteran Regular

    Joined:
    May 8, 2004
    Messages:
    1,439
    Likes Received:
    359
    Stop what? Just because you don't like the facts doesn't make them wrong. What I said is indicated in the first review linked and it doesn't take a genius to reach a similar conclusion or not reject the evidence.

    At 250-275W with balanced settings Vega LE is around 1080ti performance in Battlefield 1. Maybe 5% under. Stick 20-30% more on that from FP16 and the math is real simple. That doesn't even account for the inevitable driver fixes as obvious there are still some issues in places.

    Not sure about similar power, but the faster card may be using a bit more power in that case. Explained the math above, but doesn't seem much of a stretch based on what devs have ave achieved so far. Only a problem if buying Vega to play old games at really high framerates.

    In current titles sure, but upcoming titles that actually use the new features? With no other changes the FP16 alone would make them equal.
     
  9. silent_guy

    Veteran Subscriber

    Joined:
    Mar 7, 2006
    Messages:
    3,754
    Likes Received:
    1,380
    It'd be an even better deal for consumers.

    But it'd still be a 484mm2 chip with HBM that competes against one that's 315mm2 with GDDR5X and still a bad chip, Jerry, very very bad.

    Fury X was released shortly after the 980Ti, with almost equal perf and perf/mm2. Vega is in R600 territory.

    Forget about Navi, Vega simply has no business performing the way it does. They should fix whatever bottleneck bug there is and relaunch in 9 months or so. Though power may issue if it runs at full perf/mm2 efficiency.
     
    Lightman likes this.
  10. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,137
    Likes Received:
    3,036
    Location:
    Finland
    Err, it's the biggest change in architecture since first GCN :???:
     
    Anarchist4000 likes this.
  11. eastmen

    Legend Subscriber

    Joined:
    Mar 17, 2008
    Messages:
    11,172
    Likes Received:
    2,233
    Why forget about Navi ? AMD should be putting all its resources in getting Navi out the door . If they can have navi out in 9 months its best for them especially if its not based on GCN but something new. Leave the old stuff behind and just forge foward. In the anand review sometimes the 56 is 1-3 frames behind the 64 so its obvious something isn't scaling anymore in GCN on certian engines
     
  12. Scott_Arm

    Legend

    Joined:
    Jun 16, 2004
    Messages:
    14,209
    Likes Received:
    5,634
    dx12 performance looks pretty good from what I've seen, but that power consumption would be a non-starter for me. My apartment already gets hot.
     
    BRiT likes this.
  13. fbomber

    Newcomer

    Joined:
    Jun 9, 2004
    Messages:
    156
    Likes Received:
    17
    Are you really being serious?
     
    homerdog likes this.
  14. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,289
    Likes Received:
    3,550
    In one game with botched up API support? What about the dozens where it's actually only 1070 level of performance? What about the weak 1080p performance across the board?
    Where are the facts? I don't see them, link please!
    First there was the magic drivers, then DSBR, then now of course FP16. Yeah just like Async Compute, we all know how well that one turned out. (hint: not well).
     
  15. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    1,519
    Likes Received:
    879
    Location:
    France
    I guess he said forget about Navi because AMD need a faster card before Navi arrive. Navi is, what, 1,5 years away ? Best case scenario ? So...
     
  16. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    1,519
    Likes Received:
    879
    Location:
    France
    (Well async gave a nice small free boost for my Fury X. But yeah I agree, this "there is a secret sauce somewhere" stuff need to stop)
     
  17. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,137
    Likes Received:
    3,036
    Location:
    Finland
    Best case scenario it's about exactly 1 year away, not 1,5 (glofo claims they're ready for mass production in 2nd half of 2018 with the 7nm "Leading Performance" -process)
     
    Rootax likes this.
  18. w0lfram

    Newcomer

    Joined:
    Aug 7, 2017
    Messages:
    219
    Likes Received:
    39
    DX11 is almost 10 years old. Only outdated and re-hashed franchises are still using it. Vega is for the new era..
     
  19. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,137
    Likes Received:
    3,036
    Location:
    Finland
    Lightman and silent_guy like this.
  20. silent_guy

    Veteran Subscriber

    Joined:
    Mar 7, 2006
    Messages:
    3,754
    Likes Received:
    1,380
    Yes. That and the fact that a quick incremental fix is much lower risk to pull off. They should also move to TSMC because that will give us an apples to apples comparison wrt clocks and perf/W. :)
     
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...