Sticky

AMD Vega MegaThread! FAQ and Resources



Welcome Community, to the Official AMD Vega MegaThread!

**Updated 6/19/2017**

This thread serves as the primary discussion thread for ALL Vega information and resources. While discussing, please remember to stay within the guidelines set by Tom's Hardware and above all, DO NOT start a flame war.

AMD announced Radeon Vega earlier this year and followed up with news of its Vega Frontier Edition at its 2017 Financial Analyst Day. AMD claims the Frontier Edition is the "world's first" GPU geared for AI (Artificial Intelligence), creatives, and science pioneers.


Pricing
Radeon Vega Frontier Edition will retail at $1199 for the air cooled model and $1799 for the liquid cooled card.

Release Date
The Vega Frontier Edition is slated for release on June 27th, 2017

Memory Architecture
The card features 16GB of HBC (High Bandwidth Cache), which is AMD's new term for HBM (High Bandwidth Memory). The new card also features support for 8K displays, and it can access up to 256TB of virtual memory. This voluminous capacity is useful for crunching, video processing, and deep learning algorithms.


The Vega NCU (Next-Generation Compute Unit)
AMD is updating and revamping it's compute engine. The focus this time around is on higher clock speeds as well as an increased amount of IPCs. The NCU is very much a jack of all trades designed to excel no matter the application. Consoles, datacenters, and PC gamers all stand to benefit from the Vega's improved Compute Engine.

Performance
The Vega reveal did have some preview boards on hand, demoing id Software's Doom at 4K resolution under its most taxing detail settings. The preview board was able to t hit 70+ FPS despite having optimized software. Many in the industry expect the Vega cards to land somewhere between a 1080 and Titan X performance wise.

Performance chart courtesy of AnandTech

News and Future Updates
AMD shared some basic performance data on its blog and provided a rough overview of the new Frontier Edition.

  • Machine learning: Together with AMD’s ROCm open software platform, Radeon Vega Frontier Edition enables developers to tap into the power of Vega for machine learning algorithm development. Frontier Edition delivers more than 50 percent more performance than today’s most powerful machine learning GPUs.
  • Advanced visualization: Radon Vega Frontier Edition provides the performance required to drive increasingly large and complex models for real-time visualization, physically-based rendering and virtual reality through the design phase as well as rendering phase of product development.
  • VR workloads: Radeon Vega Frontier Edition is ideal for VR content creation supporting AMD’s LiquidVR technology to deliver the gripping content, advanced visual comfort and compatibility needed for next-generation VR experiences.Revolutionized game design workflows. Radeon Vega Frontier Edition simplifies and accelerates game creation by providing a single GPU optimized for every stage of a game developer’s workflow, from asset production to playtesting and performance optimization.
Reply to _Johnny5
694 answers Last reply
More about amd vega megathread faq resources
  1. Interesting. A lot of users who are building new computers really prefer to wait for Vega. If it is indeed going to perform between a GTX 1080 and Titan X, and the price of a GTX 1070, then that's a steal. We may see.
    Reply to Herc08
  2. The new cards will certainly drive down prices on all existing cards ... expect nVdia to play their spoiler role again, a practice that started with the 7xx series. Waiting for AMD to show their hand and then stealing the thunder / excitement by announcing the 1080 Ti is almost a certainty worth betting the farm on. Be nice if AMD is ready this time and keeps something on the shelf till nVidia shows their hand.

    As for THG promoting a "Mega Thread", from my PoV, pleas don't.... Mega Thread topics soon contain such a wide range of topics that reading them becomes too cumbersome and not worth the effort... having to read / bypass 21 posts between the ones you see that actually address the data one might be searching for tends to make them less useful as a data source.
    Reply to JackNaylorPE
  3. Hitting 70+ FPS in Doom at 4k it does make me wonder if the high end Vega will be closer to Titan X performance than the GTX 1080. I wonder if they still have some optimizations left to do in software/drivers since we are talking 4-5 months from now. Considering Nvidia has done very well in OpenGL and AMD has fared better on Directx12, Vega's 70+ FPS on an OpenGL game makes me wonder where the card will really land. I'm not in the market but I hope they do get very close to Titan X performance so we can have some serious competition and get these crazy high prices from Nvidia back down.
    Reply to JamesSneed
  4. JamesSneed said:
    Hitting 70+ FPS in Doom at 4k it does make me wonder if the high end Vega will be closer to Titan X performance than the GTX 1080. I wonder if they still have some optimizations left to do in software/drivers since we are talking 4-5 months from now. Considering Nvidia has done very well in OpenGL and AMD has fared better on Directx12, Vega's 70+ FPS on an OpenGL game makes me wonder where the card will really land. I'm not in the market but I hope they do get very close to Titan X performance so we can have some serious competition and get these crazy high prices from Nvidia back down.


    What I would like to see is nVidia push for DX12 support. Not saying I want them to be better than AMD, but Windows is pretty popular and devs will continue to use DX no doubt. I just want to see true competition. Before Polaris, it seem nVidia had a leg up in the game, but AMD is surprising us all. But getting that 70 FPS at 4K is an amazing feat. I suspect it will be around $500-600. One thing I want to knowis, will there be a card to match the GTX 1070? Not everyone has money to get a 4K monitor, and Ultra wides are popular at the 1440p level.
    Reply to Herc08
  5. At the end of the free year ... 1 of 8 eligible Win 7 / 8 users had upgraded to Win10. MS's frustration at the "failure to update" reception of the free offer was what led to the Borg like ... "You will be assimilated" tactics that only served to drive more away from taking the plunge. What I have noticed in the past is, every time there's a new API, AMD jumps on it and the marketing machine pushes out "look we're better in this new thing ... check out this game demo benchmark for a game that won't be out for a year", ... and than it's out and the gap is nothing to get excited about.

    Not that that is a bad thing in any way unless of course the focus on that results in less focus on addressing what is out now. We still see each side doing better in different games, but as we still see reviewers reporting results for DX12 games in DX11 because the DX12 implementation is still boinky, DX12 to my eyes is going to be a long transition with games started in development 3 years ago w/ DX11 now being adapted for DX12.

    Windows 8 was "gonna change everything" ... and pre-release we saw lotsa benchies showing it would ... but it didn't, most game benches were slower. So I think it's gonna be quite some time before we can makes heads or tails out of DX12.

    The 290x "was gonna crush" the nVidia 780, but nVidia dropped the Ti ... and then we found out with both cards OC'd, it didn't even top the 780. But it was close and the competition dropped prices for everyone ... I saved $300 off the prices from 2 weeks before buying 2 780s (SLI) compared to the prices on the 780s before the new cards dropped.

    As long as you don't want the card that "has the title", you made out good. The 39x series didn't work out so well against the 9xx.... w/ the 970 selling at more than 2 times the rate of all R7 / R9 2xx and 3xx cards combined. I still believe that nVidia used predatory pricing on the 970, not to compete, but to purposely hurt AMD financially during a tough stretch.

    If I won a 4k monitor, I'd sell it .... holds no attraction to me at 60 Hz. From an investment in technology perspective, when the 1st Display Port 4k monitor hits the shelves, the value of everything else at 4k is going to drop in the toilet. I noticed one think in the last year or so is that hardware srveys no longer break out display resolution above 2560 x 1440 ... now whydooyathink that is ?Seems the industry doesn't like folks having access to the fact that adoption of 4k remains at sub 1% levels and feel that having that thrown out there for public consumption is being perceived as stifling investment in new monitors.

    Right now you can do quite well at $200 @ 1080p, the 1070 is just fine at 1440p.... why the Titan keeps coming up is kinda discussion beyond me because the only person who'd pick one for a gaming box is the guy who puts to together a build by selecting the Graphics card filter on newegg or PCPP and sorts everything by highest price... with the 9xx series, the highest performing gaming card was not the one at the top; the 980 Ti was. Leave the Titan to those who wanna play games on their workstation box w/o killing rendering performance.

    But I think you hit the nail on the head with the 1070 ... that's where the market is. The 970 was priced weirdly low such that it made everybody who normally budget a 960 level card spring for the 970. From nVidia's side, it didn't work out as well as the 980 became a non-factor with many at that budget range springing for dual 970s instead of the 980. Obviously nvidia made more money selling one 980 than two 970s and I suspect that's why we see SLI performance way way down from last generation.

    That (x70 tier) is what AMD should have their sights set on....The 970 remains the most popular card in use today. AMD has only one entry in the top 20 and that's not a card but a card series. There are 4.2 970s in use for every 79xx card in use and almost 10 970s in use for every 390 series card in use. The 1070 is 63% faster than the 970 ... and with no competition, nVidia's been able to rake it in.

    So this is where they have to make their mark. Ya gotta think that AMD wouldn't have sat back for so long with no new entries until they could show a win in this sector. Great news for us,.. ya know there's $100 of room in the 1070 prices so if we see 1070 like performance at $300 from AMD, it's good news for everyone.
    Reply to JackNaylorPE
  6. JamesSneed said:
    Hitting 70+ FPS in Doom at 4k it does make me wonder if the high end Vega will be closer to Titan X performance than the GTX 1080. I wonder if they still have some optimizations left to do in software/drivers since we are talking 4-5 months from now. Considering Nvidia has done very well in OpenGL and AMD has fared better on Directx12, Vega's 70+ FPS on an OpenGL game makes me wonder where the card will really land. I'm not in the market but I hope they do get very close to Titan X performance so we can have some serious competition and get these crazy high prices from Nvidia back down.


    The demo was probably made using the Vulkan renderer for Doom. AMD's OpenGL drivers suck performance-wise for gaming, they maintain it primarily for workstations (according to bridgeman, an AMD developer who frequents Phoronix forums - the closed source driver on Linux shares most of its code with the Windows one),
    Reply to mitch074
  7. JackNaylorPE said:
    At the end of the free year ... 1 of 8 eligible Win 7 / 8 users had upgraded to Win10. MS's frustration at the "failure to update" reception of the free offer was what led to the Borg like ... "You will be assimilated" tactics that only served to drive more away from taking the plunge. What I have noticed in the past is, every time there's a new API, AMD jumps on it and the marketing machine pushes out "look we're better in this new thing ... check out this game demo benchmark for a game that won't be out for a year", ... and than it's out and the gap is nothing to get excited about.

    Not that that is a bad thing in any way unless of course the focus on that results in less focus on addressing what is out now. We still see each side doing better in different games, but as we still see reviewers reporting results for DX12 games in DX11 because the DX12 implementation is still boinky, DX12 to my eyes is going to be a long transition with games started in development 3 years ago w/ DX11 now being adapted for DX12.

    Windows 8 was "gonna change everything" ... and pre-release we saw lotsa benchies showing it would ... but it didn't, most game benches were slower. So I think it's gonna be quite some time before we can makes heads or tails out of DX12.

    The 290x "was gonna crush" the nVidia 780, but nVidia dropped the Ti ... and then we found out with both cards OC'd, it didn't even top the 780. But it was close and the competition dropped prices for everyone ... I saved $300 off the prices from 2 weeks before buying 2 780s (SLI) compared to the prices on the 780s before the new cards dropped.

    As long as you don't want the card that "has the title", you made out good. The 39x series didn't work out so well against the 9xx.... w/ the 970 selling at more than 2 times the rate of all R7 / R9 2xx and 3xx cards combined. I still believe that nVidia used predatory pricing on the 970, not to compete, but to purposely hurt AMD financially during a tough stretch.

    If I won a 4k monitor, I'd sell it .... holds no attraction to me at 60 Hz. From an investment in technology perspective, when the 1st Display Port 4k monitor hits the shelves, the value of everything else at 4k is going to drop in the toilet. I noticed one think in the last year or so is that hardware srveys no longer break out display resolution above 2560 x 1440 ... now whydooyathink that is ?Seems the industry doesn't like folks having access to the fact that adoption of 4k remains at sub 1% levels and feel that having that thrown out there for public consumption is being perceived as stifling investment in new monitors.

    Right now you can do quite well at $200 @ 1080p, the 1070 is just fine at 1440p.... why the Titan keeps coming up is kinda discussion beyond me because the only person who'd pick one for a gaming box is the guy who puts to together a build by selecting the Graphics card filter on newegg or PCPP and sorts everything by highest price... with the 9xx series, the highest performing gaming card was not the one at the top; the 980 Ti was. Leave the Titan to those who wanna play games on their workstation box w/o killing rendering performance.

    But I think you hit the nail on the head with the 1070 ... that's where the market is. The 970 was priced weirdly low such that it made everybody who normally budget a 960 level card spring for the 970. From nVidia's side, it didn't work out as well as the 980 became a non-factor with many at that budget range springing for dual 970s instead of the 980. Obviously nvidia made more money selling one 980 than two 970s and I suspect that's why we see SLI performance way way down from last generation.

    That (x70 tier) is what AMD should have their sights set on....The 970 remains the most popular card in use today. AMD has only one entry in the top 20 and that's not a card but a card series. There are 4.2 970s in use for every 79xx card in use and almost 10 970s in use for every 390 series card in use. The 1070 is 63% faster than the 970 ... and with no competition, nVidia's been able to rake it in.

    So this is where they have to make their mark. Ya gotta think that AMD wouldn't have sat back for so long with no new entries until they could show a win in this sector. Great news for us,.. ya know there's $100 of room in the 1070 prices so if we see 1070 like performance at $300 from AMD, it's good news for everyone.


    Instead they got the RX480 out, which can do 1080P flawlessly and handles 1440p rather well, and can even play in the 1070's ballpark on games that actually cater to the newer APIs out there (not that black box Nvidia proprietary GameWorks stuff) : DX12 and Vulkan : 80% of the performance for 60% of the price, it may sway some. Too bad it's such a good deal that the price for it actually went UP since it came out.
    Reply to mitch074
  8. So I finally put together a new build after many, many years. I have two old 560 ti's in sli that need replacing. They are among the only parts I retained in the new build. I was hoping, like many people, that we'd see something about the 1080ti at CES. If we are now in the game of waiting until May or June to be able to own a Vega or 1080 ti, is it likely to be worth the wait? I'm guessing my answer is going to be yes, as I want to make the jump to 4k with my next card purchase, but now it's looking like another six months to own the next gen cards instead of January. What are you guys thinking about the delay? Anyone else out there in a similar situation?
    Reply to axlrose
  9. axlrose said:
    So I finally put together a new build after many, many years. I have two old 560 ti's in sli that need replacing. They are among the only parts I retained in the new build. I was hoping, like many people, that we'd see something about the 1080ti at CES. If we are now in the game of waiting until May or June to be able to own a Vega or 1080 ti, is it likely to be worth the wait? I'm guessing my answer is going to be yes, as I want to make the jump to 4k with my next card purchase, but now it's looking like another six months to own the next gen cards instead of January. What are you guys thinking about the delay? Anyone else out there in a similar situation?


    I would wait until the vega 10/1080ti drops, when both cards are on the market and benchmarks can be compared you can bet that prices won't be as high as when just one company is at the top. I am running a acer xr34 with a 7970HD. I badly need to upgrade as my train simulators dip to 5fps all the time the 7970 just can't keep up with my monitor upgrade. I think only 3GB vram is killing me. So I am in the same boat with you but I am trying to hold off for vega myself as I strongly believe that it will make the RX 480 look like a childs toy.

    On an edit:Strong belief is that vega 10 will drop at the end of may so I would suspect that the 1080ti won't be to far off that as nvidia won't let amd dig to far into the gaming market if you want to go nvidia.
    Reply to Gforlife
  10. so close yet so far away....

    Yeah I'm in the same boat. Waiting. Waiting for CPU's and GPU's so it's twice the anticipation.
    Reply to Memhorder
  11. My own speculation is that AMD will put 2 flavors of Vega: GDDR5X and HBM2. That also implies that I think there will be 2 Vega siblings, just like with Polaris.

    Why do I think that? Because they can show off their fancy new "this is not memory, but buffer" idea. HBM2-pure cards, then HBM2+SSDs for the higher end and pro parts, then GDDR5 with no "buffer" lower tier stuff. Yes, they require different memory controllers and they are indeed different GPUs, right? Plus, we just saw with Polaris they are still on-board with using GDDR from a cost-efficiency perspective.

    So, any ideas or counter points on why it might not be good speculation?

    Cheers!
    Reply to Yuka
  12. I was probably gonna go enthusiast level to begin with, perhaps the 16GB model that might come to fruition. How ever if they over a card with onboard ssd and it only came in a 8GB model I would probably jump all over it!
    Reply to Gforlife
  13. Vega launch date as confirmed by Lisa Su yesterday*:
    April/May/June 2017.


    http://www.pcworld.com/article/3163500/components/amd-confirms-its-ryzen-cpu-will-launch-in-early-march-followed-by-the-vega-gpu.html
    "What's new, though, is a Vega timetable: Su revealed that the Vega GPUs will ship during the second quarter as well."

    Vega+Zen APU, anyone?
    "In the second half of 2017, AMD still plans to launch a Zen-based APU, codenamed “Raven Ridge,” primarily designed for notebooks but also some desktops."

    Via [H]ardOCP:
    http://www.hardocp.com/news/2017/01/31/amd_confirms_ryzen_vega_launch_timetables

    Edit: *Well, it's yesterday in my timezone. It might be tomorrow in yours... :-) #IHateTimeZones
    Reply to AndrewJacksonZA
  14. So long from now...
    Reply to axlrose
  15. AndrewJacksonZA said:

    "What's new, though, is a Vega timetable: Su revealed that the Vega GPUs will ship during the second quarter as well."



    This could be because of the HBM2 memory used if you look at the databook under graphics memory on skhynix.com the HBM2 is only 204.8GB/s not the 256GB/s that we expected. The 204.8GB/s will put it under the GTX 1080, perhaps they will roll out VEGA with the 204.8GB/s HBM2 then when its fully finished start shipping VEGA with the 256GB/s.Makes me think about getting an RX 470/480 to replace my 7970 and hold off until 256GB/s HBM2 is garanteed. The only other option is if they are gonna use samsung HBM2 but I don't know if they have reached 256GB/s either
    Reply to Gforlife
  16. mitch074 said:
    JamesSneed said:
    Hitting 70+ FPS in Doom at 4k it does make me wonder if the high end Vega will be closer to Titan X performance than the GTX 1080. I wonder if they still have some optimizations left to do in software/drivers since we are talking 4-5 months from now. Considering Nvidia has done very well in OpenGL and AMD has fared better on Directx12, Vega's 70+ FPS on an OpenGL game makes me wonder where the card will really land. I'm not in the market but I hope they do get very close to Titan X performance so we can have some serious competition and get these crazy high prices from Nvidia back down.


    The demo was probably made using the Vulkan renderer for Doom. AMD's OpenGL drivers suck performance-wise for gaming, they maintain it primarily for workstations (according to bridgeman, an AMD developer who frequents Phoronix forums - the closed source driver on Linux shares most of its code with the Windows one),


    the demo indeed running with vulkan API.
    Reply to renz496
  17. mitch074 said:
    JackNaylorPE said:
    At the end of the free year ... 1 of 8 eligible Win 7 / 8 users had upgraded to Win10. MS's frustration at the "failure to update" reception of the free offer was what led to the Borg like ... "You will be assimilated" tactics that only served to drive more away from taking the plunge. What I have noticed in the past is, every time there's a new API, AMD jumps on it and the marketing machine pushes out "look we're better in this new thing ... check out this game demo benchmark for a game that won't be out for a year", ... and than it's out and the gap is nothing to get excited about.

    Not that that is a bad thing in any way unless of course the focus on that results in less focus on addressing what is out now. We still see each side doing better in different games, but as we still see reviewers reporting results for DX12 games in DX11 because the DX12 implementation is still boinky, DX12 to my eyes is going to be a long transition with games started in development 3 years ago w/ DX11 now being adapted for DX12.

    Windows 8 was "gonna change everything" ... and pre-release we saw lotsa benchies showing it would ... but it didn't, most game benches were slower. So I think it's gonna be quite some time before we can makes heads or tails out of DX12.

    The 290x "was gonna crush" the nVidia 780, but nVidia dropped the Ti ... and then we found out with both cards OC'd, it didn't even top the 780. But it was close and the competition dropped prices for everyone ... I saved $300 off the prices from 2 weeks before buying 2 780s (SLI) compared to the prices on the 780s before the new cards dropped.

    As long as you don't want the card that "has the title", you made out good. The 39x series didn't work out so well against the 9xx.... w/ the 970 selling at more than 2 times the rate of all R7 / R9 2xx and 3xx cards combined. I still believe that nVidia used predatory pricing on the 970, not to compete, but to purposely hurt AMD financially during a tough stretch.

    If I won a 4k monitor, I'd sell it .... holds no attraction to me at 60 Hz. From an investment in technology perspective, when the 1st Display Port 4k monitor hits the shelves, the value of everything else at 4k is going to drop in the toilet. I noticed one think in the last year or so is that hardware srveys no longer break out display resolution above 2560 x 1440 ... now whydooyathink that is ?Seems the industry doesn't like folks having access to the fact that adoption of 4k remains at sub 1% levels and feel that having that thrown out there for public consumption is being perceived as stifling investment in new monitors.

    Right now you can do quite well at $200 @ 1080p, the 1070 is just fine at 1440p.... why the Titan keeps coming up is kinda discussion beyond me because the only person who'd pick one for a gaming box is the guy who puts to together a build by selecting the Graphics card filter on newegg or PCPP and sorts everything by highest price... with the 9xx series, the highest performing gaming card was not the one at the top; the 980 Ti was. Leave the Titan to those who wanna play games on their workstation box w/o killing rendering performance.

    But I think you hit the nail on the head with the 1070 ... that's where the market is. The 970 was priced weirdly low such that it made everybody who normally budget a 960 level card spring for the 970. From nVidia's side, it didn't work out as well as the 980 became a non-factor with many at that budget range springing for dual 970s instead of the 980. Obviously nvidia made more money selling one 980 than two 970s and I suspect that's why we see SLI performance way way down from last generation.

    That (x70 tier) is what AMD should have their sights set on....The 970 remains the most popular card in use today. AMD has only one entry in the top 20 and that's not a card but a card series. There are 4.2 970s in use for every 79xx card in use and almost 10 970s in use for every 390 series card in use. The 1070 is 63% faster than the 970 ... and with no competition, nVidia's been able to rake it in.

    So this is where they have to make their mark. Ya gotta think that AMD wouldn't have sat back for so long with no new entries until they could show a win in this sector. Great news for us,.. ya know there's $100 of room in the 1070 prices so if we see 1070 like performance at $300 from AMD, it's good news for everyone.


    Instead they got the RX480 out, which can do 1080P flawlessly and handles 1440p rather well, and can even play in the 1070's ballpark on games that actually cater to the newer APIs out there (not that black box Nvidia proprietary GameWorks stuff) : DX12 and Vulkan : 80% of the performance for 60% of the price, it may sway some. Too bad it's such a good deal that the price for it actually went UP since it came out.


    even with vulkan RX480 is nowhere close to 1070:

    http://www.guru3d.com/articles_pages/zotac_geforce_gtx_1080_arcticstorm_review,19.html

    the only thing i can imagine RX480 can close the gap with 1070 is with tittle that is extremely favoring AMD architecture.
    Reply to renz496
  18. AndrewJacksonZA said:
    Vega launch date as confirmed by Lisa Su yesterday*:
    April/May/June 2017.


    http://www.pcworld.com/article/3163500/components/amd-confirms-its-ryzen-cpu-will-launch-in-early-march-followed-by-the-vega-gpu.html
    "What's new, though, is a Vega timetable: Su revealed that the Vega GPUs will ship during the second quarter as well."

    Vega+Zen APU, anyone?
    "In the second half of 2017, AMD still plans to launch a Zen-based APU, codenamed “Raven Ridge,” primarily designed for notebooks but also some desktops."

    Via [H]ardOCP:
    http://www.hardocp.com/news/2017/01/31/amd_confirms_ryzen_vega_launch_timetables

    Edit: *Well, it's yesterday in my timezone. It might be tomorrow in yours... :-) #IHateTimeZones


    What I found most interesting about the article was:
    "Su added that the profit margins for both Ryzen and Vega will be well above AMD’s average."

    So much for the rumors that AMD would sell i7 level performance for i5 prices. If they plan to price Ryzen similar to competing Intel products and Vega similar to nVida, they must be very confident they have some winners there. That or they are getting really close to bankruptcy.

    Personally, I have no problems with AMD pricing things competitively, if they have the performance on par, then the price should be as well. I just hope that they finally do some advertising and let people know they aren't a cheap Intel knock-off, which is what most of my clients thought for a long time. They need to make themselves into a household name, like Intel did.
    Reply to Martell1977
  19. part of the problem is people expectation. if AMD wants to price their product competitively according to competitor performance people will expect AMD to win the benchmark. heck some people want AMD to win the benchmark while at the same time undercut nvidia/intel pricing. because to their eyes AMD is not the "greed" company like intel and nvidia is. i still remember the mantra some people said a few years back. "AMD will never going to charge more than $400 for a single GPU because AMD care about customer unlike nvidia". then bam they got slap with 7970 costing $550. forget about $400 AMD end up charging more than nvidia at the time for their top flagship.

    this is ultimately become one of AMD problem. we see a lot of Fury X being heavily discounted out there even before Vega comes out. for consumer it is definitely a good thing (in short term) but for AMD it is definitely not. that's why AMD did not coming out with a flagship that can compete directly with 1080 last year.
    Reply to renz496
  20. renz496 said:
    this is ultimately become one of AMD problem. we see a lot of Fury X being heavily discounted out there even before Vega comes out. for consumer it is definitely a good thing (in short term) but for AMD it is definitely not. that's why AMD did not coming out with a flagship that can compete directly with 1080 last year.


    There is mention of AMD taking time to "clear out the channel", which to me indicates that they are trying to clear out as much stock of old products before introducing the new. It's understandable, but it was in reference to Ryzen, but there is no reason it doesn't apply to the GPU's as well. With the money they are expecting to make, I don't think the loss they might take from discounts and clearing out the vendors stock will hurt them much.

    It will be an interesting year for sure, I just hope Vega can live up to the "leaks" or there might be some dark days ahead for AMD.
    Reply to Martell1977
  21. Martell1977 said:
    renz496 said:
    this is ultimately become one of AMD problem. we see a lot of Fury X being heavily discounted out there even before Vega comes out. for consumer it is definitely a good thing (in short term) but for AMD it is definitely not. that's why AMD did not coming out with a flagship that can compete directly with 1080 last year.


    There is mention of AMD taking time to "clear out the channel", which to me indicates that they are trying to clear out as much stock of old products before introducing the new. It's understandable, but it was in reference to Ryzen, but there is no reason it doesn't apply to the GPU's as well. With the money they are expecting to make, I don't think the loss they might take from discounts and clearing out the vendors stock will hurt them much.

    It will be an interesting year for sure, I just hope Vega can live up to the "leaks" or there might be some dark days ahead for AMD.


    but their competitor have no such problem.
    Reply to renz496
  22. renz496 said:
    but their competitor have no such problem.


    That's the problem with such small market-share. However, I see in Newegg's newsletter constant sales on nVidia cards with only a few Radeons sprinkled in.

    The market-share issue is especially relevant in their CPU business. They have gained a good amount on the GPU side, but not on the CPU side.

    Plus it didn't help that the Fury X got such a bad rep even though it was comparable to the 980ti. Then the RX 480's power draw debacle. They have great products, it's just that they get hammered much harder for minor missteps.

    I think that another issue with "clearing the channel" is the power draw of the last gen. There seems to be a plethora of computers out there with horrid PSU's and the added cost of replacing it turns a lot of people off.

    I'm waiting for Vega to possibly upgrade, which would be the quickest GPU upgrade I've ever done, but the performance gain could be awesome considering I'm at 480 levels now.
    Reply to Martell1977
  23. sometimes i was thinking this is a problem that AMD create for themselves. they start attacking with price very aggressively with HD4800 series. since then people expecting AMD to do that every time they release new GPU. hence when they tried to play the premium game by having the fastest GPU with 7970 many people are hesitant to jump right into it and waiting for nvidia offering first for price war. when you're launching your product first you should have the advantage to take more market share from your competitor (like what happen with 5k series vs 400 series) but it did not happen with 7970. instead AMD were losing market share instead.
    Reply to renz496
  24. Agreed, they have kind of cornered themselves with their price war so we will have to wait and see how it goes with Vega. the good thing is that nVidia already has everything out except the 1080ti (they love using those XX80ti's as a spoiler), but I have a hunch that AMD just might have an answer for it this time.

    AMD grabbed a good amount of market-share with Polaris and if they are on top of things, Vega could be a great spoiler. But I doubt they will catch nVidia sleeping, either they have a source inside AMD or are clairvoyant but AMD never seems to be able to blindside nVidia to get the upper hand for long.
    Reply to Martell1977
  25. if you look closely nvidia have always anticipate for AMD to have answers to their new GPU. when nvidia first launch 900 series it was like this: 980 > 780Ti > 970. nvidia are very aware that AMD should at least see this pattern. so with the new 10 series they make it like this: 1080 >1070 > 980Ti. note that this time around even the x70 are faster than previous gen flagship (at least in stock vs stock case). i think nvidia are expecting polaris are at least somehow able to match some of GP104 variant.

    also if you have been following the market share for each quarter AMD did not gain market share with polaris. AMD was gaining market share from nvidia in Q2 2016 while polaris only launch at the very last day of Q2 2016. AMD able to gain market share even without polaris so some people probably expect AMD will be able to gain much bigger with polaris. but the data from Q3 shows that AMD actually lose a very small market share to nvidia instead (less than 1% to be exact). that's mean nvidia somehow able to defend their market share with GP106 vs polaris. now we are waiting for Q4 numbers.

    http://jonpeddie.com/publications/add-in-board-report
    Reply to renz496
  26. While AMD didn't gain market share with Polaris, it did regain some mind share - up until Q2 2016, AMD was selling "heaters that could do games for cheap" while Nvidia sold sleek, quiet and powerful gaming chips with futuristic reference designs (for a price, and the odd half-hidden hardware limitation for the 970) and now that Polaris is out, powerful, not too expensive and yet building upon all the expectations AMD had built ever since they released the first GCN cards, they again became contenders for the sweet spot.

    I mean, where I live, the RX480 8Gb reference design was selling for MORE than the price I paid for it at launch (last summer) a couple months ago. Not bad for a card which was decried for its (admittedly lousy) reference cooler and (overdramatized) over-specification power draw. Current AIB cards are more powerful, more silent... And until recently regularly out of stock.
    Reply to mitch074
  27. did AMD still producing reference design for RX480/470? from what i heard those reference got very solid built but the cost probably very expensive for a card that meant to be sold at $250 (there are speculation about AMD probably intend on selling polaris on much higher price hence more expensive use of component on the reference design). some board partner like Gigabyte actually use more cheaper stuff on their custom PCB design.
    Reply to renz496
  28. renz496 said:
    did AMD still producing reference design for RX480/470? from what i heard those reference got very solid built but the cost probably very expensive for a card that meant to be sold at $250 (there are speculation about AMD probably intend on selling polaris on much higher price hence more expensive use of component on the reference design). some board partner like Gigabyte actually use more cheaper stuff on their custom PCB design.


    AMD provided a reference design for the RX480, but not for the RX470 (which would have been rather pointless, as it's the same chip with a couple of disabled shader clusters). Contrary to Nvidia, AMD's reference design was supposed to be the baseline for cooling, with the bare minimum cooling solution, and the most solid circuitry ever, with oversized voltage converters and a strangely short PCB. I'd recommend it for anybody looking to overclock the sh*t out of it on a watercooled compact rig, otherwise grab any other version which will be far more silent and/or clocked quite a bit higher.
    Reply to mitch074
  29. AMD’s Radeon RX 580 On Display Via Leaked Ashes Of The Singularity Benchmarks:

    http://hothardware.com/news/amd-radeon-rx-580-leaked-ashes-of-the-singularity-benchmarks
    Reply to jaymc
  30. Will the RX 580 be a Polaris refresh or based on Vega?
    Reply to BurgerandChips66
  31. I think the general consensus would of been Vega.. but I suppose nothing is certain.

    Vega is supposed to arrive around May. As stated earlier in this tread they may be just waiting on the HBM 2 bandwidth to increase as production is still ramping up an being tweaked. (hurry up will yis for gods sake).

    I think they should just release a slightly lower spec Vega with the slower HBM. This may let us see if Nvidia have counter punch waiting and also what it is. Think it's just the Titan for now (hopefully).. Isn't that enough to be dealing with eh.

    Although I did hear that they developed a special extra powerful GPU for the Auto-pilot in the Tesla's.. Anyone any info on this.. As I have been concerned about it, can it be used as an ace up the sleeve ??

    The taught's of a mole in AMD is disturbing to say the least..
    Reply to jaymc
  32. BurgerandChips66 said:
    Will the RX 580 be a Polaris refresh or based on Vega?


    hard to say. for one AMD has always been changing their naming scheme more often after HD7k/HD8k (OEM). with R200 and R300 they have R9, R7, R5 performance category. then they ditch that and just call them RX on every performance tier with RX400 series. RX 400 series don't even have x90 card.

    performance wise they said this "RX580" is in the ballpark of 1070. so if this is indeed a polaris based how much further you have to push the clock on the existing polaris to reach 1070 level?
    Reply to renz496
  33. jaymc said:
    I think the general consensus would of been Vega.. but I suppose nothing is certain.

    Vega is supposed to arrive around May. As stated earlier in this tread they may be just waiting on the HBM 2 bandwidth to increase as production is still ramping up an being tweaked. (hurry up will yis for gods sake).

    I think they should just release a slightly lower spec Vega with the slower HBM. This may let us see if Nvidia have counter punch waiting and also what it is. Think it's just the Titan for now (hopefully).. Isn't that enough to be dealing with eh.

    Although I did hear that they developed a special extra powerful GPU for the Auto-pilot in the Tesla's.. Anyone any info on this.. As I have been concerned about it, can it be used as an ace up the sleeve ??

    The taught's of a mole in AMD is disturbing to say the least..


    don't look at nvidia titan x (pascal). look at Quadro P6000 instead. that is the fastest GP102 configuration at the moment. if nvidia still can't beat Vega then nvidia probably have to make new chip that is bigger than GP102.

    nvidia most likely going to save volta for their 2018-2019 product line.
    Reply to renz496
  34. renz496 said:
    performance wise they said this "RX580" is in the ballpark of 1070. so if this is indeed a polaris based how much further you have to push the clock on the existing polaris to reach 1070 level?


    This very well could just be from small tweaks and process maturity. Kind of like Sky/Kaby Lake, we saw a performance boost and higher stock clocks from essentially the same architecture. However, Polaris getting a jump to 1070 performance levels is much more impressive.
    Reply to Martell1977
  35. Martell1977 said:
    renz496 said:
    performance wise they said this "RX580" is in the ballpark of 1070. so if this is indeed a polaris based how much further you have to push the clock on the existing polaris to reach 1070 level?


    This very well could just be from small tweaks and process maturity. Kind of like Sky/Kaby Lake, we saw a performance boost and higher stock clocks from essentially the same architecture. However, Polaris getting a jump to 1070 performance levels is much more impressive.


    but the question is it really possible for polaris to reach 1070 performance level? 1070 firestrike score is around 16k:

    http://www.guru3d.com/articles-pages/nvidia-geforce-gtx-1070-review,28.html

    polaris 10 at 1600mhz can get around 14000 with firestrike:

    https://www.youtube.com/watch?v=HYg123osc04

    and that's already on LN2. to reach 1070 performance RX480 probably would least need to reach 1700mhz on the core alone.
    Reply to renz496
  36. Weren't they supposed to have tweaked the fabrication process there recently.... ? I don't think it's Vega if it is they have stuck a limiter in.
    Reply to jaymc
  37. even if the process become better maybe they can push it to 1400mhz mark while maintaining current power consumption. IMO the chip is simply too small to really compete with nvidia GP104.
    Reply to renz496
  38. The problem for AMD here is how to tweak GCN to squeeze extra Mhz out of it without throwing power out of the window. nVidia did an excellent job of tweaking Pascal to get a lot of extra hertz out of the GPU.

    This is just a broad sentiment I have about GCN, but AMD is being very stubborn about not dropping some parts of GCN that just waste space in a card that would be aimed to the consumer market first and Pro market second. Having full HSA compliance when I haven't read anywhere it's being used is just stubbornness in my eyes.

    In any case, Vega... I'm still thinking they'll have a GDDR5X variant and a HBM2 variant.

    Cheers!
    Reply to Yuka
  39. Yuka said:
    The problem for AMD here is how to tweak GCN to squeeze extra Mhz out of it without throwing power out of the window. nVidia did an excellent job of tweaking Pascal to get a lot of extra hertz out of the GPU.

    This is just a broad sentiment I have about GCN, but AMD is being very stubborn about not dropping some parts of GCN that just waste space in a card that would be aimed to the consumer market first and Pro market second. Having full HSA compliance when I haven't read anywhere it's being used is just stubbornness in my eyes.

    In any case, Vega... I'm still thinking they'll have a GDDR5X variant and a HBM2 variant.

    Cheers!


    Thing is, HSA doesn't exactly consume silicon: it's a (actually successful) attempt at creating a publicly-documented GPU-based massively scalar coprocessor; the only thing you could call a "waste" are the hardware-based schedulers, graphics command units and asynchronous shader units - all elements that Nvidia also has (in some measure, although theirs is not HSA-compliant) and that games that are made to use them sure do enjoy (cf. Doom 2016 in Vulkan mode). They do require newer APIs to be used (DX12 or Vulkan, used as such and not as "DX11 with croutons") but they can be massively useful.

    Most current games still use older DX11 render paths even in DX12 mode; they mostly enjoy the lower CPU overhead on draw calls and multiple queue job submitting, but almost none make use of async compute or directly manageable VRAM addressing. Future Unreal Engine 4, Serious Engine or id Tech based games should finally start to make use of those technologies and really unleash the power of GCN cards.
    Reply to mitch074
  40. mitch074 said:
    Thing is, HSA doesn't exactly consume silicon: it's a (actually successful) attempt at creating a publicly-documented GPU-based massively scalar coprocessor; the only thing you could call a "waste" are the hardware-based schedulers, graphics command units and asynchronous shader units - all elements that Nvidia also has (in some measure, although theirs is not HSA-compliant) and that games that are made to use them sure do enjoy (cf. Doom 2016 in Vulkan mode). They do require newer APIs to be used (DX12 or Vulkan, used as such and not as "DX11 with croutons") but they can be massively useful.

    Most current games still use older DX11 render paths even in DX12 mode; they mostly enjoy the lower CPU overhead on draw calls and multiple queue job submitting, but almost none make use of async compute or directly manageable VRAM addressing. Future Unreal Engine 4, Serious Engine or id Tech based games should finally start to make use of those technologies and really unleash the power of GCN cards.


    That is very interesting, thanks for that.

    In my mind, HSA hasn't been all the glory and boom they might have hoped to be, but it seems it has paved the way to interesting things in Vulkan and DX12; probably future APIs as well. Again, thanks for pointing that out.

    So, do you have any insight as to what Vega brings to the table that might make it a worthy 1080 / 1080ti adversary?

    Cheers!
    Reply to Yuka
  41. Not gunna lie I fully expect the Vega 10 flagship to beat the Titan XP, if it doesn't then AMD has let us all down pretty badly.
    Reply to neiliohep
  42. if the rumored die size it's true then Vega most likely able to compete head to head with nvidia GP102. but the target is not Titan XP. but Quadro P6000 instead. it depends if AMD can beat it with wide margin or not. if they can then nvidia most likely forced to bring in 600mm2 behemoth to the game. if anything there is one thing that nvidia will not going to give up: single GPU crown. that is what lead to their market share dominance since Fermi generation.
    Reply to renz496
  43. Yuka said:
    [...]

    So, do you have any insight as to what Vega brings to the table that might make it a worthy 1080 / 1080ti adversary?

    Cheers!


    No idea :p I'm happily rocking my Polaris, and since I'm not looking to change it any time soon I've pretty much lost interest in GPUs for the moment.
    Reply to mitch074
  44. mitch074 said:
    No idea :p I'm happily rocking my Polaris
    I just bought an MSI RX470 Gaming X 4GB yesterday for +-USD165! I'm soooOOOooOOOooo excited to start playing BF1 when the card arrives! I haven't played a proper AAA game in a number of years!

    My 6670 has served me well for the past six years and deserves to be heading into a happy retirement, perhaps as an HTPC box.
    Reply to AndrewJacksonZA
  45. renz496 said:
    if the rumored die size it's true then Vega most likely able to compete head to head with nvidia GP102. but the target is not Titan XP. but Quadro P6000 instead. it depends if AMD can beat it with wide margin or not. if they can then nvidia most likely forced to bring in 600mm2 behemoth to the game. if anything there is one thing that nvidia will not going to give up: single GPU crown. that is what lead to their market share dominance since Fermi generation.


    Are they going with a ~600mm2 die like the Fury siblings then? I hope they don't do another Fury though... HBM was good like performance at high resolutions, but the card still wasn't the star we wanted.

    Plus, the bigger the die, the harder it will be to cool and keep the costs down. Following the trend and if they continue to push the 300W power envelope, it's going to be another big room heater. Performance not withstanding, it's going to be a hot card, me thinks.

    mitch074 said:
    No idea :p I'm happily rocking my Polaris, and since I'm not looking to change it any time soon I've pretty much lost interest in GPUs for the moment.


    Same here, but I still want to know if I should give my RX480 to my GF and get myself a Vega or just get her an RX480 and keep mine.

    Still, I can understand where you're coming from. If you come up with any ideas, please let us know anyway. I'd say it's still interesting to speculate and see how the technology unfolds.

    Cheers!
    Reply to Yuka
  46. Yuka said:
    renz496 said:
    if the rumored die size it's true then Vega most likely able to compete head to head with nvidia GP102. but the target is not Titan XP. but Quadro P6000 instead. it depends if AMD can beat it with wide margin or not. if they can then nvidia most likely forced to bring in 600mm2 behemoth to the game. if anything there is one thing that nvidia will not going to give up: single GPU crown. that is what lead to their market share dominance since Fermi generation.


    Are they going with a ~600mm2 die like the Fury siblings then? I hope they don't do another Fury though... HBM was good like performance at high resolutions, but the card still wasn't the star we wanted.

    Plus, the bigger the die, the harder it will be to cool and keep the costs down. Following the trend and if they continue to push the 300W power envelope, it's going to be another big room heater. Performance not withstanding, it's going to be a hot card, me thinks.

    mitch074 said:
    No idea :p I'm happily rocking my Polaris, and since I'm not looking to change it any time soon I've pretty much lost interest in GPUs for the moment.


    Same here, but I still want to know if I should give my RX480 to my GF and get myself a Vega or just get her an RX480 and keep mine.

    Still, I can understand where you're coming from. If you come up with any ideas, please let us know anyway. I'd say it's still interesting to speculate and see how the technology unfolds.

    Cheers!


    in the past VCZ try to predict GP104 die size based on the leak picture. they estimate GP104 to be around 333mm2. real GP104 measured at 314mm2.
    https://videocardz.com/59266/nvidia-pascal-gp104-gpu-pictured-up-close

    using the same method they did with GP104 before they estimate Vega to be in the range of 520mm2-540mm2

    https://videocardz.com/65477/amd-vega-gpu-pictured-features-two-hbm2-stacks
    Reply to renz496
  47. renz496 said:
    in the past VCZ try to predict GP104 die size based on the leak picture. they estimate GP104 to be around 333mm2. real GP104 measured at 314mm2.
    https://videocardz.com/59266/nvidia-pascal-gp104-gpu-pictured-up-close

    using the same method they did with GP104 before they estimate Vega to be in the range of 520mm2-540mm2

    https://videocardz.com/65477/amd-vega-gpu-pictured-features-two-hbm2-stacks


    I can agree to that seeing the pictures put up in the scale they showed. Good they try to make it a tad smaller, but I kind of miss seeing 4 HMB2 blocks there as "cache".

    I wonder if they'll keep GDDR5 or move onto GDDR5X as general memory. Although, it kind of makes little sense... That memory shenanigan sounds like it will add a lot of latency when you have a cache penalty.

    EDIT: According to different sites, the AMD conference at GDC didn't really give any interesting insight to Vega... It will be called "RX Vega". AMD is really flushing the brandings this generation, haha.

    Cheers!
    Reply to Yuka
  48. for cost reason i think AMD will stick with regular GDDR5 for low end cards. not sure about mid range like RX480 though. because GDDR5X higher spec it probably a bit more expensive than older GDDR5. if there is performance advantage to be had they most likely going to use them. but if there is not much then why raise the cost of the product just for the sake of having much "modern" stuff slap on the card?

    for consumer Vega most likely going to top at 8GB VRAM. and this time they really try to do something so the lack of VRAM will not going to hamper the card's performance (that highbandwidth cache thingy). second reason most likely because of cost and and yield.

    interesting thing will be FP16 usage in games. it seems AMD try to encourage for game developer to take advantage of FP16 while nvidia seems doing the opposite (hence nvidia limiting FP16 performance for geforce starting with pascal).
    Reply to renz496
Ask a new question Answer

Read More

Vega Next Generation AMD Graphics