John Carmack's VR head set

Rodéric said:
Screens can have severe latency, up to 50ms and sometimes more, depending on the electronics.
On serious screen reviews you'll see delay compared to CRT, screens with 0 delay are *very* rare.
Click to expand...

0 seems impossible as the fastest LCD pixels have been managed to shut off is 2ms apparently, according to Wiki (while 8ms+ is still happening). Of course, at 8ms in theory you can get to 120 refreshes p/s, but if the pixel doesn't manage a full off (I think many pixels are still somewhat more lit then off at that point, often half values being considered acceptable?), that's still a limitation. All the additional scaling and processing delays comes after that, and that is far too often 40ms or more in total.

Since I have the Vita I've become more aware of this than previous. But Foosball 2012 on the PS3 was also interesting, as because it has almost no controller assists on the move controller, the lag when using the Move controller seems to be extremely low, 66ms or maybe even 50ms (which would be one of the very, very few games at that response time if true).
 
You can't get transition time to 0, but you can get input-output delay from HDMI to screen to defacto 0 (overdrive can be implemented with a purely causal algorithm, even though not all manufacturers do it, lookahead is not strictly necessary).

If you can use a TN screen then really compared to other delays in the system even with transition time it's all close enough to 0 not to matter (colour/intensity angle dependence is not necessarily a problem ... the angles to the eye are fixed, in theory you could compensate).
 
MfA said:
As I said tv's can buffer a few more frames because they do more complex conversions, but this has little to do with LCDs
Click to expand...
I was referring to LCD TVs. You rarely connect bare LCD panel to PS3/360 and do all the plumbing yourself, don't you?
 
This is however a thread about the oculus rift, and the current discussion started when I responded to the claim that OLED would solve latency problems for VR ... which it doesn't, because as I said, latency is mostly in the rendering, not the screen.
 
MfA said:
This is however a thread about the oculus rift, and the current discussion started when I responded to the claim that OLED would solve latency problems for VR ... which it doesn't, because as I said, latency is mostly in the rendering, not the screen.
Click to expand...

No that was only part of what was said. Feel free to go tell Carmack he is wrong, we are just the messengers, he was adamant throughout the entire keynote that as displays it exist and it's a problem.

What was also said is that John Carmack thinks that most traditional LCD screens have a big problem with stereoscopic 3D. He gave two main reasons ;

1. They can't do true black (off).
2. The grey to grey latency is too high causing ghosting effects.

I'm guessing you haven't watched it yet, he does go into quite a bit of detail about it and it was very interesting to watch. So you should watch it, that way you can argue with John and not us :smile: .
 
Chill, itsmydamnation. There's really nothing to argue over. :) MfA is correct in at least one thing - I did not review the flow of the entire conversation. There are actually two "threads" here in, not one: one about the viability of LCD TVs, another about LCD displays themselves and feasibility of these in VR headsets. Both got confused at one point and here we are.
 
I'm completely relaxed and my post isn't supposed to be aggressive or defensive in any manner. Reality is i dont have an opinion because I haven't tried all these different screens in a headsets, John has. John has said its a problem for LCD's there's no point arguing with us about that.

He explicitly states that when you try the RIFT that the three major problems are:

1. Screen resolution
2. The LCD latency causing ghosting in 3D
3. Head tracking when you do more complex head movements or things like to were standing your now sitting.
 
Last edited by a moderator:
itsmydamnation said:
No that was only part of what was said. Feel free to go tell Carmack he is wrong, we are just the messengers
Click to expand...
What does that have to do with anything? You weren't discussing Carmack in general, you were replying to Billy Idol ... who said :
"His enthusiasm during the keynote really got me to the point where I want VR gaming myself...but it seems that the tec is still far away. I think he is right, that one of his major concerns is simulation sickness. The smallest movements of your head need a consequence in the simulated world."
What was also said is that John Carmack thinks that most traditional LCD screens have a big problem with stereoscopic 3D.
Click to expand...
Billy Idol was talking about VR specifically ... or in other words, HMDs ... the problems traditional LCD screens have with shutter glasses are neither here nor there, and OLEDs do not solve the fundamental latency problem Billy Idol was talking about (since rendering latency is the largest part of it).

As for Carmack, where exactly did he call response time latency?
 
Its been around for ages
Cat Stabilization was First pioneered by Eddie "The Eagle" Edwards during the 1998 winter Olympics
 
How cute ... !

I'm about to get my 10 yo cat from the vet, had to have one eye removed (virus induced cancer :( ), looked just like that one when I got him first (though he was a mess, stray cat in bad condition, cleaned up nicely though stunted in his growth).

Anyway back on topic:

http://nl.hardware.info/reviews/283...ratie-testresultaten-reactietijd-en-input-lag

shows that 0-100-0 pixel values for panels are indeed not great in general, much worse than I thought, and IPS panels are worse. So there is indeed an important win here for OLED.
 
Anyone have an educated guess as to what sort of rendering load and resulting new/different bottlenecks there would be with this method of split screen stereoscopy? Being that each field is doubled, but half the pixels, any sort of per-pixel shading/texturing/etc should be roughly the same pixel load as having just a standard single full frame? Scene geometry would be presumably doubled though?
 
You must log in or register to reply here.
Back
Top