Slashdot Mirror
Valve To Demo Prototype VR Headset, "Steam to Support and Promote VR Games"
An anonymous reader writes "The itinerary for Steam Dev Days 2014 lists two talks by Valve's internal virtual and augmented reality researchers, Michael Abrash and Joe Ludwig. Abrash's talk, titled 'What VR Could, Should, and Almost Certainly Will Be within Two Years' will feature a demonstration of Valve's secret prototype VR headset that is 'capable of stunning experiences.' Ludwig's talk 'Virtual Reality and Steam' will discuss how Valve will be adapting Steam to VR to 'support and promote Virtual Reality games.' Rift inventor and Oculus co-founder Palmer Luckey will also be taking to the stage at Steam Dev Days to speak on best-practice for VR development."
There's a hint that they might be showing off a head mounted display featuring a low persistence display, which would be great news for those of us that get the urge to hurl when playing Doom on a conventional display. If you missed it you might want to check out the slides and notes (PDF) from Michael Abrash's GDC2013 talk on VR.
Low persistance displays are a tricky issue.
They obviously don't have the issues that high-persistance displays have of holding frames for too long. However, they have another annoying effect, commonly referred to as the strobe effect. This has to do with each pixel being lit for only a minute duty cycle on the display. This causes bad flicker at low refresh rates.
Early low persistance displays obviously were not very good on this issue. This is because the displays used very slow technologies such as oscolating mirrors.
By the details I've read on their blog, I'm pretty certain Valve has gotten down that they need a high refresh rate to get the VR to work right. They have identified strobe effect as a problem, and have identified that while the traditional 60Hz rate, while tolerable, is far from ideal for low persistance displays. They seem to believe they can push the refresh rate high enough to deal with strobe effect. I have confidence that they can.
Higher refresh rates also have other advantages for gaming as Internet router designs improve and ping times drop, the latency produced by interpolation becomes more substancial, and the best way to reduce it is to push more physical frames. If you are pushing more physical frames, there are clear advantages to pushing more visual frames to match.
VR
V = 22
R = 18
V - R = 6
Two letters in VR: 6 / 2 = 3
Half-Life 3 Confirmed
Kinda both. TFA mentions that the Oculus guys were shown a hands-on demonstration of Valve's headset and plan on implementing some stuff they learned there in the Rift. I have no doubt they'll be sort-of competitors, in that people are unlikely to buy both, but Valve's position on hardware so far has been to let other people handle the manufacturing and only do prototype and design work themselves, so it remains to be seen if Valve even directly produces a VR headset. They are, however, producing at least a prototype version of their own set.
However, competition is good. It'll drive anyone who makes VR headsets to produce a headset better (at least in some ways) than the other guys, and give consumers a bit of choice. Plus highlight problems that one set has that others don't, and maybe give everyone some ideas how to fix those problems. Provided, of course, games are intercompatible with all sets on the market (otherwise you just get fragmentation which in practice can be worse than a monopoly).
"None can love freedom heartily, but good men; the rest love not freedom, but license." --John Milton
... featuring a low persistence display, which would be great news for those of us that get the urge to hurl when playing Doom on a conventional display.
Note that many CRTs (basically all modern color CRTs and most modern monochromes) are low- (not zero-) persistence, with phosphor decay on the order of microseconds to tens of microseconds. Some of the visual effects are bound to be very different with a exponential or power law decay than with the sharp cutoff of scanning devices, but it does suggest ways for some of us who aren't in secret VR prototype labs to experiment with some of the stuff he's talking about.
I'm not about to do this now, but back when I was about 12 and could do such things without getting dizzy, I tinkered with motion perception by making programs that would scroll an image horizontally across a CRT, lay on my belly across a swivel chair with my feet in the air, and spun myself by pushing off the legs of the chair with my hands to get a near-constant speed that synced with the monitor.
If I didn't have better things that needed doing, I'd strap a CRT display and an LCD onto a lazy susan, together with an Eee or such to drive one of them (swap plugs to repeat experiment with low- or high-persistence), and spin it instead of myself. (It's kinda sad that I do have better things that need doing, and yet I'm posting this on /. instead of either doing them or doing visual perception experiments...)
Also, VR is the state-owned railroad company in Finland. And railroads were a quite recurring theme in HL2. Pure coincidence? I think not!