Sony VR Headset/Project Morpheus/PlayStation VR

I expect both native resolution and high frame rate to be the priorities over core graphics. OR/Vive both have higher base frame rates and native resoutions than Morpheus so most (but not all) of the additional performance the Occulus baseline spec (GTX 970) offers will be absorbed by that. Hopefully core graphics will scale for more powerful setups though and the law of diminishing returns may apply less in this case due to the more basic core graphics being used as baseline.
 
I expect both native resolution and high frame rate to be the priorities over core graphics. OR/Vive both have higher base frame rates and native resoutions than Morpheus so most (but not all) of the additional performance the Occulus baseline spec (GTX 970) offers will be absorbed by that. .

That's only if all PS VR games use a 60fps baseline. Morpheus is actually capable of displaying 120hz and from what I remember from an old digitalfoundry article; the demo with the robots actually ran at that native framerate.
 
I think every PSVR game is 1080p, I ever encountered any talk that some games are sub-native.

the demo with the robots actually ran at that native framerate.
There is another native 120fps game, space shooter that was quickly made by a very small indie team.

Anyone has info about framerate of Super Hyper Cube and Hyper Void?
 
Last edited:
Oh! Nice! I hope they don't change the pixel clock and just add longer blanking time. So it would keeps the screen latency at 8ms instead of raising it to 11ms.

the reprojection technique works best when using exact multiples of frames (like 60Hz to 120Hz), so 90Hz native display is actually preferable to 90Hz » 120Hz reprojection, according to the Sony rep.
Yep... because of judder.
 
Last edited:
If PC VR scene starts developing like mad, with better displays arriving each year, PS4 will still remain highly viable because higher PC system requirements will keep the games from aiming at higher visuals. With better displays [4K@90 or more], games will remain simpler to render, and therefore, PS4 will have a great shot to render everything because devs will aim "only" at 1080p60.

Epic talked about their process of porting that street firefight VR demo from PC to PS4. They integrated Morpheus SDK into UE4, optimised some shaders, and then had to change nothing asset wise. Game immediately worked great on PS4.
25:50 mark

I think that's short sited. The bulk of Vive / Rift sales will happen the second half of 2016 and there will be large jump in pc component power. New gpus , skylake and crabbylake , zen and so on . Then there will be improvements in api's .

So while its first year it might not be a problem , I suspect it to change very quickly. Not to mention as better headsets appear on the sony will be stuck with the same gear. I don't think a yearly playstation vr headset would work .
 
That's only if all PS VR games use a 60fps baseline. Morpheus is actually capable of displaying 120hz and from what I remember from an old digitalfoundry article; the demo with the robots actually ran at that native framerate.

Yes, although while native 120fps is possible, I don't expect it to be used in anything but the most basic of games. I suspect that the 120hz capability is primarily to allow the doubling of "normal" 60fps gameplay through re-projection. So in the vast majority of cases, especialy in the more demanding games, the PC will have to work harder than the PS4 on the frame rate front, absorbing 50% of that additional power.

Very interesting news, Sony will allow devs to target native 90hz display mode on PSVR! This mode will not use upscale to 120fps.
http://www.roadtovr.com/sony-confirms-new-90hz-display-mode-for-playstation-vr-formerly-morpheus/

Supported PSVR modes:
  • 60Hz rendering » 120Hz display reprojected
  • 90Hz rendering » 90Hz display
  • 120Hz rendering » 120Hz display

That's interesting indeed. I'd also say it confirms, or at least strengthens the argument that 90fps native > 60fps re-projected to 120fps, otherwise there would be no reason to include this mode.
 
That's interesting indeed. I'd also say it confirms, or at least strengthens the argument that 90fps native > 60fps re-projected to 120fps, otherwise there would be no reason to include this mode.

Or they included it because they just want to give devs more choice.
 
It depends on the game. The devs said the reprojection doesn't feel like 120 with specific games which have high velocities and specially objects flying close to the camera. Those games are better at 90, other games are better with 60-to-120. This update is very cool because it means the difficult games don't have to reach 120. I assume slow paced adventure games and exploration would be better reprojected, since there's no fast translation to mess it up.

They also said the most popular mode will be 60-to-120 which means it's worth it for most games.

They said 90 is better than 90-to-120, which means 60 is better than 60 to 90, which explains why oculus/vive cannot use reprojection to increase the frame rate.

Also the fixed 90 from Sony has a little less latency than the 90 from competitors. I think the frame reaches the display 2ms earlier despite being the same frame rate.

- Adventure, exploration, platformers, and games requiring high graphics fidelity, should be 60-to-120.
- Shooters and racers will be a mix of 90 fixed and 60-to-120 depending on which one works best.
- Any game that doesn't need high graphic fidelity will be 120.

No other platforms have that flexibility, they will all be 90 fixed.
 
Is it ironic that someone in the pc master race interprets more choice as a bad option?

Well done for completely missing the point.

My point... was that the choice they are giving is clearly 60fps re-projected to 120fps OR 90fps native, OR 120fps native. Now given that 90fps native is going to require more performance than 60fps re-projected to 120fps, there would be very little point in offering that option if it wasn't going to come along with better quality that 60fps re-projected to 120fps.
 
I agree they wouldn't offer it if it was useless in every situations, but nobody claimed this. It's a case by case basis. It's awesome how much the current devs in VR are discovering the fundamentals of this new medium (this new medium from 1993). Nothing is simple here.

A slow moving environment will be clearly better with 60-to120, while a high speed game will be clearly better in 90. Anything inbetween is going to be very interesting. The multiple frame rate choices are giving an edge to Morpheus in efficiency. If most game use this reprojection feature unavailable on PC, it reduce the ps4's horsepower deficiency and provides better VR for the price.

Your claim that 60-to-120 is the inferior choice is an error in basic logic.
 
Last edited:
I think every PSVR game is 1080p, I ever encountered any talk that some games are sub-native.


There is another native 120fps game, space shooter that was quickly made by a very small indie team.
Would certainly need some nice MSAA/EQAA, especially for VR.
Graphics are so simple that I'm sure that 8xMSAA with coverage samples wouldn't cost too much to fit frame within 8ms.
 
Well done for completely missing the point.

My point... was that the choice they are giving is clearly 60fps re-projected to 120fps OR 90fps native, OR 120fps native. Now given that 90fps native is going to require more performance than 60fps re-projected to 120fps, there would be very little point in offering that option if it wasn't going to come along with better quality that 60fps re-projected to 120fps.

Obviously with re-projection not being black magic, there will *have* to be artifacts, small or not. There can be instances for very fast moving games that going native framerate would be beneficial, but 120fps could prove to hard to handle. I don't see why having this option can be seen as a negative.
 
A slow moving environment will be clearly better with 60-to120

Is there evidence of this?

The obvious advantage of the 60 -> 120 is that it requires less performance for an experience that doesn't suck. I've seen no evidence yet though that its a fundamentally better experience - even in just slow moving games - than 90 native fps.
 
At 60 fps you can have more eye candy. In a slow moving game, the interpolation will work successfully (positional deltas small enough). that gives you 50% more GPU time to make for pretty graphics. Faster games that the interpolation can't handle robustly will need a higher framerate, requiring a compromise in art.

The choice of 60 or 90 Hz gives devs the best option for their game. When it was only 60 Hz reprojected or 120 Hz native, games would either be very simple (120 Hz) or have to stick to being slower. The addition of 90 Hz gives a nice middle ground, so 3 flavours of game presentation are supported (and quite probably 120 Hz native is ignored).
 
Is it theoretically possible that VR is able to switch frequency on the fly without visible delays/artefacts? So a game could switch the modes dynamically and during fast paced scenes increase framerate by dimming detail/effects.
 
I don't imagine it could do so seamlessly unless the display has some variable timing system (FreeSync). AFAIK displays have some time-out when switching frequency. You'd probably rather have the devs set chapters/scenes at the preferred refresh.
 
I think the original assertion was that Morpheus would give an inferior experience because games will always run at a lower framerate. That is fundamentally untrue since the device actually has the highest hz of any headset.

While it's true that there will be minimal games that run at native 120hz, those games that actually do (as basic as they may be graphically), will actually give a better presentation on Morpheus than the HTC Vive or the Rift regardless of the GPU/CPU combo in the relating PC. Unless the minor resolution increase negates it - which I doubt.
 
Back
Top