Guru3D Thermal Paste Roundup - Round 2 (2021)
ClockTuner v2.1 for Ryzen (CTR) Guide
TRENDnet 5-port Unmanaged 2.5G Switch review
Powercolor RX 6700 XT Hellhound Spectral White review
Resident Evil Village: PC graphics performance benchmark review
Enermax LIQMAX III aRGB 240mm review
PCIe Resizable BAR Performance AMD and NVIDIA benchmarks
Guru3D Rig of the Month - April 2021
Intel Core i7 11700KF processor review
Acer Predator Apollo and Predator Talos DRAM Review
AMD VEGA (Radeon RX 490) Launch Imminent ?
As you all know, AMD is yet to release the enthusiast SKUs for their graphics card line. The Radeon RX 490 series will be based on a GPU called VEGA 10. We already know that AMD Radeon Technologies Group (RTG) chief Raja Koduri tweeted that development of "Vega10" had just crossed a milestone, although it's a long way to go before you can see it".
AMD was supposed to launch their high-end class VEGA (codename) product in early 2017, the latest rumor was that the launch was being moved forward towards October. However today on Facebook Chris Hook, Sr. Director, Global Marketing and Public Relations, Radeon Technologies Group at AMD posted a photo with the comment "This is the Vega launch venue. Shh, don't tell the press". Considering half the worlds tech press is on his facebook, that obviously was a rather subtle joke slash hint.
Not exactly what you'd expect for a product launch - but for gaming series graphics card it certainly looks like a scenery from a Batllefield'ish game?
If this isn't a complete pun (and I do not expect it to be) then AMD is preparing a launch event for VEGA. Let's say you start planning such an even two months up-front, then indeed you can expect launch time to be October ...
So After Polaris comes Vega, Vega is the brightest star in the constellation Lyra, the fifth brightest star in the night sky and the second brightest star in the northern celestial hemisphere, after Arcturus. On earlier roadmaps VEGA was tagged as a HBM2 product.
On the venue ... I better get my Tetanus shots.
ddelamare
Senior Member
Posts: 221
Joined: 2013-11-18
Senior Member
Posts: 221
Joined: 2013-11-18
#5321296 Posted on: 08/15/2016 01:17 PM
That is exactly why i moved to nVidia this time around.
Was waiting and waiting and waiting on AMD for something to replace by 2x OC R9 280x's, and a little bit more, and gave up.
I'll see whats on the market from AMD this time next year, maybe they have a better product then.
I'd find it strange if AMD itself didn't know if they can release a product already this october or only months later in '17. It makes little sense they would have been telling 2017 if they are releasing it already in a couple of months. Some people who didn't want to wait no doubt went for 1070/1080, but they might have waited if it was only october.
That is exactly why i moved to nVidia this time around.
Was waiting and waiting and waiting on AMD for something to replace by 2x OC R9 280x's, and a little bit more, and gave up.
I'll see whats on the market from AMD this time next year, maybe they have a better product then.
TerminatorUK
Junior Member
Posts: 1
Joined: 2009-12-17
Junior Member
Posts: 1
Joined: 2009-12-17
#5321298 Posted on: 08/15/2016 01:20 PM
Not really the best sign when people are deathly afraid to hype cards. I mean I hope the cards are honestly good and comparable to Nvidia and at a little cheaper pricepoint.
But without Gameworks, DSR, and the rest of the Nvidia stable there is no reason to switch to AMD sadly. Not with games like Witcher 3 using these features. I use DSR on so many indies as well the SGSSAA I cannot live without in older games.
Still, AMD can win some fans back with this some decent cards here.
Now if they release a card that is even a little more powerful than Nvidia that would be even better for competition. I don't expect that at all but would be nice.
I have to disagree slightly.
AMD already have their own version of DSR called VSR which effectively does the same thing.
AMD cards are showing strong results (currently at least) in DX12 and Vulkan API titles - I realise these are few and far between but the current GCN architecture does seem better suited for asynchronous compute tasks (with Ashes of the Singularity, Hitman and Doom providing some examples) and therefore potentially has better 'future proofing' if games adopt this approach more and more.
I'd personally take that over Gameworks features, PhysX (which tends to be framerate sapping, niche and often poorly implemented) and additional features such as Ansel which I would never use myself.
I think people have been burnt before with AMD products (and not just GPUs - look at Bulldozer on the CPU side for example) not meeting expectations and rather see why the finished product is actually capable of before jumping to conclusions.
I too fell victim to the RX480 hype train that it'd be GTX 980 / Fury levels of performance for $200 even though AMD never stated it would be at that level.
AMD did exactly what the set out to do (and said the high-end was coming later this time around).
The only slight disappoint is that I expected it to use less power than it does (again, we need to remember it is still a big improvement over the previous generation) and the sheer lack of overclocking headroom, even on the 8-pin custom RX480 solutions was frustrating for me.
Still, one can hope that it was a limitation of the Polaris architecture that held back potential overclocking headroom rather than an inherent trait of the Global Foundaries / Samsung 14nm process...
Time will tell anyway - I'm more excited that they are quoting 'Vega' rather than a dual-GPU Polaris board which I would (will?) steer clear of personally - I've lost my faith in multiple GPU solutions at the moment (both SLI and Crossfire).
Not really the best sign when people are deathly afraid to hype cards. I mean I hope the cards are honestly good and comparable to Nvidia and at a little cheaper pricepoint.
But without Gameworks, DSR, and the rest of the Nvidia stable there is no reason to switch to AMD sadly. Not with games like Witcher 3 using these features. I use DSR on so many indies as well the SGSSAA I cannot live without in older games.
Still, AMD can win some fans back with this some decent cards here.
Now if they release a card that is even a little more powerful than Nvidia that would be even better for competition. I don't expect that at all but would be nice.
I have to disagree slightly.
AMD already have their own version of DSR called VSR which effectively does the same thing.
AMD cards are showing strong results (currently at least) in DX12 and Vulkan API titles - I realise these are few and far between but the current GCN architecture does seem better suited for asynchronous compute tasks (with Ashes of the Singularity, Hitman and Doom providing some examples) and therefore potentially has better 'future proofing' if games adopt this approach more and more.
I'd personally take that over Gameworks features, PhysX (which tends to be framerate sapping, niche and often poorly implemented) and additional features such as Ansel which I would never use myself.
I think people have been burnt before with AMD products (and not just GPUs - look at Bulldozer on the CPU side for example) not meeting expectations and rather see why the finished product is actually capable of before jumping to conclusions.
I too fell victim to the RX480 hype train that it'd be GTX 980 / Fury levels of performance for $200 even though AMD never stated it would be at that level.
AMD did exactly what the set out to do (and said the high-end was coming later this time around).
The only slight disappoint is that I expected it to use less power than it does (again, we need to remember it is still a big improvement over the previous generation) and the sheer lack of overclocking headroom, even on the 8-pin custom RX480 solutions was frustrating for me.
Still, one can hope that it was a limitation of the Polaris architecture that held back potential overclocking headroom rather than an inherent trait of the Global Foundaries / Samsung 14nm process...
Time will tell anyway - I'm more excited that they are quoting 'Vega' rather than a dual-GPU Polaris board which I would (will?) steer clear of personally - I've lost my faith in multiple GPU solutions at the moment (both SLI and Crossfire).
Denial
Senior Member
Posts: 13323
Joined: 2004-05-16
Senior Member
Posts: 13323
Joined: 2004-05-16
#5321301 Posted on: 08/15/2016 01:28 PM
There are two Vegas the bigger Vega 10 and the smaller Vega 11. The rumored Vega launching this year likely to be the smaller Vega 11 which competes with GTX 1080/1070. The big Vega 10 in reply to GP102 Titan X Pascal probably release on Q1 2017 at the earliest.
Vega 11 is big Vega - Vega 10 is small Vega. The numbers swap for Vega.
I have to disagree slightly.
AMD already have their own version of DSR called VSR which effectively does the same thing.
AMD cards are showing strong results (currently at least) in DX12 and Vulkan API titles - I realise these are few and far between but the current GCN architecture does seem better suited for asynchronous compute tasks (with Ashes of the Singularity, Hitman and Doom providing some examples) and therefore potentially has better 'future proofing' if games adopt this approach more and more.
I'd personally take that over Gameworks features, PhysX (which tends to be framerate sapping, niche and often poorly implemented) and additional features such as Ansel which I would never use myself.
I think people have been burnt before with AMD products (and not just GPUs - look at Bulldozer on the CPU side for example) not meeting expectations and rather see why the finished product is actually capable of before jumping to conclusions.
I too fell victim to the RX480 hype train that it'd be GTX 980 / Fury levels of performance for $200 even though AMD never stated it would be at that level.
AMD did exactly what the set out to do (and said the high-end was coming later this time around).
The only slight disappoint is that I expected it to use less power than it does (again, we need to remember it is still a big improvement over the previous generation) and the sheer lack of overclocking headroom, even on the 8-pin custom RX480 solutions was frustrating for me.
Still, one can hope that it was a limitation of the Polaris architecture that held back potential overclocking headroom rather than an inherent trait of the Global Foundaries / Samsung 14nm process...
Time will tell anyway - I'm more excited that they are quoting 'Vega' rather than a dual-GPU Polaris board which I would (will?) steer clear of personally - I've lost my faith in multiple GPU solutions at the moment (both SLI and Crossfire).
Max frequency is more related to path optimization then the process. The problem is the best optimization is done by manually, by physical process/design engineers. It's been mentioned before by Anandtech that AMD simply can't afford to hire the people necessary to do that type of work in their CPU division, I'm sure the case is the same with their GPUs.
There are two Vegas the bigger Vega 10 and the smaller Vega 11. The rumored Vega launching this year likely to be the smaller Vega 11 which competes with GTX 1080/1070. The big Vega 10 in reply to GP102 Titan X Pascal probably release on Q1 2017 at the earliest.
Vega 11 is big Vega - Vega 10 is small Vega. The numbers swap for Vega.
I have to disagree slightly.
AMD already have their own version of DSR called VSR which effectively does the same thing.
AMD cards are showing strong results (currently at least) in DX12 and Vulkan API titles - I realise these are few and far between but the current GCN architecture does seem better suited for asynchronous compute tasks (with Ashes of the Singularity, Hitman and Doom providing some examples) and therefore potentially has better 'future proofing' if games adopt this approach more and more.
I'd personally take that over Gameworks features, PhysX (which tends to be framerate sapping, niche and often poorly implemented) and additional features such as Ansel which I would never use myself.
I think people have been burnt before with AMD products (and not just GPUs - look at Bulldozer on the CPU side for example) not meeting expectations and rather see why the finished product is actually capable of before jumping to conclusions.
I too fell victim to the RX480 hype train that it'd be GTX 980 / Fury levels of performance for $200 even though AMD never stated it would be at that level.
AMD did exactly what the set out to do (and said the high-end was coming later this time around).
The only slight disappoint is that I expected it to use less power than it does (again, we need to remember it is still a big improvement over the previous generation) and the sheer lack of overclocking headroom, even on the 8-pin custom RX480 solutions was frustrating for me.
Still, one can hope that it was a limitation of the Polaris architecture that held back potential overclocking headroom rather than an inherent trait of the Global Foundaries / Samsung 14nm process...
Time will tell anyway - I'm more excited that they are quoting 'Vega' rather than a dual-GPU Polaris board which I would (will?) steer clear of personally - I've lost my faith in multiple GPU solutions at the moment (both SLI and Crossfire).
Max frequency is more related to path optimization then the process. The problem is the best optimization is done by manually, by physical process/design engineers. It's been mentioned before by Anandtech that AMD simply can't afford to hire the people necessary to do that type of work in their CPU division, I'm sure the case is the same with their GPUs.
Ryu5uzaku
Senior Member
Posts: 7041
Joined: 2006-09-24
Senior Member
Posts: 7041
Joined: 2006-09-24
#5321308 Posted on: 08/15/2016 01:44 PM
Vega 11 is big Vega - Vega 10 is small Vega. The numbers swap for Vega.
Max frequency is more related to path optimization then the process. The problem is the best optimization is done by manually, by physical process/design engineers. It's been mentioned before by Anandtech that AMD simply can't afford to hire the people necessary to do that type of work in their CPU division, I'm sure the case is the same with their GPUs.
I hope they can afford in really near future. But that all comes down to how Vega then Navi and Zen does in the end.
Vega 11 is big Vega - Vega 10 is small Vega. The numbers swap for Vega.
Max frequency is more related to path optimization then the process. The problem is the best optimization is done by manually, by physical process/design engineers. It's been mentioned before by Anandtech that AMD simply can't afford to hire the people necessary to do that type of work in their CPU division, I'm sure the case is the same with their GPUs.
I hope they can afford in really near future. But that all comes down to how Vega then Navi and Zen does in the end.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 7041
Joined: 2006-09-24
It won't be hbm card for sure.