No DX12 Software is Suitable for Benchmarking *spawn*

[trinibwoy]

I think AoTS was carefully chosen for the dual cheap GPU beating expensive single card narrative, but if there substantial gains anywhere else I would had expected that game to be showcased as well.

I gotta hand it to AMD. They already have people calling the next generation based on a randomly generated benchmark of a game that doesn't look that great and that nobody plays. There's already a general view that AMD is "better at DX12" and we don't even have any actual DX12 games yet.

Well done.

 

[lanek]

Well just because they have Nvidia Gameworks should not exclude them as AMD went out of their way to compare HW, and in a way this is where AMD needs to compete - as Carsten rightly mentioned AoTS has a limited narrative because it already plays to the strengths of AMD GCN architecture.

The benefit of both those games is that they have aspects worth showing, if AMD does compete.
They may had missed a trick not showing those games if the 480 does have great performance, but that would take away a bit of shine from the 2 cheap cards perform as well as a very expensive one narrative, which is what they wanted to push I think.
Not disagreeing just their narrative could had been a bit broader in terms of performance comparison because in theory the improvements to GCN should also reflect in both those games while also giving them an edge in terms of fidelity in one with setting maxed on Pure Hair (RoTR), while the other is known to be a headache in terms of tessellation for both GPU companies, so both good when promoting your product.
On the plus side from an AMD marketing perspective, a lot the talk-disussions since the event is again around DX12 and looking at AoTS, so maybe this is a success.

Cheers

DX12 in TR is completely bugged ( on both AMD and Nvidia, but particulary on AMD and this is said by the developpers themselves), this said, Witcher3 is certainly the most balanced "Nvidia" title to date when it concern performance of both brands, so like you, i will be happy to see it benchmarked. The W3 is certainly one of the best games who have been released thoses last years, and suprisingly run really good .

Again as i have said, i really think that AMD dont want to show performance of thoses gpu's during this show ( and i can tell you, many have the numbers from Macau presentation, but are peoples who dont break an NDA too easely )..

Honestly i camp in my position, that AMD have do an error to dontrelease an higher sku yet, copying the Nvidia method with the 104 ( who was initiate first with Kepler ).

Its like they offer all the 350-500$ market to Nvidia.. im really not happy of this positionnement from them. For me its an error.. Its even a big risk, as even with the " under 350$ market" .. Nvidia could react quickly, releasing first a 960TI and then a 960 based on the GP104.

 

[AlNom]

How aboot Quantum Break.

 

[lanek]

How aboot Quantum Break.

The problem, witth Quantum, is its QB... many think the graphism are a shame, some who like technics, think this game use most of the most advanced graphics technics.. its a controversy games on his technical aspect and even a controvery games as it is released by Microsoft windows 10 store.. This technical and commercial controversy have been enough strong for erase the "first controversy with this game who was the mix between TV series and games.. Seriously, when the game have been released, i was thinking to see a lot of TV shoow here, speaking about this aspect.. and finally, nothing.

Honestly, i have not play QB, all i have seen about it is what i have read on different forums.

 

[Otto Dafe]

How aboot Quantum Break.

Another secret Canadian outs himself with one slip of the tongue.

 

[itsmydamnation]

I gotta hand it to AMD. They already have people calling the next generation based on a randomly generated benchmark of a game that doesn't look that great and that nobody plays. There's already a general view that AMD is "better at DX12" and we don't even have any actual DX12 games yet.

Well done.

I gotta hand it to someone, im just not sure who though. I have never seen so many people of a certain persuasion attack a single product that most of them have never played. The funny thing is went you point out what is actually good about the game they shut up because they have no retort because they have never played it themselves. Its far from a perfect game, but it does some of the hardest things brutally well.

 

[CSI PC]

How aboot Quantum Break.

They could not compare it to Nvidia as it is clearly broken on Nvidia HW, perspective from the average person would be this is even more in favour of AMD than AoTS.
While RoTR is sort of broken on DX12, there is performance there (and in DX11) for AMD to make up.
Which is why it plays a better narrative if the 480 indeed has performance gains over previous GCN.

Cheers

 

[lanek]

They could not compare it to Nvidia as it is clearly broken on Nvidia HW, perspective from the average person would be this is even more in favour of AMD than AoTS.
While RoTR is sort of broken on DX12, there is performance there (and in DX11) for AMD to make up.
Which is why it plays a better narrative if the 480 indeed has performance gains over previous GCN.

Cheers

TR Dx 12 is broken in both hardware. over DX 11 .. what will you expect ? better broken performances ?

Honesty, i dont knw what have happpend to it, but on my personal view, if Nvidia had not get the contract, maybe we will have a proper DX12 release of TR (.. wait to see how run Deus EX based on similar work .. .. honestly... aiee)

 

[trinibwoy]

I gotta hand it to someone, im just not sure who though. I have never seen so many people of a certain persuasion attack a single product that most of them have never played. The funny thing is went you point out what is actually good about the game they shut up because they have no retort because they have never played it themselves. Its far from a perfect game, but it does some of the hardest things brutally well.

The game itself probably has its merits but the real question is whether AoTS is representative of what DX12 brings to the table. It's definitely not the generation defining game that it's being made out to be.

I'm looking forward to seeing what devs can do with all API features - conservative raster, raster ordered views etc in more visually impressive titles.

 

[CSI PC]

TR Dx 12 is broken in both hardware. over DX 11 .. what will you expect ? better broken performances ?

Honesty, i dont knw what have happpend to it, but on my personal view, if Nvidia had not get the contract, maybe we will have a proper DX12 release of TR (.. wait to see how run Deus EX based on similar work .. .. honestly... aiee)

While DX12 is broken you would expect improvement with DX11, and tbh what other recent games could they show with GPUOpen features - which may also makes sense.
They are not forced just to present DX12 games, they can do that with AoTS.
Because the narrative so far is the 480 is great with a niche game that hardly anyone has purchased, and we cannot use any other game because DX12 is busted in all of them
Luckily for AMD so far the general discussions are about the visual difference in AoTS and back to the DX12 and async compute focus.

Ideally they need to show a few games where Polaris improves on its previous generation, and Nvidia had no issues themselves presenting their performance gains over a 980 with the 1080 using RoTR DX12 along with Witcher 3 (where performance between AMD and Nvidia is reasonably close).

Cheers

 

[Silent_Buddha]

While DX12 is broken you would expect improvement with DX11, and tbh what other recent games could they show with GPUOpen features - which may also makes sense.
They are not forced just to present DX12 games, they can do that with AoTS.
Because the narrative so far is the 480 is great with a niche game that hardly anyone has purchased, and we cannot use any other game because DX12 is busted in all of them
Luckily for AMD so far the general discussions are about the visual difference in AoTS and back to the DX12 and async compute focus.

Ideally they need to show a few games where Polaris improves on its previous generation, and Nvidia had no issues themselves presenting their performance gains over a 980 with the 1080 using RoTR DX12 along with Witcher 3 (where performance between AMD and Nvidia is reasonably close).

Cheers

Heck, and even in AOTS, Dx12 is apparently broken for the GTX 1080 as it isn't rendering everything the game is asking it to render.

Regards,
SB

 

[Genotypical]

Heck, and even in AOTS, Dx12 is apparently broken for the GTX 1080 as it isn't rendering everything the game is asking it to render.

Regards,
SB

Is it "broken" or changed? I like my tinfoil hat so I am of the mind that nvidia did it to boost perf where fury x was getting too close. But Oxide are looking into it. I expect nvidia won't fix it till it no longer matters for sure, no matter the reason behind it.

https://www.reddit.com/r/pcgaming/c...ng_the_aots_image_quality_controversy/d3t9ml4

We (Stardock/Oxide) are looking into whether someone has reduced the precision on the new FP16 pipe.

Both AMD and NV have access to the Ashes source base. Once we obtain the new cards, we can evaluate whether someone is trying to modify the game behavior (i.e. reduce visual quality) to get a higher score.

In the meantime, taking side by side screenshots and videos will help ensure GPU developers are dissuaded from trying to boost numbers at the cost of visuals.

Can expect they will let us know whats going on. Very nice communication in the past

 

[Kaotik]

I gotta hand it to AMD. They already have people calling the next generation based on a randomly generated benchmark of a game that doesn't look that great and that nobody plays. There's already a general view that AMD is "better at DX12" and we don't even have any actual DX12 games yet.

Well done.

There are nine actual games that beg to differ on the DX12 front

edit: regarding Ashes, it's a bug between the game version and drivers, apparently same has happened to AMD (can be found on YouTube) and doesn't happen on all NVIDIA cards (not 110% sure on if there's GTX 1080 benches without the glitch, but at least earlier GeForces do it right)

 

[Genotypical]

There are nine actual games that beg to differ on the DX12 front

edit: regarding Ashes, it's a bug between the game version and drivers, apparently same has happened to AMD (can be found on YouTube) and doesn't happen on all NVIDIA cards (not 110% sure on if there's GTX 1080 benches without the glitch, but at least earlier GeForces do it right)

this doesn't seem to have it. Oxide will have to explain whats going on there but looks like game version/driver or some such. The benchmark should prominently overlay the game version and driver version (if possible)

DF has the difference in their video.

sorry about the videos embedded taking up space. I had linked to actual points in the videos but it got formatted.

I'd guess this is something from review/launch drivers. The video without it was uploaded June 1. The DF was middle may and AMD used 368.19 which was before 368.22 was released. Thats something that users can test themselves. If its a consistent difference between 368.19 and 368.22. If so then what is the performance impact.

 

[Freemantle]

Maybe it's just time to stop using AotS as the gold standard DX12 benchmark?

When the GTX 1080 is doing it wrong, but it actually looks better you know comedy will ensure.

It's getting farcical now.

 

[Kaotik]

Maybe it's just time to stop using AotS as the gold standard DX12 benchmark?

When the GTX 1080 is doing it wrong, but it actually looks better you know comedy will ensure.

It's getting farcical now.

It's just illusion of "looking better", because heavier snow on top of terrain hides details from the end image

 

[Freemantle]

It's just illusion of "looking better", because heavier snow on top of terrain hides details from the end image

I rest my case.

 

[Genotypical]

Maybe it's just time to stop using AotS as the gold standard DX12 benchmark?

When the GTX 1080 is doing it wrong, but it actually looks better you know comedy will ensure.

It's getting farcical now.

graphics is always subjective. Can't discard a good benchmark just because it could look better if something was darker, there was less snow etc. Personally I'd prefer it was a darker world with more detail on the maps and less colorful units, but that's neither here nor there regarding how good it is as a benchmark. Especially nice when the company behind it is open.

 

[Silent_Buddha]

I'd guess this is something from review/launch drivers. The video without it was uploaded June 1. The DF was middle may and AMD used 368.19 which was before 368.22 was released. Thats something that users can test themselves. If its a consistent difference between 368.19 and 368.22. If so then what is the performance impact.

Wouldn't be the first time that Nvidia has benefitted from launch drivers mysteriously not rendering everything correctly and thus obtaining a higher score and then weeks later a driver is released that fixes everything with a hit to performance. By which time almost no one is going to bother to re-bench it for a review.

Regards,
SB

 

[CarstenS]

It's just illusion of "looking better", because heavier snow on top of terrain hides details from the end image

I wonder whether or not the terrain is rendered back to front or front to back with this procedural generation of whatever it is they generate procedurally. And yes, that would have serious implications, either way the answer turns out.

Is there an actually correct read up on how AotS does things? Their FAQ is a little shallow from what I saw.