WHAT DO WE DO!!!!!! 1070 IS KILL!!!!!!! http://www.guru3d.c

>>55525812
>1% gains from vulkan
what the fuck nvidia

VULKAN WAS SUPPOSE TO BE OURS

FUCK THIS RETURNING THIS SHIT AND GETTING A RX 480

>>55525821
What? No gains
It lost performance vs OGL

>>55525821
>gains
I mean if you look it loses some frames moving to Vulkan. lel novidya, I guess I can get a 480 after all.

Nvidia can't do async compute properly

These cards are gonna be shitty for future games

>>55525812
irresponsible amount of NO GAINS

>>55525812
Nvidia can't into DX12 and Vulkan

>>55525834
>VULKAN WAS SUPPOSE TO BE OURS
Said who? The AMDfags have been chanting this for what feels like centuries now.
>Oh yeah? Well just wait for vulkan and dx12 and stuff! We'll show you!
>They're mantle, just wait and see!
>W-w-we'll get the last laugh ;_;
Alright, you were right you fuckers. Do your victory dance so we can be done with this shit.

File: sour throat.jpg (115KB, 900x900px) Image search: [Google]

115KB, 900x900px

D-drivers will fix it.

>>55525812
>Nvidia cards can't into new technologies

This isn't news. They're designed to the strengths of the most popular API at the time and ignore everything else. That's why they age like milk left out in the sun.

>>55525812
DELETE THIS

File: 480.jpg (26KB, 645x440px) Image search: [Google]

26KB, 645x440px

>>55525862
Some other sites are showing even bigger gains for the 480.

http://www.legitreviews.com/doom-gets-vulkan-implementation_183882

Waiting finally paid off, guys. Nvidia are done.

File: 10d.png (205KB, 482x355px) Image search: [Google]

205KB, 482x355px

>>55525812
>>55525862
>>55526031
>AMD wins again
It's not really shocking. We always win, eventually.

File: image.jpg (121KB, 1024x768px) Image search: [Google]

121KB, 1024x768px

>>55525862
>rx 480 gets 10 more fps than the 1070 at 1080p

Not getting a 250$ aib 480 with 90% the 1070 performance it's like you like getting railed

>>55525862

Now release a single card that competes with the top end and maybe people will actually buy AMD products again and the market can have some competition.

I've waited 5 months but it's finally here, I finally get this play this game on my favorite api.

-said no one ever

>>55526031
Holy shit. Savage.

>>55525821
Vulkan (like DX12) needs Async Compute capable hardware to make better use of the GPU
Also Nvidia's OpenGL drivers are highly optimized

File: 1407414879301.jpg (87KB, 953x960px) Image search: [Google]

87KB, 953x960px

>>55526067
>115-150=10
I don't see how you could have possibly made this mistake.

>>55526067
You're reading the graph wrong. Look at the text before you make an idiot of yourself again.

File: 1467763868487.jpg (48KB, 472x315px) Image search: [Google]

48KB, 472x315px

We warned you guys about the lack of full hardware async compute support on nvidiot cards.

Just played the final two levels of Doom with Vulkan enabled. I have a old Radeon 7850 1gb which could somewhat manage the game at 1080p at low graphic settings.
With Vulkan I managed to bump it up to medium graphics with some added bells and whistles and got between 40-60fps. Thats a MASSIVE improvement.

What you're seeing is the same thing that occurred with DX11 to DX12. Nvidia's cards are already performing where they are supposed to be even under OpenGL. AMD's weren't. A 980 beat a Fury X in OpenGL.

http://www.guru3d.com/index.php?ct=articles&action=file&id=21951

This discussion comes up with every new DX12/Vulkan title. Just because one company is seeing massive gains doesn't mean the other company has something wrong.

1080p on ultra the RX 480 was pulling 98/99 fps average with whatever the default is for D44M according to a lot of benches. Now under Vulkan its a ballhair short of 120fps average.

Pretty impressive gains.

File: image(1060).jpg (99KB, 596x628px) Image search: [Google]

99KB, 596x628px

>>55526031
Getting a mere gain of thirty fps for a 250$ card

>>55526159
bu-but

nvidia dx12_1 support!

what do mean? nvidia lied? AGAIN?

DUDE GRAPHICS CARDS LMAO XXDDDDD

>>55526193
The whole DX12 feature levels was just tech illiterate retards running amok. The levels aren't even "levels" they're categories, and higher isn't necessarily better. All they describe is whether certain features are implemented via hardware, or handled in software.

All of the
>hurr durr Nvidia is better 12_1 true DX12 durrr
was super cringey bullshit not even remotely accurate

>>55526193
I was about to post this...some guy was on here saying that shit like 4 days ago. Said AMD is not up to spec with DX12

Originally Posted by id
Does DOOM support asynchronous compute when running on the Vulkan API?
Asynchronous compute is a feature that provides additional performance gains on top of the baseline id Tech 6 Vulkan feature set.
Currently asynchronous compute is only supported on AMD GPUs and requires DOOM Vulkan supported drivers to run. We are working with NVIDIA to enable asynchronous compute in Vulkan on NVIDIA GPUs. We hope to have an update soon

AMDead
M
D
e
a
d

>>55526222
i know lol its funny how they suddenly changed it back in december and no one said anything lol

>>55526145
NVIDIOTS BTFO AND ON SUICIDE WATCH

>>55526229
>working with NVIDIA to enable asynchronous compute


lol

File: tumblr_o8rga0xYPz1s373hwo1_400.gif (2MB, 320x196px) Image search: [Google]

2MB, 320x196px

This just proves how AMD sucks with drivers.
They really need to rewrite the whole DX11\OGL part of the driver.
It should be easier now since they maintain only GCN and I think Cayman is still on life support but not for long.

>>55526229
Nvidia STILL can't into asynchronous compute? Fuck me. I still have 5 days to return my 1070 for a full refund... seriously considering it right now.

File: rceQJnY.jpg (67KB, 620x388px) Image search: [Google]

67KB, 620x388px

>>55526229
the only way you gonna see async

>>55526288
It really has nothing to do with drivers, you tech-illiterate little butterball. OpenGL and DX11 literally can't take advantage of AMD's Asynchronous Compute Engines, which is where a huge chunk of their performance is hidden. It's not possible to fix that via a driver. These cards were designed from the ground up with DX12 and Vulcan in mind, even though they didn't exist at the time (hence we got Mantle as a stopgap). Nothing AMD can do will ever bring their DX11 and OpenGL performance anywhere near these levels.

>>55526288
>They really need to rewrite the whole DX11\OGL part of the driver.
Fucking why? Those will be irrelevant going forward.

asynchronous compute won't show up on nvidia cards until volta, and volta has been delayed until 2018

>>55526289
What good DX11 game came out this year? What good DX11 game is expected to be coming out?

>>55526289
Prove that you have one.

>>55526338
They can, and in linux, in some games the open source driver actually outperforms the proprietary one by 50% or more.
I would rather they focus on Dx12 and vulkan performance, as well as developing tools and resources for devs to work with.

>>55526338
>hence we got Mantle as a stopgap
To be fair to AMD, it wasn't a stopgap. They played a long, strategic game and somehow won with what was arguably a losing hand. Let's see if that means they can avoid having to fold in the coming rounds.

Sorry, that metaphor was so strained that it was physically painful to type.

Even so, the GTX 1070 is still faster than the RX 480.

I'm glad that AMD got a performance boost.

>tfw waiting for 3rd party Radeon 480 with proper cooling and 8 pin power plug.

File: 1070.png (80KB, 1920x1080px) Image search: [Google]

80KB, 1920x1080px

>>55526366
Sure. What do I do, lads?

>>55525812
Isn't Vulkan an AMD technology ? If so it's not surprising.

>>55526396
GTX 1070 in the UK is almost twice the cost of RX 480. Scan has it on for £400+ and I can get a 380 for £215.

>>55526462
get it back and get a 480 and then be baptised by ed
https://www.youtube.com/watch?v=nloRGBdkOeM

>>55526382
>They can

No, they can't. DX11 and OpenGL cannot use the ACEs, so will never come close to the extra performance gained from utilizing them. AMD's Linux drivers being absolute shit is a completely seperate issue.

>>55526462
Send it back. Seriously.

>>55526462

It has to go back

>>55526492
its a open standard, even google is going to support it on their anpoors

>>55525812
its still around 40% worse then nvidia? okey, I get it the rx480 is better perf / $ because if the pricing was to scale it should have been 50% or so but still, we all know that you get dimishing returns at the high end.

>>55526462
Is your display 144hz? If yes then keep the 1070, if not then get a new display or send the 1070 back and get a 480.

>>55526462
1070 is terrible card anyway. if it actually sold for 380 dollars like promised then it'd be pretty good but as of now either get 480 as budget card or go all the way to the deep end with 1080 and 1440p/4K.

>>55526462
Who gives a fuck about AC? What you gona do? The 1070 is still faster than all AMD cards and Nvidia dominating.

>>55526582
>open standard

Then, Nvidia needs to work on it. If the 480 can reach 80% of the 1070 performance at half the price, they will now have real competition. And now, the 1060 is beaten by the 480.

>>55526462
OP here, I will be honest. I would keep it the RX480 wont be able to preform as well at the 1070 even with DX12, it's just preforming well in its own area. Unless someone can prove me wrong? But the RX is quite a beast for $250 if the conditions are right, that being Vulkan/DX12 and proper development.

>>55526462
>buying a slower card because it support XYZ feature in 0.001% of the games

Good work /g/

>>55526626
>1060
hO shit I forgot about that fucker! Let me laugh louder!! AH HAHAHAHAHAHAHAHAH

>>55526359
>came out this year
Dark Souls 3
>expected to be coming out
Space Hulk: Deathwing

>>55526626
if 480 is reaching the 80% of a gameworks game then there is no hope
period

>>55526626
Nvidia is on the fucking board of directors for Khronos group and is generally more involved. AMD donated the mantle API to Khronos and Microsoft to be implemented into Vulkan and DX12.

>>55526288
>They really need to rewrite the whole DX11\OGL part of the driver.

we need moar gimpworks. full steam ahead. gimp it harder. we outta rewrite the entire from ground up again.

File: 1457506668996.png (357KB, 397x402px) Image search: [Google]

357KB, 397x402px

Vulkan is shit to be honest, you should go with DirectX12™

>>55526162
I don't believe you

>>55526708
>Dark souls 3
>The fourth game in a series of very samey games that people are kind of getting tired of, of which three have been very badly ported to PC
>Space Hulk: Deathwing
>Part of the flood of 40k games being released because Games Workshop decided they really didn't care any more and gave their rights away like a charity hooker
Yeah, not that guy but he said *good* games.

>>55526031
Nice fake, here is some real data http://www.guru3d.com/news-story/new-patch-brings-vulkan-support-to-doom.html

>>55525812
Nvidia's OGL drivers are multi-threaded while AMD's aren't
BIG SURPRISE IT RUNS SAME!

>>55526838
This link we posted by OP, the other guys was just saying he seen other things on another site.

>>55526288
You make it sound like nvidia/amd should rewrite their dx9 drivers because a lot of indie games/mmo still use them. No one's gonna do shit about old tech.

>>55526838
>Two sites have slightly different data
>That must mean that one is a fake
I struggle to follow your logic.
>>55526816
I looked into Space Hulk: Unnecessary colon. It's using the unreal engine so it's going to be DX12/Vulkan too.

>>55526509
i love him, this show is more entertaining than any nvidia announcement

>>55526516
They can still optimize though.

>>55526384
>losing hand
>owns the console market with that hand
Fanboy or not, you have to admit it's going exactly as they planned it.

>>55525812
>1070 IS KILL
not really, but the 1060 is

>>55526708
UE4 supports dx12.

>>55526462
I wouldn't send it back, but If you need better cpu/ram/other shit, then go ahead and put the funds into those instead.

>>55526049
>we
You're fanboy trash unless you actually work for AMD.

File: 1461556229115.jpg (62KB, 450x557px) Image search: [Google]

62KB, 450x557px

>tfw didn't fall for the GTX 1070 meme

Feels good desu. Got me my R9 390X for $250 CAD.

>>55526095
Let's imagine that chip performance scales with chip size at a factor of about 75%
AMD currently has a 230mm chip and the 1070 is a (cut down) 320mm chip, 40% larger
40% at a 75% scaling rate is 30%

With these Vulkan scores, comparing only Guru3D's results, a 40% larger GCN4 chip could get about 116 * 1.3 = 150FPS, shocker, the same score as a 1070
The good news for you is that AMD's Vega chip is going to be a bit larger than 320mm, and considering Polaris' power draw 350mm2 (150 * 1.5 = 225w TDP) is the most likely size.

1080 killer? No, but it could slot in between the 1080 and 1070 in performance. Let's hope all vega chips use at the least GDDR5X, and at the high end HBM2, to control for some of that power use.

>>55526980
>Fanboy or not, you have to admit it's going exactly as they planned it.
I will happily admit to being an AMD fanboy. I'm typing this on a ULV trinity laptop that I use to do steam in home streaming from my a10 7850k desktop. That replaced my athlon x2 245/ HD7750 build. I am a massive fanboy.

But before they got the consoles, there's no denying that they were in a tight spot. Mantle was used to incredible effect, essentially bringing them back from the dead. Now they have a real chance to get some proper market share.

>>55526626
There's only so much you can do with software if you don't have the proper hardware for it. It's like trying to play a ps2 game on a pc. You can emulate it with software, but you're gonna need a way better pc than a ps2 and it still won't run perfect.

>>55527044
#rekt

>>55527104
vega wont even compete with 1080 in the first place...
its about to compete with volta
1080/70 will be DOA by 2017

>>55527149
That is some wishful thinking, I'm an AMD fanboy but come on history is not on their side.

File: AMD-GPU-Roadmap-July-2016-900x534.jpg (96KB, 900x534px) Image search: [Google]

96KB, 900x534px

>>55527104
This slide is from a AMD conference today. They still show HBM2...do you think they can still switch it to GDDR5X for 2017? I'm thinking they might be going all the way wtih HBM?

>>55527160
dont know 480 is pretty much spot on where they said...

and vega having 15 to 18 billion transistors as per amd well it wont even be close to begin with..

>>55526805
but im a girl

>>55525812
What CPU does this bench use?

Vulkan's benefit is from reducing CPU load. If the CPU is badass and never chugged to begin with then of course there is no benefit. Doom is a heavily GPU bound game. I'm running full 60FPS on max settings on a fucking Phenom II X4

>>55526462
It'd be a waste to switch out the 1070 for the RX 480. I have no preference between AMD and Nvidia, I just like whatever's best for my budget. The GTX 1070 is better than the RX 480 and it's silly idea to think otherwise. I love the 480, I'm overjoyed that there's a new great budget GPU out there. But the 1070 simply performs better, it's a higher end card. That's all there is to it. The RX 480 might get more out of Vulkan than the 1070, according to these benchmarks, but the 1070 still outperforms the 480. You do you man. But personally I'd stick with what you have already. Also from my experience /g/ isn't always the best place to go in for advice, to much biased opinions and all that.

>>55526049
>AMD wins again
>performance overall is less than the Nvidia
What?

All this is showing is that AMD is less restrained by their shitty drivers thanks to Vulkan and shows that Nvidia was never restrained by shitty drivers to begin with

>>55527234
Vulkan also benefits from other GPU side things. You cant do async on dx11 or opengl 4.5

>>55527249
>performance overall is less than the Nvidia
For little over half the price, with a significantly smaller chip.

>>55527238
no nvidia card will be better.. because having more fps doesnt mean squat
https://www.youtube.com/watch?v=uxnFv8cnjZA

>>55527169
There've been rumors of a "low end" and a "high end" 2-chip-each Vega set and while I don't believe AMD has the budget for four products in the upper range the lower end most definitely couldn't have HBM2 due to supply and yield issues.

Having said that I really do hope AMD has a >350mm Vega chip and a >450mm Vega chip because it would attack Nvidia's market segmentation very very well

>>55527277
Apparently async is still disabled in Doom for nvidia cards even o nVulkan

>>55527350

Yeah no Shit because Nvidia doesn't have async

>>55525812
All this shows is that Nvidia's OpenGL drivers were already really good.

>>55527100
Used?

File: 1381227569841.png (2MB, 1809x1357px) Image search: [Google]

2MB, 1809x1357px

I'm glad nvidia is finished and bankrupt.

>>55527372
http://forums.guru3d.com/showthread.php?t=407947&page=10

This is false, user reporting performance increase on Nvidia card @Vulkan.

While AMD user also reporting performance loss with Vulkan.

Come one guys, Vulkan and drivers are not even final, wait some time.

>>55527283
Holy fucking shit that 1070 stutters when panning and zooming. If that were happening to me I'd spend days googling how to fix it.

>>55527408
Doesn't change the fact that async is disabled for Nvidia

https://bethesda.net/#en/events/game/doom-vulkan-support-now-live/2016/07/11/156
>Does DOOM support asynchronous compute when running on the Vulkan API? Asynchronous compute is a feature that provides additional performance gains on top of the baseline id Tech 6 Vulkan feature set. Currently asynchronous compute is only supported on AMD GPUs and requires DOOM Vulkan supported drivers to run. We are working with NVIDIA to enable asynchronous compute in Vulkan on NVIDIA GPUs. We hope to have an update soon.

>>55527388
Manufacturer refurbished from NCIX, but it's literally brand new. It's the Sapphire Nitro with the backplate and what not. Also I'm only interested in 1080p and still use a 60Hz monitor, so I'm restricting it's use a lot. All the shit I read about heat/power hungry is basically bullshit but only if I turn vsync off does it actually start getting hot.

Anyway, for me I'm getting basically 70-75% of the 1070 performance and it cost me basically 60% less than the 1070 (at current prices/exchange) so I really don't think it was a bad deal.

>>55527408
>Results are not consistent across the board
Well no fucking shit Sherlock. There are thousands of potential hardware combinations, if not millions at this point. There will be a large degree of variance.

>>55527388
Also, not that long ago they had brand new ones going for ~$360 CAD: http://www.ebay.ca/itm/Sapphire-Nitro-R9-390X-8G-GDDR5-PCI-E-DVI-D-HDMI-TRIPLE-DP-TRI-X-OC-W-BACK-Pl-/282078697152?hash=item41ad3356c0:g:kz8AAOSwjXRXbckL

I just happened to get 1 that wasn't brand new that they had in stock on the actual website really cheap but this was like a 3 day sale or some shit.

>>55527459
So? Whats your strawmann about? I never said anything different you retard. People who use THEIR OWN hardware see performance increase in Vulkan on nvidia cards you spastic.

>>55527443
Ye, now imagine when they enable AC.

http://www.dsogaming.com/news/doom-benefits-greatly-vulkan-vulkan-versus-opengl-performance-comparison/

I would buy an rx480, but their Linux drivers have been consistently crap.

>>55527499
Snot bad.

>>55527503
A lot of people on their own hardware are reporting some performance losses too.

We'll just see what happens when (or if) they enable async

>>55526133
>highly optimized
That's some fluent gamer jargon

>>55527169
what scalability mean for gpus?

>>55527583
multiple chips inside the interposer cutting out the middle man (alx chip)

File: rx480-13b.jpg (157KB, 1920x1080px) Image search: [Google]

157KB, 1920x1080px

>>55527104
You can't extrapolate performance from die area, Polaris 10 was designed specifically to disregard density to maximize yields. The components in the CU are spaced incredibly far apart where in all other GCN designs they're packed in incredibly tight.

AMD could have given the design 64 ROPs so it would out perform the R9 390X, but the design would have had more transistors, been more complex, and more expensive. Pixel throughput is the only limiting factor for the design, its why the clocks are so high for a GCN card, and why power consumption is so high. Its essentially a factory overclocked card to make up for the higher yeilding 32 ROP layout. If it weren't for that it would be a bit closer to the GTX 1070 in performance at markedly lower power, but it was just made to be cheap so they could shit out millions of them.

Designs targeting higher price points won't have the same low density of Polaris 10. Vega will be much more dense, and they're going way over 64 CU.

>>55526626
gtx 1060 = rx 470

>>55527554
Yes, on AMD and Nvidia cards. It only means that this Thread is completly missleading and says nothing about the future of Vulkan.

The only think we still know that nvidia is still dominating the highend market and AMD is unable to catch the gap, they release to late.

DX12 games coming out 2016

Battlefield 1
Deus Ex: Mankind Divided
Star Citizen
Ark: Survival Evolved
Watch Dogs 2
Halo 5: Guardians Edition
Gears of War 4
Forza Horizon 3
Halo Wars 2

>>55527646
>Halo 5: Guardians Edition
source: your ass

>>55527646
dishonored is on id tech 6 as well

>>55527670
http://arstechnica.com/gaming/2016/05/halo-5-online-multiplayer-forge-mode-are-coming-to-windows-10-for-free/

>>55527667
this image makes NO FUCKING SENSE

>>55527646
>DX12 games
Vulkan is the future m8
Get with the format

>>55527709
Sure it does

For Talos, Vulkan is just a wrapper for OpenGL, it's not really a game made or optimized for Vulkan, the devs themselves have said this

>>55527739
>>55527709
And in that regard, it still performs better than their strictly OpenGL mode, that's what they were trying it out to see, however both are an afterthought compared to their DX11 mode

File: 1415375070001.png (42KB, 1070x868px) Image search: [Google]

42KB, 1070x868px

That feel when ordered 1080 and goy-sync monitor today and already getting second thoughts. It's gonna be obsolete in 2 years isn't it?

>>55527646

hey look games I give zero fucks about

>>55527755
Any GPU you could possibly get will be obsolete in 2 years.

You picked the wrong platform if you don't want this feel.

>>55527754
So thats why the 1070 and RX480 are faster in DX11 than 1080?

Makes no sense

>>55525862
That's just because AMD OpenGL drove is terrible

>>55527755
Yes, in 2 Years you wont be able to start games on this hardware, get cucked nvidia fanboy

>>55527667
>Single 480 is 50% slower than 1070
>Dual 1070 is 50% faster than single 1070
>Dual 480 is 10% slower than 1070
>Dual 480 is 95% faster than single 480

WHAT THE FUCK IS GOING ON?

>>55527780
They're not, only the ones in SLI mode, SLI 1080 would be faster

>>55527811
Ah, its CF and SLI, ty.

File: 1466632216961.jpg (130KB, 1200x900px) Image search: [Google]

130KB, 1200x900px

>>55527755
It won't really be obsolete, but with Nvidia's current track record I wouldn't expect to be good $/performance as times goes on. AMD has already proven that they have drivers that optimize their older cards where as Nvidia has literally cucked Kepler owners starting with Maxwell, and it won't surprise me if the same things happens with Pascal once a new architecture comes out.

http://www.overclock.net/t/1529108/are-nvidia-neglecting-kepler-optimization-since-maxwell-release

>>55527852
wohoo conspiracy

>>55527852
nah they dont need to gimp them anymore

they just throw them inti legacy mode

File: launch vs today 290x performance.png (396KB, 828x1041px) Image search: [Google]

396KB, 828x1041px

>>55527755
Yep, just look at where the 780 and 780 Ti is now relative to the 290x. Don't buy Nvidia unless you plan to upgrade every generation.

File: article-2709502-2018CFED00000578-833_634x458.jpg (33KB, 634x458px) Image search: [Google]

33KB, 634x458px

>>55527924
>conspiracy
>popular marketing strategy
Pick one

>>55527852
>Pascal once a new architecture comes out.

Pretty much a given considering how shit it is with Async compute. They can't ignore it with Volta considering that AMD has all the consoles.

>>55527809
Crossfire scaling has been really good for a while now(that is, when it works as intended) while SLi has been lackluster.

>>55528004
Huh, you'd almost think that a new iPhone release coincided with a massive software update that gave the old iPhones absolutely no new features. Funny that.

>>55525812
ITT: shilling and countershilling
everyone, go back to >>>/v/

>>55528099
"Searches for 'iPhone slow' show a spike ahead of launches"

Try again

>>55527283
>https://www.youtube.com/watch?v=uxnFv8cnjZA
Explain it nerds. Is the GTX catching wildly different render times and spitting it out asap?

>>55525812
>every AMD shill thread ever
>muh ASYNC COMPUTE THAT ONE GAME USES
>Doesn't realize that vulkan is about as stable as a bag of schizophrenic dicks

>>55528174
>one game
I don't even play games anymore but I know that there are at least 5 games right now using async compute.

>>55528200
And 99.99% of all games don't.I really don't see the benefit of early adopting APIs just so you can play 5-6 games WOW LOOK AT AMD BEATING NVIDIA XD WE DID IT WE BEAT NVIDIA AT ASYNC COMPUTE and lose every DX11 and OpenGL benchmark.

>>55528145
>Searches for 'iPhone slow' show a spike ahead of launches
No, they don't. If you find the source to go with the graph it states "the days after". What do you want me to try again?

>>55528240
>.I really don't see the benefit of early adopting APIs just so you can play 5-6 games WOW LOOK AT AMD BEATING NVIDIA XD WE DID IT WE BEAT NVIDIA

jokes on you, Nvidia was the first one who adopted Vulkan Doom in their 1080 conference.

>>55528285
>jokes on you
No it isn't, and vulkan support =/= YOU'LL GET MASSIVE FPS BOOST FROM USING VULKAN :DDDDD

>>55526031
OH MY GOD

>>55528240
>still 2 whole years until nvidia's async card
>it only takes a couple months to convert dx11/opengl to dx12/vulkan

>>55528240
>i don't see the benefit being able to play the newest game on the highest settings
why do you even buy a gpu then? To play 3 years old titles? I would undestand this argument on the last gen, but nowadays A LOT of AAA games are being built with dx12 and Vulkan in mind.

Also, Microsoft is really pushing dx12 so it can sell their scorpion thing or whatever is the name next year. AMD will slowly, but surely take over the gpu market on the next 2 or 3 year. Maybe by than Nvidia will find a way to compete again.

>>55528305
>vulkan support =/= YOU'LL GET MASSIVE FPS BOOST FROM USING VULKAN

Yeah, i can see that from OP image.
Poor nvidia, got beaten on their own Vulkan sample game..

>>55528285
Think of it this way: AMD's DX 11 and OpenGL performance is so bad that it makes laughable gains in DX 12/ Vulkan benchmarks.

Nvidia's DX 11 and OpenGL drivers are so well optimized that it doesn't really make a difference if you use some new meme APIs that will crash every 3-4 minutes.

WAIT FOR VOLTA

That will kill the AMDrone

>>55528338
>why do you even buy a GPU then?
Why are you implying that I'm going to buy any current gen die shrink pieces of garbage with GDDR5?

>To play 3 year old titles?
>buy new shiny graphics cards
>It only works with 3 -5 games but has shit performance in other titles you already own

You tried pajeet.

>>55528357
>Nvidia's DX 11 and OpenGL drivers are so well optimized that it doesn't really make a difference

I would agreed with you if Nvidia 1080 doesn't get 25% performance boost with Vulkan.

>>55528389
Not him but are you honestly saying that you only intend to play games released before your GPU was made and not touch any that come after?

>>55528427
>25% performance boost

see

>>55525812

Is this good or bad. You GPU fags change everyday

>>55528447_(you)

>>55528442
>not him

No, I'm saying that I won't buy / switch graphics cards to one that only offers support to meme APIs that aren't widely used yet and have mediocre performance everywhere else.

>>55528447

Are you blind? I don't see GTX 1080 in OP pic.
Current Doom bench showing GTX 1080 run at 160 fps on openGL, nvidia conference showed 200fps with Vulkan.

No performance gain, right?

>>55528263
You're right, surely ou must have a delusional response to the fact that the problem doesn't exist for Android phones too.

Is apple genuinely worse at designing software?

>>55528500
>No performance gain right?
What settings you stupid fuck?
Just the raw FPS doesn't mean donkey dick.

>>55528535

https://www.youtube.com/watch?v=jN0BTk6ZewA

see for your fucking self.

>>55528500
>nvidia conference showed 200fps with Vulkan

On what settings? don't know?

Then stfu and look at the real benchmarks already out.A 1070 is a cut down 1080, it's not magically going to not scale just because you want to make believe.

>>55528563

I'm just gonna laugh at you.

>>55526462

>70 GlobalistBucks in tax

Holy shit

>>55526873
Amd can't into multi-thread so they've done a lower level api so the game devs can do the job for them.

>>55528511
Wait what? You think I'm defending Apple because I'm stating that it happens after a new phone is launched and not before? Holy shit that's dumb.

No, I honestly believe that Apple is fucking over their iPhones with software "updates" but you know, they kind of have to be available for people to get a hold of them.

>>55528551
>both touch 200fps
>Wow Vulkan is gr8 meme
Please die

>>55528574
Yeah, it's painful.

Thanks for all the advice, lads. Gonna think it over.

>>55525812
I don't get it, does running it on Vulkan make the game look nicer?

File: VULKANGAINZ.png (795KB, 850x473px) Image search: [Google]

795KB, 850x473px

>>55528551
>vulkan means 25% fps gain

File: Untitled.jpg (279KB, 1911x873px) Image search: [Google]

279KB, 1911x873px

>>55528595

>It touch 200 while he standing doing nothing staring at a wall
>WOW

>>55527104
All vega chips will have hbm2 up to 16gb. At 4k it will kill the 1080 for sure

>>55528675
see

>>55528663

''''''no''''''

>>55528663

I love how you find a part where both of them showing 200fps on some occasion while ignored most of the time the openGL only got average of 150fps framerate.

I'm here only to discuss the new API and you guys really need to be so autistic about it.

>>55528686

See. >>55528675 and >>55528700

File: openglfaster.png (898KB, 864x480px) Image search: [Google]

898KB, 864x480px

>>55528700
>>55528717

File: openglfasterer.png (891KB, 858x475px) Image search: [Google]

891KB, 858x475px

>>55528717
>>55528700

>>55528728

And? the whole video the game ran on 150fps average on openGL.
What the hell are you gonna prove here beside that it sometime both of then can reach same number of FPS in small occasion?

Are you posting just to spite me or actually want to discuss something?

>>55528757
That vulkan is a meme

>>55528788

Okay.

>>55528677
>At 4k it will kill the 1080 for sure

vega 10 will be equivalent performance to the 1080. AMD isn't even making a 600mm2 chip this generation so they won't put out any competitors to the titan p and 1080ti.

gg

File: 1463617354030.png (254KB, 399x385px) Image search: [Google]

254KB, 399x385px

>>55525812
Why do you type like that?
Are you brain damaged?

File: Fury X1.png (14KB, 733x151px) Image search: [Google]

14KB, 733x151px

>>55525812
Sure am glad I didn't fall for the Pascal Meme

File: DOOM_Vulkan.png (4MB, 1296x1460px) Image search: [Google]

4MB, 1296x1460px

Switching from OpenGL to Vulkan, Ultra Settings with TSSAA@1440P with a stock Fury.

>>55527646
The only non shitty looking game on that list is Forza Horizon 3.

>>55529102
Acquire some decent taste

>>55529020
Nice. Although I think that the Fury's 4gb will become a limiting factor very soon.

>>55529121
Sorry, I forgot Halo Wars 2 and maybe Star Citizen. Everything else is definitely shit looking.

>>55529137
>I think that the Fury's 4gb will become a limiting factor very soon.
Only for Ultra Mega Resolution Textures, some games don't even show a difference with various textures despite it eating up more VRAM

>>55529140
The Mainline Deus Ex Titles are always a decent playthough

>>55529165
I really didn't like HR so I have absolutely no interest in MD. Also le mechanical apartheid meme storyline is putting me off.

I'm not going to miss much by skipping out on it.

>>55529012
I'm also thinking about this. Like deeply!

>>55529140
>Star Citizen.
How long has that game been delayed? My vaporware sense is tingling.

File: ef9ad3d781bb280d4bd0732b1d72fe09.jpg (7KB, 250x242px) Image search: [Google]

7KB, 250x242px

DELET THIS

JUST FUCKING DELETE THIS

>tfw bought a 1080 because I truly thought this was the end for AMD

File: Tech Power Up.png (77KB, 507x875px) Image search: [Google]

77KB, 507x875px

>>55529177
>>55529012
It's also aging like a fine wine, much like most modern AMD cards, it's gained 10fps in GTAV through the years

top chart was from Fury X launch

>>55529190
it's been in development for about as long as Cyberpunk 2077, so maybe vaporware.

>>55529216
Why would you buy a card that can't properly push 4k 60hz or 1440p 144hz?

>>55529236
Because I play at 1440p/60hz you donut

File: 1467480257828.jpg (240KB, 1356x916px) Image search: [Google]

240KB, 1356x916px

>>55529236
>properly

>>55529248
but a Fury X or 1070 could do 1440p 60hz high/ultra no problem

File: 1437431525356.jpg (44KB, 634x500px) Image search: [Google]

44KB, 634x500px

>>55525812
>novidiafags think their pascal is an all-new uarch
>realize it's just maxwell 3.0
>realize it's just tesla 5.2
>realize they're buying deprecated tech
mfw pic related

>>55527719
Maybe if DX12 wasn't around.

The thing is the xb1 already supported DX12, so if you are going to make a game for the xb1 and the PC why not have both use DX12 ?
It won't make sense for them to go through the hassle of dx12 AND vulkan
Not to say we will never see a game support dx12 and vulkan, just when resources are tight developers will go with dx12 by default since they have no choice on the xb1

>>55527249
>Amd didn't trample a card that costs almost triple
Shock!

>>55529259
No.

File: GTAV.png (24KB, 500x530px) Image search: [Google]

24KB, 500x530px

>>55529318
Yes?

GTAV, nvidia optimized title Fury X hits 66fps average

File: witcher3_2560_1440.png (25KB, 500x530px) Image search: [Google]

25KB, 500x530px

>>55529318
>>55529438
The Witcher 3, Fury X hits 50fps, 1070 60fps

File: bf4_2560_1440.png (23KB, 500x530px) Image search: [Google]

23KB, 500x530px

>>55529318
>>55529438
>>55529452
Battlefield 4, over 60fps for both GPUs

So what the fuck are you talking about? The 1080 is a bad purchase because it's overkill for 1440p, but not 100% there for high refresh rate 1440p or 4k 60hz gaming

>>55529137
Different anon here. This is why I'm switching to dual 480's.

>>55529482
multi gpu support just isn't 100% there, just keep that in mind.

>>55529482
Eh, crossfire is risky, you have to wait for special drivers all the time and sometimes they don't come at all. You basically can't play games in crossfire immediately after they come out. It would maybe be smarter to wait for Vega.

>>55529482
>>55529564
I know it's a risk but I'm betting on Multi-GPU support getting much better with DX12. It could easily backfire and if it does I'm planning on doing a whole new build in a couple of years anyway. Alternatively I could buy a 2nd Fury but the fuckers price is literally never going to drop so why bother.

>ze germans showing an 8350 paired with a nano going from 36fps to 50fps at 4k

Honestly as impressive as all this gpu talk is i'd like to see some cpu focused benchmarks in the coming days. Hell for all DX12 games released so far cpu scaling is far more interesting than seeing gpus gain performance.

Hey Nvidiots. Turn vsync off in Doom settings and in your Nvidia game profile.

You might see some gains.

File: xBgSksM.jpg (18KB, 340x260px) Image search: [Google]

18KB, 340x260px

Quickly, shoop this so DX12 is replaced with vulkan.

File: DX11Vulkan.png (16KB, 524x175px) Image search: [Google]

16KB, 524x175px

>>55529783

All that is happening is AMD hardware is being brought up to the levels it would have had if DX11 had been a level playing field. It's not about AMD beating Nvidia hardware. It's about AMD's architecture now being enabled to run at it's optimized settings via DX12 and Vulkan. Nvidia is not losing anything since they were always already optimized for DX11 and OGL. The reason you are not seeing significant gains (and sometimes losses) in DX12 and Vulkan on Nvidia's side is just that their architecture is not optimized towards it. If they did that then they would lose any gains in DX11 and OGL and be on the same level of performance as AMD.

People see it as black and white when it is not.

All that DX12 and Vulkan are doing is leveling the playing field for once. The hardware is pretty much so close on each team anyhow it's not worth fighting over.

So a benchmark shows that AyyyMD's implementation of opengl sucks and they are trying to spin this as an huge win?

>>55525812
This is patethic. THIS IS PATETHIC.

For real. It sound like for half of the people, it was alll a lie.

>>55525821
I fucking know right?

>>55525980
I wish they could. Won't happen.. FUCK

This just show how and will always be ft the budget build and nvidia will always be for the enthusiast

>>55527408
>http://forums.guru3d.com/showthread.php?t=407947&page=10
Linux.........

>>55525812
While I have no doubt that Nvidia's cards will have smaller performance gains in Vulkan than Nvidia due to lack of async, is the 1070 being bottlenecked by the CPU or is it really just that bad?

DOOM definitely uses Async in Vulkan as it uses it on PS4 and Xbox One.

>>55531438
Async allows the GPU to do stuff faster / in parallel and shouldn't affect CPU load

>>55532275
>Async ... shouldn't affect CPU load
That's exactly what it's for.

>>55532334
No, that's what Vulkan/DX12 in general is for

>>55527044
>>55527249
Buttmad Novidiots can't handle the truth

>>55526621
It is today, but did you missed the part where they added Vulkan support to Doom and suddenly the RX480 is within 10% of the 1070? That's pretty fucking good for a card that costs HALF of a 1070.

>>55527349
>There've been rumors of a "low end" and a "high end" 2-chip-each Vega set and while I don't believe AMD has the budget for four products in the upper range the lower end most definitely couldn't have HBM2 due to supply and yield issues.

Not that impossible. Low bin for 490 with GDDR5 and a new Fury with HBM2, high bin for 490x with GDDR5X and a new Fury X with HBM2.

Nvidiots on suicide watch! https://www.youtube.com/watch?v=GlzKPBIjZPo

>>55532493


It would be more likely to see

low bin 490 with 8gb GDDR5x, high bin with 8gb HBM2,
low bin rage with 8gb HBM2, high bin with 16gb hbm2

Why are 99% of devs such fags? There are massive improvements in both dx12 and vulkan yet they keep making dx11 games. It's not like they'll lose customers, vulkan is available on all platforms.

Is Nvidia just drowning them in money? There's literally no reason to make a dx11/opengl game now other than to shill nvidia cards.

nvidia user with sandy bridge cpu seem to be benefit from vulkan.

http://www.neogaf.com/forum/showthread.php?t=1245006&page=7

>>55532528
pajeet, pls

give it a break with the spam of your designated shitting forums

>>55532556
Because change costs money in the real world. In the delusional world of /g/ you can just flip a switch and change everything without any downside.

>>55526462
You better send that shit back m8

ok I thought that nvidia had a better architecture than AMD in this generation because all benchmarks say nvidia has better performance per watt but it now looks like AMD architercture is superior but only for vulkan and directx 12 and that original AMD 480 slice of 2X480 only using 50% is actually true when using vulkan+async
look in this video vulkan+async with 50% GPU usge vs OpenGL 4.5 using 100%

>>55525862
>b-but the 480 is worse than a 970!!
>JUST GET THE GTX 970 so you can play The Way It's Meant To Be Played!!

File: .png (2MB, 1920x1080px) Image search: [Google]

2MB, 1920x1080px

>>55528338
>AMD will slowly, but surely take over the gpu market on the next 2 or 3 year

>>55527667

Vulkan for The Talos Principle wasn't even good enough to be considered a half arsed job, they simply created a wrapper and pushed it out calling it "experimental" literally a day after the API's release.

Shit implementation which looked bad on the API all so they could cash in on calling their game the first Vulkan title, wouldn't recommend the game for any benchmarking purposes..

>>55525812
>nVidia wants Vulkan because it has more performance than AMD
>AMD has better performance

NoVidya shills will be in so much denial.

https://www.youtube.com/watch?v=jN0BTk6ZewA

vulkan test with 5960x 4.4 ghz and gtx 1080.

>>55532775
Haven't AMDrones been saying that for the last five or six years?

File: 1467276904862.jpg (100KB, 734x811px) Image search: [Google]

100KB, 734x811px

>>55525812
>vulkan performs worse than opengl

Opengl forever

>>55532841
Yeah AMDfags are always talking about how they'll take over in a few years
And Nvidiafags are always talking about how their latest card is going to kill AMD.

>>55532739
>just get the gtx 970
Everyone who wants one already does, if you're a poorfag and want a burnt motherboard buy a reference 480. Otherwise shell out an extra $50 for a 1060 instead. Nvidia doesn't make anymore 970's shoo poo in loos

>>55532866
You mean kill them more than how dead they already are?
>10% of the GPU market
kek

Oh shit I forgot about Vulkan
Get Rekt nvidiots

>>55532895
>burnt motherboard
And this right here is why I'll never fucking buy the 480.
The worst part is that AMD won't even admit they're killing motherboards.

>>55532941
>AMD needs to admit my assumptions.
worst part is that you believe people on the internet.

>>55532941
http://www.tomshardware.com/news/zotac-nvidia-gtx-1060-amp,32227.html

This kills the AMD

>>55532962
>they believe people on the internet
Yeah nobody should believe your lies.
>AMD comes up with a fix for a non-issue through drivers that doesn't actually fix the overcurrent problems just lessens it to burn your motherboard more slowly now.

>>55525812
my r9 380 already runs the game with everything on ultra. How does this affect me?

>>55532962
>assumptions
Seriously? All those burned PCI-E slots and you seriously going to say it's just a 'conspiracy'.
It's not rocket science, just overclock your 480 and wait a few days for your slot to die.

>>55525862
Vulkan can go suck a proprietary dick.

I like OpenGL, naysayers be damned.

>>55532962
They even released a fix for it without admitting it caused the issue.
So they just decided to gimp out their card to fix something that wasn't even happening?

>>55532989
>people had trouble running rx480 on AM2 and LGA755 platforms
>AMD fixed the issue

>Oooh, its because it burns the motherboards.
yup, sound logic you have there.

>>55532993
It doesn't but AMD OpenGL performance is so shit that it makes their Vulkan performance look like godmode.

>>55533020
>people had trouble running rx 480

http://lmgtfy.com/?q=amd+overcurrent

Yes the whole internet is lying about AMD just for shits and giggles

>>55533002
>All those burned PCI-E Slots
yea, i remember seeing that corroded to shit Asrock motherboard blowing couple of tracks, nothing to do with corrosion and everything to do with rx480.

>>55533043
nobody is saying they don't overdraw, you fucking faggot.

>>55533047
http://www.guru3d.com/news-story/amd-gives-statement-on-the-pci-express-overcurrent-problems.html
Jesus Christ you fucking shill.
THEY EVEN FUCKING ADMITTED THEMSELVES.
You can stop doing damage control

>>55533056
Then why are you pretending it isn't burning PCI-E slots when it's literally overcurrenting at stock?
If you OC a reference 480 it's just a matter of time before you burn out your motherboard

>>55533062
>THEY EVEN FUCKING ADMITTED THEMSELVES.
You need to have pair of glasses or a new brain, they never admitted that any of motherboards burn.

worst part of this is, you can't proof that it does.

>>55533062
It's the new narrative dumbdumb, AMD is lying about themselves because they're secretly controlled by the Nvidia illuminati

>>55533076
>If you OC
Then you get disclaimer saying that you are on your own and of something happens to your PC its your fault.

Good job on proving my point so quickly.

>>55533082
>they never admitted that any of the motherboards burn
Yeah they just admitted to holding the matches and the gasoline at the time your motherboard burns.

File: geforce_gtx_480_egg_cooking.jpg (97KB, 400x287px) Image search: [Google]

97KB, 400x287px

>>55533047
lol, sure

>>55533062
Don't put your effort into it. You will be up all night arguing with them. Nvidia will always be the motherboard wrecker.

>>55533082
If you overcurrent a pci slot you will eventually burn it.
Do I need to explain that you put your hand in fire you burn it as well? Do I have to prove to you that if you exceed the voltage on the motherboard you will burn it?

>>55533107
>It's okay if AMD destroys your mobo once you attempt to OC.
but it reaches 1600MHZ!!!!!

>>55533111
no, they just said "Fine, we will strap your child with dog leash when we go on a walk"

because you all are pussies.

>>55533132
>if you're don't want your motherboard to burn after a few days/weeks of use you're a pussy

>>55533120
What will you burn?
How will it show?

Basically if you have no proof even theoretical then you are full of shit.

>>55533132
Yeah what a pussy who doesn't want their hardware to be destroyed. Better crossfire 4 of them so you can have a deluxe housefire.

>>55526359
uhhhh literally all of them? dx12 is the new one

Hey /g/uys, any news on the 1060?

>>55526462
Keep as it is still better than everything AMD has to offer you dumb fucking Amdrone. You fucking Amdrones are delusional as fuck holy shit. Having 78% share in discrete graphics Nvidia will make sure the games will run shit on AMD on pc. Hope AMD goes bankrupt soon because amdrones are most cancerous and delusional idiots I have ever seen. Keep grasping at straws you dumb motherfuckers

>>55533153
>Basically if you have no proof even theoretical then you are full of shit
Just google overcurrent.
next you'll ask me to prove to you that water is wet

>>55533207
Proof that motherboards have 75w hardware limit and that it's not just conservative specification limit.

if you cannot do that, you are guessing that 75w is max and talking out of your ass.

File: fury sale.png (134KB, 710x597px) Image search: [Google]

134KB, 710x597px

Hey guys, if anybody is interested, Newegg has the Sapphire Tri-X OC R9 Fury for $350. Pic related, just ordered one.

Nvidia will always be the housefire King

https://www.youtube.com/watch?v=sRo-1VFMcbc

>>55527646
Star citizen dx12. No.

Faggot Roberts wants more luck money

>>55532739
>get the latest amd turd so you can have +- 10 fps in one game while terrible performance in all the other mainstream games that all favor competition!!!!

Hmm I'm convinced. What could go wrong

>>55533184

Actually almost all AAA games are made for consoles in mind first, the PC being an afterthought.

AMD provides cpu/gpu to Xbox, Playstation, Nintendo.

Do the math here. AMD pretty much has all bases covered. All games will be developed for their hardware in the future.

I've been waiting for this day for so long. Not for AMD to beat Nvidia, but to finally play a game I bought on release with my 260x. Fuck you, Bethesda for making me wait so long for something promised "shortly after release."

Doesn't matter now though, been planning to buy the Nitro 480 for a while now.

>>55533281
Are you fucking stupid? The kind of optimization that goes into consoles has absolutely fucking nothing to do with how the port is gonna come out.

>>55533281
>their hardware
nobody cares, problem here is that their hardware uses exactly the same instruction sets and their competition.

So, now that the dust has settled.
Is AMD the way to go?
Or will it fry my mobo?

>>55533341
Ask again in about a week when we see how the 1060 does.

>>55533341
if you beliebe novidya shills, it fries your mobo
if you beliebe AMDevil shills, it doesn't
if you are rational then you would wait for after market cards that are full sized and are better constructed.

>>55533341
CPU wise? no, intel is still better.

RX480? avoid the XFX version.
It wont fry your MOBO but will fuck it up if you are using a really low quality motherboard (ASRock, prebuilt, chink, etc).

>>55533281

Let's add in AMD providing hardware to Apple PCs. Polaris is tailored for Apple. High yields -> high margins. iMac users don't care about +10-20% performance gains. They want "enough" and VR.

It's checkmate already really. NVIDIA will shortly abandon Pascal to go for full dx12/Vulcan supported hardware. Think Kepler was gimped? You haven't seen ANYTHING yet.

>>55530338
Talk all you like but from 2017 onward Vulcan and Dx12 are the standard.

Nvidia-age like rotten cheese
Amd-age like fine wine

>>55529020
>vulkan
>nice even cpu load spread
>opengl
>one core is at 0%
kek

>>55528943
>AMD isn't even making a 600mm2 chip this generation so they won't put out any competitors to the titan p and 1080ti.
So what? The less than 1% of people who buy those things? How would a 1080/Vega equivalent not be enough for almost everyone?

>>55533011
Vulkan is by the same guys as OpenGL son

>>55532911
real number is 25-30% though since last year just on 390 rep., now it might be about 35% if those 480 actually sell that well

So, anyone has info about Vulkan on phones?

Sounds like will do miracles on them.

>>55533315

Next gen is different. What you need to understand is Moore 's law breaking down in high end but still viable low-mid range, and the drive towards multi core/gpu. This is what ALL parties want, even NVIDIA , but they lost it when AMD won contracts from Sony, Msoft, Nintendo and monopolized console harware market. This can NOT be fixed for NVIDIA.

AMDs problems have been because they've lacked software support for their hardware. Will not be problem in the future.

Expect cheaper, better cards in the future. Probably made by AMD...if NVIDIA even bothers with mainstream markets anymore.

>>55533485
>nvidia also lost intel for igpu to amd

and nobody who worked with nvidia want to work with them again, we seriously need third player in both cpu and gpu markets

>>55533355
1060 cuntfirmed DOA when Vulcan doom bench released. I'm going with under 480 perf.

>>55533538
>Vulcan doom
>doom
Just one game? Good job carrying on the tradition of AMDrone cherry-picking.

>>55533485
oh, also, nobody believed that we will see avalanche of ports to pc this gen and it happened => nvidia pushed a lot of money for gimpworks to be in every game to stay "the king"

now with half-gen consoles announced which are GCN things will only escalate, nvidia would have to make some changes to architecture

File: AZRUy5L4mIGWPzYF54kpIkAqVKkaErCg1RT0HqdClTk.png (37KB, 519x397px) Image search: [Google]

37KB, 519x397px

>>55533576
It show what can be done when devs use new APIs and not DX11 crap So yea I'm also going the 480 route. The Refrence has its problems but over all the card it good.

>>55533603
>It show what can be done when devs use new APIs
And are devs going to use new APIs or continue to stick with what they are familiar with, and cheaper?

You guys are like the linux crowd always believing next year will be the year of linux.

>>55533619
API are free, son, all that matters is work hours which stay the same

>>55533640
>API are free
Paying devs to learn a new API is not, "son."

>>55533619
He never mentioned linux.

>>55533659
Might I suggest you go look up what an "analogy" is.

>>55533649
>Paying devs to learn a new API is not
you either stay relevant in the field or go do something else
we'd still use dx8 by your logic

>>55533687
>we'd still use dx8 by your logic
No. We'd use whatever is the current version of DirectX, in this case we are transitioning into DX12. But certainly not Vulkan. The only engine that uses it is Idsoft's. Everyone else is using their own engine or Unity or UE which runs on Vulkan? Oh wai-

File: 1467860102737.jpg (22KB, 480x461px) Image search: [Google]

22KB, 480x461px

>Nvidia BTFO on DX12/Vulkan
>Paid shills can't come up with valid arguments so pull the copypasta PCIe destroying motherboards card.

>>55533619
>Consoles use DX12 and Vulkan
>Win10 uses DX12
>Hurr nobody is gonna switch
>Most games coming in the fall are DX12

This damage control is desperate.

>>55533720
>>Consoles use DX12 and Vulkan
What console game uses Vulkan?

>>55533714
>tfw don't even have an RX480 but my card is going to stay relevant for years to come.

>>55533225
The Fury X is only $50 more, you should cancel and get that

>>55533711
Source supports Vulkan. Also, pretty much every modern game supports OpenGl, why do you think it will not be replaced with Vulkan?

>>55533279
It's just the start. That list is only going to get longer and your butthurt stronger. Rise of the Tomb Raider and Doom are only the beginning. More triple A titles are already lined up and if devs see they can get such a boost on both platforms and it's easy to implement then it means more sales on all platforms for them.

>>55533731
Not him but none that im aware.

Nintendo and sony might in a future.

>>55533619
>devs going to use new APIs or continue to stick with what they are familiar with

Well fuck I guess if you were running a game development company you would still be making DX9 games huh?

>>55533619
did devs jump DX9 or did they stick with DX8?

devs will jump to what is going to sell their games, this is exactly why DX10 was commercial failure.

>>55533744
None? Who in this thread is desperately attempting damage control? The guy arguing for Vulkan on consoles?

>supports
Let's just ignore that no one is actually using that potential.

>>55533222
It's the PCI-Sig spec you fucktard, i.e. the MAX CURRENT ALLOWED is 5.5A, it's pulling more than 6A which is over the maximum allowed. We should call the police and fbi on them.

>>55525812
install gentoo

>>55533776
>>55533759
You two certainly came to this thread late. That's argument was already noted as stupid and addressed.

>>55533784
>We should call the police and fbi on them.
ah ha ha ha kid go to sleep No one cares about this shit.

>>55533784
Good that rx480 is not PCI-SIG certified and doesn't need to adhere to their standards, innit?

>>55533431
That doesn't matter.

It's hard as fuck to find just which license(s) Vulkan falls under, and it uses parts of AMD's proprietary Mantle API, further complicating things. I can't find a straight answer. I don''t know if it's free, open, or freeware.

Alternatively, OpenGL is much easier. It fucking says it right here: https://www.sgi.com/tech/opengl/?/license.html

>>55533797
>was noted as stupid
by you, you also never actually addressed the argument.

your argument is "hurr durr expensive to learn new things"

>>55527613
This is what we are thinking is possible.

>>55527583
usually, it means they can just add more to the chip and it will work.

>>55533742
I don't want the AIO.

>>55533797
It's ok I understand your opinion but look at the post we are on right now. DX12/Vulkan making a really good performance on AMD cards. You can ignore it but its happening.

>>55533828
>by you
Yes. Because one would have relearn the entire API switching from DX8 to DX9 or DX9 to DX 10 or . . . and is in any way analogous to jumping from DX11 to Vulkan.

Do you have any more stupid arguments to present?

>>55533776
No, Devs will jump to whatever becomes the de-facto standard while requiring the minimum amount of effort to do.
Big publishers won't because the current consoles are basically stuck with their APIs for at least another 2 years.
So it's basically Microsoft first party shovelware and indie garbage for you if you like vulkan/dx 12.
>Valve
>In charge of not being steam jew and making games again

MSI Gaming X 1070 has gone up from the £419 I have it pre-ordered at for Thursday to £489. I'm thinking of scalping it on Ebay and grabbing a 480 as it's more than enough for 1080@60.

>>55533841
>You can ignore it but its happening.
Devs are switching from DX to Vulkan because of one game, which was originally in OpenGL?
Is that really your argument?

>>55533849
>We are making new engine from scratch
>We can't use Vulkan that will work on all the platforms but Xbone because we need to learn new stuff

no, it's good and sound argument and not retarded as fuck at all.

>>55533813
Then it shouldn't use the PCI-E slot and build their own custom mobo/slot