2016 : A Pivotal Year For AMD, Nvidia, PC Gaming And VR

D3K said:
I've only tried cardboard, which I felt was equally marred by low resolution, low FPS, poor FOV, and poor graphics.

I assumed desktop VR kits would overcome all of those, with only res and fov being potential issues. Very happy to hear that the final res could be of high enough quality. The FOV is still of concern, but 1 out of 2 aint bad :)

Have they shown async to be better in real terms, and not just on paper yet? (serious question, I'm out of the loop on async)
Click to expand...

low rendering latency is fundamental in VR, it makes or breaks immersion.
it's not an issue under low workload or DX11, that have limited multi-threading, nvidia is faster than amd.
but when the workload is big or under DX12, Asynchronous compute comes really handy for AMD, and where nvidia need to find a workaround, althoug this doesnt seem to be a simple matter to solve.
so my guess would be that will depands on the game and which API it uses DX11 or 12, we will see basic concepts of VR games on DX11, but my guess every big VR project will pick DX12 or vulcan, they will have much more control closer to the hardware, expecialy that VR games seem to have very specific needs, here is an article that might help.
AMD presentation of async here
more articles about DX12 and async here and here
 
Last edited:
D3K said:
:rolleyes:

I don't doubt any of it on paper, but the AotS hurf blurf has me waiting cynically to see who is gonna perform better in VR.
Click to expand...

personally i think AMD got a clear advantage when it comes to VR, not only because of the async, but also think that the major VR API will be vulcan, not DX12.
after occulus got aquired by facebook, there was a shift into what these mounted displays are going to be used for, facebook made it clear that they are more interested in the social, medical, educational, military aspect of it more than gaming.
on the other hand Valve Headgear is mainly focused on gaming, steamOS will be based on vulcan, the steam platform is a heaven for indie devs, who will bombard it with VR games, and explore new concepts.
the Xbox doesn't have VR headset, while PS4 does and probably the new nintendoNX will too, gives even more incentive for devs to first develope on vulcan, that will make it easier to port between PC & the PS4 because of the unified GCN architecture.
seriously i dont see why a VR developer would need DX12 at all, vulcan works on windows/mac/linux(steamos), and mantle(integrated into vulcan) makes PS4/nintendo port easier and cheaper, and you dont have to wait for the install base to grow, come to think of it, things look gloomy for DX12 :D
i honestly wouldn't be surprise if valve takes a good chunk of market share from microsoft, if VR takes off correctly.
 
Last edited:
Got a 970 recently with my new build so will have to cope with that eagerly waiting on pascal and hoping it delivers...thanks for the read

Where are Nvidia when it comes to Vulcan support
 
Last edited:
D3K said:
I've only tried cardboard, which I felt was equally marred by low resolution, low FPS, poor FOV, and poor graphics.

I assumed desktop VR kits would overcome all of those, with only res and fov being potential issues. Very happy to hear that the final res could be of high enough quality. The FOV is still of concern, but 1 out of 2 aint bad :)

Have they shown async to be better in real terms, and not just on paper yet? (serious question, I'm out of the loop on async)
Click to expand...

Meh, whilst cardboard does a semi decent job, it doesn't do a very good job but it does show what can be done, however in comparison to the DK2, it is **** lol. The FOV is much improved apparently on the CV1, although it isn't an issue for me on the DK2 (I understand it can be for some).

As for the ACE aspect, I can't see this being of importance, when Nvidia can cope just fine with it here and now. The biggest issue is grunt and this is what I have concerns for regarding AMD. They are not the best when it comes to MGPU support (neither are Nvidia) and the best case scenario is a single GPU and this is where Greenland and Pascal come in. Now all rumors suggest a Q1/possibly Q2 launch for Pascal but for Greenland, all rumors suggest Q3/possibly Q4, so AMD could be too late to the party.
 
Suppose if they guess enough they'l get one of their rumors right, always just rumors at the end of the day.

Imagine baby Pascal will be first, much like Maxwell, doubt big Pascal will have much of a lead on high end AMD, but you never know.

Personally not touching VR for a while but hoping anyone that does can get a decent £1200 gpu that can run it.:p:D
 
AlamoX said:
personally i think AMD got a clear advantage when it comes to VR, not only because of the async, but also think that the major VR API will be vulcan, not DX12.
after occulus got aquired by facebook, there was a shift into what these mounted displays are going to be used for, facebook made it clear that they are more interested in the social, medical, educational, military aspect of it more than gaming.
on the other hand Valve Headgear is mainly focused on gaming, steamOS will be based on vulcan, the steam platform is a heaven for indie devs, who will bombard it with VR games, and explore new concepts.
the Xbox doesn't have VR headset, while PS4 does and probably the new nintendoNX will too, gives even more incentive for devs to first develope on vulcan, that will make it easier to port between PC & the PS4 because of the unified GCN architecture.
seriously i dont see why a VR developer would need DX12 at all, vulcan works on windows/mac/linux(steamos), and mantle(integrated into vulcan) makes PS4/nintendo port easier and cheaper, and you dont have to wait for the install base to grow, come to think of it, things look gloomy for DX12 :D
i honestly wouldn't be surprise if valve takes a good chunk of market share from microsoft, if VR takes off correctly.
Click to expand...

there's an awful lot of assumption and very little fact checking going on in here

Vulkan isn't even released yet - games take 2-3 years to develop as a minimum so anyone developing a VR game right now can't be using Vulkan
PS4 doesn't use Vulkan so Sony would have to be pressed in to adding support

people have spent the last 10 years wishing for the death of DX and it hasn't happened yet - it would be great if true but I can't really see it happening in the next 5 years

in fact of all of the currently announced big VR games, all of them have said DX support, none have stated Vulkan support that I can find (if you have that info then please share)

HTC have hinted at a "major breakthrough" that they'll be unveiling at CES next week, interesting to see what that will be
 
While not disagreeing with you entirely, SW BF has unreleased Vulcan/DX12 capable engines, long shot but guessing AOS/\UE4 will have it too, who knows what Steam is doing also.

DX12/Vulcan development is happening.
 
tommybhoy said:
While not disagreeing with you entirely, SW BF has unreleased Vulcan/DX12 capable engines, long shot but guessing AOS/\UE4 will have it too, who knows what Steam is doing also.

DX12/Vulcan development is happening.
Click to expand...

Which one? DX12 yes, but not heard anyone say they are definitely supporting VR with vulkan as of yet

Happy to hear different
 
I would love to see something like Vulkan become a success but with over 200 million people installing Windows 10 as of yesterdays readings, I can't see it shaking the industry. DX12 will be the future but with only AoTS to run thus far, we can't gauge anything substantial from that. AMD do tend to kick their heels but hopefully their finger is on the pulse for DX12/VR.

I am sure there will be some super lovely slides out to show what they will be doing before long :D
 
What's with the AMD slide chuckles all of a sudden, it's not as if Nvidia don't do slides, they let the 970 spec slide for long enough:p

andybird123 said:
Which one? DX12 yes, but not heard anyone say they are definitely supporting VR with vulkan as of yet

Happy to hear different
Click to expand...

Steam support Linux=logical progression to Vulcan, common sense I imagine is enough to assume their VR implementation won't be DX only considering Steams direction, do you think Steam VR won't be supported on Steam OS?
 
Last edited:
Avenged7Fold said:
Excuse me while i visit half the threads in the 'life' forum to post that i am not interested...
Click to expand...

Half life 3 VR confirmed! I knew it!


As for VR im going to be all over the Vive when it is realesed 4k and 8k screens be damned, so many people are being blown away by this already. As for GPU grunt, I would happily turn down the eye candy and stick with a single card solution.
 
KelDG said:
Half life 3 VR confirmed! I knew it!


As for VR im going to be all over the Vive when it is realesed 4k and 8k screens be damned, so many people are being blown away by this already. As for GPU grunt, I would happily turn down the eye candy and stick with a single card solution.
Click to expand...

Absolutely. With a better resolution and 90Hz, you won't need all the AA on etc, so I will also be looking at what I currently have plus a CV1 or the Vive in truth. I am interested to see what the big announcement is about the Vive before committing either way.
 
Gregster said:
Absolutely. With a better resolution and 90Hz, you won't need all the AA on etc, so I will also be looking at what I currently have plus a CV1 or the Vive in truth. I am interested to see what the big announcement is about the Vive before committing either way.
Click to expand...

Apparently they are announcing the Vive breakthrough at CES so should know in a couple of weeks, hope its something interesting and not just PR posturing.

I'm going to go Vive anyway as I am excited about room scale, also the wife gets terrible motion sickness so I am hoping that the 1:1 tracking allows her to at least experience some content.
 
You must log in or register to reply here.
Back
Top Bottom