AMD "Vega" Demoed in Sonoma, California

[LightningJR]

This performance is very close to what I get on my overclocked 1070. Was just playing with the Doom demo today too bad Vulcan lowers FPS at 4k for me

 

[OneMoar]

so even running doom THE BEST Title in recent memory for AMD cards ON brand new hardware
it still only manages to best the 1070/1080 by 11 and 8 fps respectively and uses god knows how much power doing so

this does not bode well just more pr non-sense from amd

 

[Camm]

so even running doom THE BEST Title in recent memory for AMD cards ON brand new hardware
it still only manages to best the 1070/1080 by 6 and 4 fps respectively and uses god knows how much power doing so

this does not bode well just more pr non-sense from amd

We don't know if this was small vega or big vega. If this is small vega going up against a 1070, performance is more than acceptable.

 

[OneMoar]

We don't know if this was small vega or big vega. If this is small vega going up against a 1070, performance is more than acceptable.

yes we do its big vega (not that small vega is even a thing yet)
its using 8GB HMB2 amd would not put that on anything but a top end part the 25TFP/s number also points to that

 

[TheGuruStud]

yes we do its big vega (not that small vega is even a thing yet)
its using HMB2 amd would not put that on anything but a top end part the 25TFP/s number also points to that

So it performs the same as a Fury X? That would be the end of AMD video cards that cost more than the 480.

 

[OneMoar]

So it performs the same as a Fury X? That would be the end of AMD video cards that cost more than the 480.

the fury x gets about 45 or so fps in doom Vulcan at 4k if memory serves
the 1070 was getting about 50

the big thing is power consumption unless amd has worked some serious magic is probably going to be well north of 225W

 

[Camm]

yes we do its big vega (not that small vega is even a thing yet)
its using 8GB HMB2 amd would not put that on anything but a top end part the 25TFP/s number also points to that

I'm not trying to be rude, but I haven't seen any indication that AMD has mentioned what arch this is, outside it being 'vega', as both rumours on 10 and 11 have indicated HBM stacks. Shrug. Be better if AMD just releases the fucking thing.

 

[TheGuruStud]

the fury x gets about 45 or so fps in doom Vulcan at 4k if memory serves

Depends on where you're at. Over 60 in some areas.

 

[OneMoar]

doom is a horrible measure of actual gpu power for one reason its well optimized the game engine scales very well and does a bunch of stuff in the background regardless of detail settings to maximize performance/image quality and because it can use aync for those extra operations amd hardware sees a pretty big benfit,and that makes it inaccurate as a graphics benchmark



the only metrics you should be caring about are the ROP's/TMU's shader count and FP math are mostly second banana when it comes to running games

but hey it looks really good on paper if you have no idea what any of that means

 

[refillable]

The PC is taped (covering air holes) and thus has a very bad airflow. This is what I call "Koduri Cube".

Anyway, this preliminary result is as expected. There's no small or big Vega yet. The only thing ever spotted was a 4096 shader Vega, which is likely to be this very card.

so even running doom THE BEST Title in recent memory for AMD cards ON brand new hardware
it still only manages to best the 1070/1080 by 11 and 8 fps respectively and uses god knows how much power doing so

this does not bode well just more pr non-sense from amd

Wow, you must have fallen from the hype train before posting this.

 

[ViperXTR]

speculations, rumours and reveal.
Good times.

 

[ShurikN]

the fury x gets about 45 or so fps in doom Vulcan at 4k if memory serves
the 1070 was getting about 50

the big thing is power consumption unless amd has worked some serious magic is probably going to be well north of 225W

1080 was getting 50, 1070 was at 40 avg.
https://www.techpowerup.com/reviews/MSI/GTX_1070_Quick_Silver_OC/13.html

 

[jigar2speed]

This performance is very close to what I get on my overclocked 1070. Was just playing with the Doom demo today too bad Vulcan lowers FPS at 4k for me

How is this possible, even GTX 1080 makes max 52FPS @ 4K resolution -

 

[cdawall]

so even running doom THE BEST Title in recent memory for AMD cards ON brand new hardware
it still only manages to best the 1070/1080 by 11 and 8 fps respectively and uses god knows how much power doing so

this does not bode well just more pr non-sense from amd

On a prerelease driver, with a cpu fan glued to it. This isn't exactly a final product. God knows what we can expect for clock speeds and final tdp numbers.

I expect 1070 performance in dx11 and 1080+ for vulkan/dx12 for small Vega and a Titan competitor for the big Vega.

 

[ShurikN]

On a prerelease driver, with a cpu fan glued to it. This isn't exactly a final product. God knows what we can expect for clock speeds and final tdp numbers.

I expect 1070 performance in dx11 and 1080+ for vulkan/dx12 for small Vega and a Titan competitor for the big Vega.

And maybe that Polaris 10xt2 will slot a bit above 1060 to fill that gap

 

[cdawall]

And maybe that Polaris 10xt2 will slot a bit above 1060 to fill that gap

Better bins on the 480 will take care of that on their own. Honestly AMD shouldn't bother with a name change. Just pump out some faster bins and call it good.

 

[Dippyskoodlez]

How is this possible, even GTX 1080 makes max 52FPS @ 4K resolution -

The trick is the usual AMD hype machine: Either leave out significant, pertinent information or leave out significant pertinent information.

For anyone that actually has doom, it doesn't have a benchmark tool.

This means review site numbers are not comparable directly, and AMD can cater this 'sample' to whatever they want.

Because here's my 980Ti @ 4k ultra:

http://i.imgur.com/diFvE7i.jpg

I don't believe any 'astonishing' numbers until it's in at least w1zz's hands.

Bonus Vulkan screenie:

http://i.imgur.com/xMEkWxe.jpg

 

[ShurikN]

Better bins on the 480 will take care of that on their own. Honestly AMD shouldn't bother with a name change. Just pump out some faster bins and call it good.

Wont that create a lot of confusion. You'd be left with the luck of the draw. Some sort of name change would have to exist, or at least something to differentiate the two.

 

[cdawall]

Wont that create a lot of confusion. You'd be left with the luck of the draw. Some sort of name change would have to exist, or at least something to differentiate the two.

Why? I would be more pissed if they added more junk to the lineup when they are the same card.

 

[Camm]

Why? I would be more pissed if they added more junk to the lineup when they are the same card.

Vega 11 is supposed to be in the same space as Polaris 10, so I'm starting to expect a number jump from 4xx to 5xx (or something else new.

 

[HD64G]

MI25 is not a dual gpu

Raja Koduri answered that on twitter

That's good to know.

 

[HD64G]

so even running doom THE BEST Title in recent memory for AMD cards ON brand new hardware
it still only manages to best the 1070/1080 by 11 and 8 fps respectively and uses god knows how much power doing so

this does not bode well just more pr non-sense from amd

Too hasty in your assumptions there. Do you know the clocks of this demo GPU? The optimisations coming? The power consumption? The price of the final product? Probably not.

 

[HD64G]

Why? I would be more pissed if they added more junk to the lineup when they are the same card.

You simply name the polaris 10 refresh GPUs RX475, RX485 and sell them in almost the same price and leave the RX470, RX480 in a discount to quickly clean up the inventories. nVidia would fall in despair imho being unable to respond without losing money.

Any moderator that could merge my 3 posts, please do it and sorry for not doing that correctly from the beginning.

 

[m1dg3t]

the numbers only indicate the design order, so polaris 10 was designed before polaris 11, and vega 11 was designed after vega 10, its not any indication of their segment or performance

That was true for Polaris, not sure about Vega.

Source?

In fact, Anandtech had an article a while back saying 4096 cores is about as much as the architecture can take, Fiji or otherwise. So, if this Vega 10 is touted at 12.5 T/flops at a certain speed (1525Mhz), it works out at 4096 shaders.



http://videocardz.com/64706/amd-vega-doom-4kultra-gaming-performance-demo-possible-specs

I don't see Vega 11 being bigger than 4096 shaders.

Source? My shitty memory! LoL Sorry, but I aint diggin' through the internets looking for a PR slide from sometime during the last 6 or so months. So take it as you will. I also recall I saw a slide indicating the possibility of a ~6k shader core but can't recall if it was referring to Vega or Navi? :/

 

[the54thvoid]

While everyone is talking speeds, even Fury X does 60+ fps.. We wont know shit until this thing is released. Until then even the genuine leak benchmarks are unusable.

Hexus do use Vulkan API for their testing.