Who has better framerates for this new tech?
Any recommended cards
It's just rendering two images at once and applying a simple distortion to them. There's nothing about it that favours either side.
>>51659361
Nvidia has less microstutter, which is important for this tech. Basically if it plays Fallout 4 on ultra then it should do well with oculus.
>>51659361
if the game has horrible optimization then it doesn't really matter
>>51659394
>microstutter
yeah if you're using an HD7990. all the newer chips have less stutter than Nvidia equivs; unless you go over memory budget on the fury.
AMD easily. Nvidia didn't plan for VR or DX12 with Maxwell. Probably will be fixed with Pascal.
http://wccftech.com/preemption-context-switching-allegedly-best-amd-pretty-good-intel-catastrophic-nvidia/
http://www.extremetech.com/extreme/213519-asynchronous-shading-amd-nvidia-and-dx12-what-we-know-so-far
>I’ve been told by folks at Oculus that the preemption context switching is, and this is prior to the Skylake gen 8 architecture which has better preemption, but the best preemption context switching was with AMD by far. Intel was pretty good, and NVIDIA was possibly catastrophic.
>You have to be super careful because you can miss a frame boundary because the preemption is not particularly well latency. And again, it’s not like this is a bad decision on the part of NVIDIA, it’s you know, that’s just what made sense. Preemption isn’t something that was super important when the chip was designed and the API support was, there wasn’t much bang for your buck. Now, I’m sure they will improve it for Pascal, NVIDIA is full of good, sharp architects and they’ll probably fix it in Pascal.
>>51659835
Nividia probably wasn't expecting VR to actually take off and become consumer grade
>>51659835
the funny thing is that context switching is one of the things axed from Fermi to make Kepler, and was moved to software.
>>51659951
They invested a lot of R&D in it. They definitely thought it would take off, they just didn't think it would be so soon.
>>51659373
It has to render everything twice. So it takes about double the GPU power. I am not sure about memory or anything like that.
>>51660024
>find a video card that can at least maintain 120fps.
boom, you now have 60fps. as for memory, no; it just moves the camera between each frame. some games already have this mode if you turn it on.
>>51660055
What resolution for 120fps?
>>51660136
rift2 is only 960x1080
>building a machine that can handle 4k at high resolution just to strap a fucking screen to your face
>>51660285
and do you realize how fucking shitty that resolution looks when you essentially put it inches from your face and use a magnifying glass to make it seem big?
so uh... do we have VR sex games/ VR fleshlight apparatuses yet?
>>51659394
Look at these nvidialiars!
Amd frametime is way lower than Nvidia.
Which is kills microstutter.
>>51660370
they need an ultra fast screen. the rift1 had such bad ghosting it would cause projectile vomiting. they're using bleeding edge OLED displays, so the resolution is going to suck.
>>51659361
Does anyone know what the game being played is, and or recommend a similar game?
>>51659394
Wroooong. OR programs don't use DX11.
Nvidia actually has slightly worse frametimes with the OR.
>>51661516
So what like DX9 or OpenGl?
Please for the love of god wait until the displays get better. The upcoming consumer release is going to be shit. Wait until the resolution is at least 4k. 8k is the point where pixels wont be visible and 16k is where it will be perfect. A 4k display now is really hard to see the difference in pixels but you're not having it strapped to your faces with lenses that blow up the image so think about it.
Also current gpus suck and we have no idea how good the next gen are really going to be.
>>51661985
Fuck you. I want my waifu simulator right now.
![1449129659176[1].jpg](https://i.imgur.com/bZwwBpQ.jpg "1449129659176[1].jpg")
>>51661453
Can I take advantage of VR if I'm running a 390? What's the effective resolution of these headsets? I haven't really been paying attention since I never thought it'd take off (still weary desu, after the 3D meme died off fairly quickly)
>>51662053
How dare you comparing ponies to 9gag and male feminists desu senpai you were like family to me but not anymore......
>>51661885
Don't know what he's talking about, it does support it, in fact Oculus dropped DX9 support.
>>51661453
So the onahole is connected to the PC
Sweet.
So mean to tell me its like display is like phones with higher resolution?
But this time where it matters.
>>51662084
Can you run games on 2160x1200 at 90hz perfectly? If so, you're good. AMD should have some sort of super sampling so you can test out that resolution and then check FPS in games.
>>51662162
Display manufacturers for phones were the ones sponsoring most of the displays being used in vr headsets. This is why you see Samsung and HTC products. They're just re-using and selling off what they don't need since it was ready made.
>>51659361
isn't the 980ti the recomended Specs for most VR demos on the oculus?
https://www.oculus.com/en-us/oculus-ready-pcs/
I think waifu sim is going to burn me around $2-3k probably with new PC, OR, onahole apparatus, and oppai silicone of different size
>>51662656
No, even the link you posted says it's GTX 970 or AMD 290
Nvidia has done a lot of stuff to improve performance, like supporting VR SLI (one GPU per eye) in programs that are built for it, and multi-res shading to save on processing power for the peripheral vision where it doesn't matter as much.
AMD is also engaged, with LiquidVR, but I don't know of exactly what features that involves. There are rumors that AMD cards are better at async shaders which is important for VR though.
Im all for VR, tried it recently and it was awesome. Whats with all this faggy animu bullshit though?
>>51660349
>4k at high resolution
>>51662887
>async
Here we go again
I dont think Pascal will have that problem tho
>>51662934
Cause it most likely have the least hardware demanding games
>>51659361
What game is that?
>>51663205
Wish I knew
>>51660024
>It has to render everything twice. So it takes about double the GPU power.
I think most games are limited by GPU pixel processing / bandwidth or just the CPU at this point. Drawing two images at half the pixels each will just add work to the code issuing the draw commands, potentially making crappy games even more CPU bound.
You could probably avoid that with geometry viewport multicasting or something, but that could take considerable effort to patch into existing games.
>>51663866
Is 256 the recommended ones?
>>51663866
Owner of a 120Hz display reporting in.
Nine of ten DX9/10/11 games are CPU bound, when it comes to maximum FPS. Got a 290X and an overclocked 5820K. I usually need to lower CPU bound graphics settings (like shadows) to even get in the vicinity of stable 120Hz. But no matter the game, there will always be dips into the 80s or 90s. GPU usage often hovers in the 50s to 70s in all but the newest games.
>>51664207
Oh, even worse than I figured then. Maybe Vulkan / DX12 will improve things in the API end.
Probably doesn't help that many games are made to run at 30 FPS (~33 ms /frame) and accept drops even below that. Who cares if the shitty resource loader sometimes stalls for 20 ms? Low FPS is, like, cinematic and so on.
If you losers didn't know, the game is custom maid 3d 2
>>51664352
That's like the exact thing Vulkan/DX12 helps the most. Reducing the CPU workload in rendering
>>51664412
t-t-thanks senpai
