>Just watch guys, Nvidia cant even do DX12, AMDs gonna take the GPU market back!
>>55031688
Why does this thread exist? It seems rather silly to me. Are you mentally challenged?
>>55031688
we know pascal has better async support though. it's not perfect, but its better nevertheless.
>>55031688
>Rise of Tomb Raider
>DX12
Pick one.
Negative improvements on DX12? Literally how the fuck is this possible?
AMD 4 life
>>55031764
rottr is broken shit. dx12 has been broken on that since it was implemented which was practically a half job.
I have the 960 and it works fine if your building a 1000 dollar pc a 70 or 60 will work fine. I run arms 3 ultra settings with the 60 and 60fps but I'm getting the 1080 soon it looks great
that game's dx12 uses a wrapper to call dx11 code. it's only using dx12 in the vaguest sense of the word
cuck
>>55031764
I hate to be that guy but that test was done at 1080p. Show the benchmarks at 4k and we get an entirely different story.
This is coming from someone who's got a GTX 1080 on the way.
>>55031688
Oh look, another /v/-style CALLOFDOOODYY progamer GPU thread
>>55031722
>Better
When you turn on ASync on the 1080 its loses a frame.
>>55031885
I couldn't find any other doing dx11 vs dx12 benchmarks on rottr
Closest I could get is this resolution. Even still the dx12 performance is worse.
As a note, this game dev partnered with Nvidia. This should explain the discrepancies on why it runs shit on DX12 and why it runs dogshit on AMD.
>>55031990
>loses a frame
it lost 0.1 of a frame according to your own picture. 50.2 > 50.1
that's margin of error shit and just goes to show how efficient the new pascal cards run since it has no performance gain due to the card already being fully utilized compared to the, lets say, fury x which is very under utilized because of shitty drivers.
not being a fanboy here, just stating what i see.
>>55031833
did you mean Arma3? Arma3 isn't really a graphics intensive but a CPU intensive game. upgraded from i7 950 to i7 6700k and got about 4x the fps (GTX 970)
>>55031833
Literally the worst possible choice of game to prove your point mate. Game depends entirely on your CPU.
Witcher 3 is the current gold standard for graphics bench marking.
>>55032088
Somehow its always AMD drivers issue when Games that nvidia partnered run dogshit on AMD cards.
But its not the same driver when the games that are either neutral or AMD partnered runs like a champ.
Its the same driver, if you factor out what changed, you'll see the obvious outlier (nvidia partnered games).
>>55032088
>not being a fanboy here, just stating what i see.
Yes you are being a fanboy here... It's well known that non-async render pipelines are never 100% efficient or even close to.
The reason why Nvidia cards suck at async is because unlike AMD cards, they don't have hardware asyc and there isn't any clear info on when they will get it. AMD went for async well before there were any API's to actually support it apart from their own Mantle API and suffered because of the wasted die space. However now it's paying dividends.
>>55032002
Probably wouldn't matter anyway, as I just realised you were comparing it to the Fury X and not the 295x2. That would be a much fairer assesment.
>>55032135
that's just how it goes i'm afraid. companies do this all the time to each other, it's just business. i personally don't give a shit about any of that and i buy which ever works best over all titles so i got a 980 on the cheap a year ago. i could have bought amd to give them my support but like i said, i don't really care about muh ethics or any of that.
Dx12
8 games
EIGHT FUCKING GAMES
Not to mention half of them "are not real dx12" as people have allready pointed out.
>>55031764
>>55031841
>>55032182
You're not getting the point of the benchmark. Its not nvidia 980ti vs amd fury X. Its AMD cards DX11 vs 12. Nvidia card DX11 vs 12.
Notice the HUGE discrepancies? On Nvidia's DX11 vs DX12, the difference is minimal but still a small net negative. On AMD's DX11 vs DX12, the difference is a huge net negative.
>>55032175
please, tell me where i was being a fanboy. i didn't explicitly side with one company or another, i simply made an observation.
you sound mad.
>>55032248
That game had a poor implementation of dx12 tho.
>>55032248
I think that test may be on older drivers. I've been looking up shitloads of benchmarks recently to help decide between the 1080 and 1070 and the AMD cards usually tend to beat the 980 and 980 TI in higher resolution, and especially on DX 12.
The only thing I think that may be tipping the scale in their favour in the benchmarks you've posted is if they had Nvidia Hairworks on.
>>55032240
>8 games
Try one game, which is why AMD shilling only uses one benchmark.
>>55032240
That's now, however with the rise of Windows 10 usage, and the fact that more cards will be able to implement it better, DX 12 will become the standard soon.
>>55032213
Here's the most recent AMD partnership
This has only been an Nvidia thing far as I remember. I recall the first Tomb Raider that introduced the TreeFX. AMD later opensourced that. Nvidiafags still cry "foul" at new AMD tech exclusivity that they made opensource after launch. They are blind to the gameworks being completely closed on AMD side of things
>>55032375
Enjoy your first party Xbone ports.
>>55032375
dx12 won't be the standard for a few years yet.
>>55032467
Honestly wouldn't mind. I already have a PS4 so it'd just mean that I'd have access to all exclusives.
>>55032399
>tressfx
They open sourced it because literally nobody would use it
