• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD VEGA confirmed for 2017 H1

Status
Not open for further replies.
AMD need to be delivery at the very least 30% faster than the 1070 across the board never mind the FX or they might as well not bother at this point.

Well no, AMD just need to deliver the right performance for the right price, if its only 10% faster than the 1070 but cheaper its a winner.
 
Yeah but as i said, thats running in Vulkan, which gives AMD a big boost, so we should be knocking about 20+ fps off, for when its not used. :p

I don't understand your point here? Vulkan is the latest API replacing OpenGL so why would or should AMD showcase a GPU running on an API that they have put little time and money into?
Where on the other hand they have been with Vulkan from day one, amd have put a lot of effort into Vulkan/Mantle that Vulkan is very much a part of AMD they should be happy to help showcase this excellent API.

Strange, maybe it best we start seeing latest GPUs running DirectX 9 games again forget 11 or 12 :p

You better than this LoadsaMoney the hate for AMD really comes out in your posts lately.
 
Well no, AMD just need to deliver the right performance for the right price, if its only 10% faster than the 1070 but cheaper its a winner.

Its already a fairly saturated market - and if they are pricing it to pickup the sales from people who've had to stick with a tier or 2 down, etc. then they probably aren't making the money they need to be. AMD needs to be making a ripple in the market at this point to make the money they need going forward not a drop in an already crowded ocean.
 
I don't understand your point here? Vulkan is the latest API replacing OpenGL so why would or should AMD showcase a GPU running on an API that they have put little time and money into?
Where on the other hand they have been with Vulkan from day one, amd have put a lot of effort into Vulkan/Mantle that Vulkan is very much a part of AMD they should be happy to help showcase this excellent API.

Strange, maybe it best we start seeing latest GPUs running DirectX 9 games again forget 11 or 12 :p

You better than this LoadsaMoney the hate for AMD really comes out in your posts lately.

Yea i know AMD have been pushing the GFX technology hard these years. Better APIs that allow games to be developed better. Better technology such as HBM they have helped to develop. They are pushing VR technology with liquid VR. They are pushing the envolope. What has nVidia done in recent years? Lock things down with gameworks. Fair enough the latest APIs have not took off but neither did DirectX 11 it took a fair few years for Devs to get to know this tech and develop games and engines intended for it. just the same for Vulkan and DX12. People's hate towards AMD astounds me sometimes. HBM fail. Mantle Fail... AMD cry at game works and so on. Nvidia gimped even some keplar GPUs with how much tessellation was used making Maxwell look even more better in games that used game works but that was okay and only AMD cry.

These upcomming years are going to be great when the APIs start coming into play and HBM and nextgen memory comes along.
 
Its already a fairly saturated market - and if they are pricing it to pickup the sales from people who've had to stick with a tier or 2 down, etc. then they probably aren't making the money they need to be. AMD needs to be making a ripple in the market at this point to make the money they need going forward not a drop in an already crowded ocean.

This. Releasing a card that is around or barely faster than a 1070/980Ti/Fury X in the middle of this year isn't going to get them a lot of sales unless it is VERY cheap (in comparison). They need to target those that have cards at the level of the 1070/980Ti/Fury X and tempt them to upgrade. Lets face it, anyone who wanted that level of performance and was willing to spend £400+ has most likely already got one of those three cards.
 
Last edited:
I don't understand your point here? Vulkan is the latest API replacing OpenGL so why would or should AMD showcase a GPU running on an API that they have put little time and money into?
Where on the other hand they have been with Vulkan from day one, amd have put a lot of effort into Vulkan/Mantle that Vulkan is very much a part of AMD they should be happy to help showcase this excellent API.

I get what he's saying, Look at it like this, How many Vulcan games are we expecting over the next year? How many OpenGL games have we had in the last 5 years? My guess is not many and it will continue to be a small percentage of games overall.
When Doom released the Vulcan patch allowed AMD cards to get a 20% increase in their performance. So if we remove that advantage from the Vega result we are left with a card that's significantly slower than it is there when compared to a 1080.
I don't mind personally as long as it's priced right, but, if AMD does a Fiji and price it higher than what the performance justifies because they used HBM2 they've gone about it all wrong.
 
Last edited:
This. Releasing a card that is around or barely faster than a 1070/980Ti/Fury X in the middle of this year isn't going to get them a lot of sales unless it is VERY cheap (in comparison). They need to target those that have cards at the level of the 1070/980Ti/Fury X and tempt them to upgrade. Lets face it, anyone who wanted that level of performance and was willing to spend £400+ has most likely already got one of those three cards.

What magic 8-ball did you use to come up with those performance estimates for Vega?
 
I get what he's saying, Look at it like this, How many Vulcan games are we expecting over the next year? How many OpenGL games have we had in the last 5 years? My guess is not many and it will continue to be a small percentage of games overall.
When Doom released the Vulcan patch allowed AMD cards to get a 20% increase in their performance. So if we remove that advantage from the Vega result we are left with a card that's significantly slower than it is there when compared to a 1080.
I don't mind personally as long as it's priced right, but, if it does a Fiji and is priced higher than what the performance justifies because AMD used HBM they've gone about it all wrong.

OpenGL in general has lower performance than DX11 unless you put a lot of effort into it for all vendors. OGl has many problems in general.
 
I get what he's saying, Look at it like this, How many Vulcan games are we expecting over the next year? How many OpenGL games have we had in the last 5 years? My guess is not many and it will continue to be a small percentage of games overall.
When Doom released the Vulcan patch allowed AMD cards to get a 20% increase in their performance. So if we remove that advantage from Vega we are left with a card that's significantly slower than it is there when compared to a 1080.
I don't mind personally as long as it's priced right, but, if it does a Fiji and is priced higher than what the performance justifies because AMD used HBM they've gone about it all wrong.

OpenGL is forgotten about, if you buy Doom you hardly going to run it with OpenGL.
We might has well make this claim that dx11 helped boost performance from dx9 when them GPUs was also coming along, so forget dx11 run them in dx9

Do we all not want the PC industry to move forward? We are the first to complain when we get "console ports"
Showcasing the latest and greatest is what we want to see not what the past holds. Vulkan API shows the hard work that has gone into this from AMD they have a right to showcase Doom.
 
well perhaps, but then combining your CPU with the absolute best GPU money can get won't tell a real story either. It could be an absolute monster, overclocked to hell and back. Doesn't mean you, as a real user with a consumer grade card will get anywhere near that performance. They could have paired it with a 480 for example, that would have given us a better platform to compare. or a 1070 ... you are much more likely to find people with a 1070 than people with a titan.

You combine it with the highest performing card available in retail that people know the results for so they can see that Zen isn't costing any performance.

If you pair it with a 480 and get lets say 40fps, but a Titan X gets 80fps, then all you can say is Zen is fast enough to support a 480, but maybe if a Titan X was in there it would only get 60fps, where Titan X with a 6900k would get 80fps.

The point is to show, a 6900k + Titan X gets 80fps, and so does Zen, so Zen won't hold a Titan X back.

In the same way imagine Vega is 20% faster than a Titan X, so they do the demo and Zen + Vega gets 100fps.... but then an 6900k and Vega might have gotten 120fps. That is why you show your new product with a known product, but in the way that best highlights what it's capable of, that is not holding back the fastest GPU available.
 
What magic 8-ball did you use to come up with those performance estimates for Vega?

I didn't give any performance estimates for Vega. I, like everyone else, don't have a clue how fast it will be. It might be worth following the whole conversation and then reading it in context.
 
Hopefully the uptake of Vulcan will be higher.
As for OGL I can only recall RAGE at the moment,

Well Open GL Next which was going to be 4 was supposed to strip a lot of the old crud and refactor OGl.

Vulcan is essentially a fresh start for OGl with a solid basis.
 
I get what he's saying, Look at it like this, How many Vulcan games are we expecting over the next year? How many OpenGL games have we had in the last 5 years? My guess is not many and it will continue to be a small percentage of games overall.
When Doom released the Vulcan patch allowed AMD cards to get a 20% increase in their performance. So if we remove that advantage from the Vega result we are left with a card that's significantly slower than it is there when compared to a 1080.
I don't mind personally as long as it's priced right, but, if AMD does a Fiji and price it higher than what the performance justifies because they used HBM2 they've gone about it all wrong.

GCN gets a lot of optimisations in Doom under Vulkan.... there is nothing to suggest that Vega has had the same level of optimisation nor that GCN optimisation automatically works and helps with Vega optimisation.

IE the Vega performance in Doom could be unoptimised with a further 20% to be gained with intrinsics for the new architecture.
 
I didn't give any performance estimates for Vega. I, like everyone else, don't have a clue how fast it will be. It might be worth following the whole conversation and then reading it in context.

Quote

"Releasing a card that is around or barely faster than a 1070/980Ti/Fury X in the middle of this year isn't going to get them a lot of sales unless it is VERY cheap (in comparison)."

That is pretty much magic 8-balling it.
 
Quote

"Releasing a card that is around or barely faster than a 1070/980Ti/Fury X in the middle of this year isn't going to get them a lot of sales unless it is VERY cheap (in comparison)."

That is pretty much magic 8-balling it.

Seriously?

Please read the conversation we were having and what I responded to.
 
On early hardware with very early drivers. And yet it was still on average faster than an overclocked 1080. not 1070/980ti/furyx.

Seriously, read the conversation and the context! I was responding to the conversation based on Humbugs hypothetical, which was "Well no, AMD just need to deliver the right performance for the right price, if its only 10% faster than the 1070 but cheaper its a winner. "

In no way was anyone saying that Vega will definitely be around 1070 performance.
 
Status
Not open for further replies.
Back
Top Bottom