Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John Carmack: Latency Mitigation Strategies

altdevblogaday.com

Post reply on HN
Print view View original

41–50 of 106 posts

Re: John Carmack: Latency Mitigation Strategies

#41

Post by knodi »

stcredzero wrote: post #37

> Updating the imagery in a head mounted display (HMD) based on a head tracking sensor is a subtly different challenge than most human / computer interactions. Doesn't the military already have this solved for head mounted displays for attack helicopters and 4++ generation fighter jets? Heck, they have augmented reality displays for that matter. EDIT: Many of these problems could be solved by putting an entire purpos…

Yes, they have with very powerful hardware thats not very consumer wallet friendly.

Re: John Carmack: Latency Mitigation Strategies

#42

Post by seanalltogether »

breckinloggins wrote: post #31

It sounds like we would benefit from GPU and display manufacturers providing a set of standard low-level control primitives so that people like Carmack who know what they are doing can really play around with the entire pipeline without having to worry about all the things the cards and the displays are doing behind their backs. For example, a GPU could have a set of standard settings with full buffering and all the…

Carmack has also mentioned in the past we could solve other buffer problem with non-isochronous displays, essentially a display that doesn't run on a fixed hz cycle but rather outputs lines to the screen as it comes down the pipeline at whatever framerate the host system can handle

Re: John Carmack: Latency Mitigation Strategies

#43

Post by Erwin »

dman wrote: post #15
dchichkov wrote: post #12

> Conventional computer interfaces are generally not as latency demanding as virtual reality, but sensitive users can tell the difference in mouse response down to the same 20 milliseconds or so, making it worthwhile to apply these techniques even in applications without a VR focus. I wish people would start treating text editors that they are developing this way. As hard real time systems. Don't care much about virt…

Would you be willing to pay a recurring fee for using an editor? I have toyed with writing an editor at multiple points but have backed away because I think in the grand scheme of things its a one time sale to a pretty small market (The number of people who care enough about text editing that they would pay for a good editor when they see one)

I pay, effectively, a recurring fee for PyCharm. The personal license is €89 and comes with 1 year of upgrades. As they've been able to release upgrades after that year expired that looked interesting, I paid another €50 for it the next year support.

Personally I'm not keen on a recurring payment unless there's a clear benefit. For an editor, that'd be having a team committed to making it better. The JetBrain guys have shown good progress.

Re: John Carmack: Latency Mitigation Strategies

#44

Post by niggler »

stcredzero wrote: post #37

> Updating the imagery in a head mounted display (HMD) based on a head tracking sensor is a subtly different challenge than most human / computer interactions. Doesn't the military already have this solved for head mounted displays for attack helicopters and 4++ generation fighter jets? Heck, they have augmented reality displays for that matter. EDIT: Many of these problems could be solved by putting an entire purpos…

"Doesn't the military already have this solved for head mounted displays for attack helicopters and 4++ generation fighter jets? "

Far from it. There was a recall a few years ago over glitches in a fighter jet HMD. It's still an active area of research and no less than 5 different systems employed by the US military divisions

Re: John Carmack: Latency Mitigation Strategies

#45

Post by laureny »

stcredzero wrote: post #36
dchichkov wrote: post #12

> Conventional computer interfaces are generally not as latency demanding as virtual reality, but sensitive users can tell the difference in mouse response down to the same 20 milliseconds or so, making it worthwhile to apply these techniques even in applications without a VR focus. I wish people would start treating text editors that they are developing this way. As hard real time systems. Don't care much about virt…

> I wish people would start treating text editors that they are developing this way. IDEs. Most of the things people do in modern IDEs were done in Smalltalk (albeit, with uglier interfaces and some manual steps) over a decade back. One of the differences, though, is that these things were mostly highly responsive, because they were bloat-free. (Can't speak for IBM VisualAge. Also, one could rightly complain that the…

> One of the differences, though, is that these things were mostly highly responsive, because they were bloat-free.

That's not my recollection of the Smalltalk IDE at all. I especially remember how selecting was crazy slow: press down the button, move the mouse to the end of the line and the selection takes two seconds to catch up with the cursor.

Re: John Carmack: Latency Mitigation Strategies

#46

Post by ealloc »

neverm0re wrote: post #21
marshray wrote: post #16

Earlier quoted context omitted.

> mismatch between what you're seeing and your inner-ear pressure How does your inner-ear pressure change such that it cannot be fixed with better display and rendering?

Because the working theory on what causes simulator sickness is related to the inner-ear detecting motion. If your eyes are detecting motion, but it does not line up with what your inner-ear pressure is telling your body, the end result is the body emptying its contents under the theory that it's been poisoned. Simply improving the quality of the picture isn't going to solve that.

It seems to me the oculus rift can accrately account for all head motion when you are standing still. There is no need for eye-tracking: All the rift has to do is detect all head bobbing & rotation, and update the view acocrdingly. Your eyes can still look around on the lcd screen.

What the rift _can't_ do is account for long term linear acceleration (eg in a car or walking around), it's true. If you had a large empty room to walk around in with the rift on, even this could be accounted for, however.

Re: John Carmack: Latency Mitigation Strategies

#47

Post by breckinloggins »

Scene_Cast2 wrote: post #40
breckinloggins wrote: post #31

It sounds like we would benefit from GPU and display manufacturers providing a set of standard low-level control primitives so that people like Carmack who know what they are doing can really play around with the entire pipeline without having to worry about all the things the cards and the displays are doing behind their backs. For example, a GPU could have a set of standard settings with full buffering and all the…

Regarding your last point: it's getting better. Modern TVs have "game mode", where latency is minimized. The latest LCD panels use embedded Display Port, eDP. The latest eDP standard introduces a frame buffer on the LCD itself, where you can just write deltas and tell it to "swap". I think this is quite similar to what you're proposing, in concept.

I found this [1], which seems to be a good introduction to the new features and also why eDP is poised to replace LVDS (which I, as a non hardware guy, had to deal with in a previous embedded device project and hope to never deal with again).

[1] http://embedded-computing.com/articles/embedded-greater-disp...

Re: John Carmack: Latency Mitigation Strategies

#49

Post by drivers99 »

neverm0re wrote: post #30
modeless wrote: post #19

Earlier quoted context omitted.

Eye tracking would be awesome but it's not required for good VR. I fail to see how "using your head as a dumb camera joystick" is a problem; that's the whole point of VR. Motion sickness is a real problem for some people (but not all). With low latency and thoughtful game design I think it can be mitigated. The bigger problem is the social acceptability of blocking your entire field of vision for long periods of time…

> I fail to see how "using your head as a dumb camera joystick" is a problem; that's the whole point of VR. Perhaps you should try playing some VR games for a while. I clocked quite a few hours playing the old Virtuality SU2000 games like Dactyl Nightmare and I've idly kept up with HMDs. The chief problem is that without eye tracking, it's incredibly UNFUN to use your head for movements that your eyes could otherwise…

> The chief problem is that without eye tracking, it's incredibly UNFUN to use your head for movements that your eyes could otherwise have done for you. Mind you when the HMDs were much heavier back then it sucked a lot more, but it's still pretty shitty not being able to glance aside.

Isn't that more of a problem of field of view? With a wide enough field of view, you could just look at whatever you wanted to. Basically make it like real life, where what you look at is the combination of where your head is pointed, which the computer needs to use to update the screen, and where your eye is looking, which the computer doesn't need to care about. Or are you thinking of using eye tracking to something else, like determining what you're pointing at like a mouse? Yet another option you have in VR is what you're targeting, like how Dactyl Nightmare uses the gun you're holding to map directly into the virtual world's gun. Using the head position to AIM rather than just LOOK seems like a bad way of doing things.

Re: John Carmack: Latency Mitigation Strategies

#50

Post by nitrogen »

roc wrote: post #25
nitrogen wrote: post #20

Earlier quoted context omitted.

I think the PS4's insane memory bandwidth (8GB of memory on a 170GB/sec GDDR5 bus) makes it a very interesting contender against PCs (dual-channel DDR3 maxes out at 34GB/s).

Given the PS4 is using an off-the-shelf x86 processor, how long do you think it'll be until other OEMs are pitching laptops with these same CPUs that have GDDR5 controllers for main memory? How long until Intel matches? How long until basically every CPU supports it and every manufacturer has one or more such systems?

They may be using an off-the-shelf x86 core, but their actual CPU die may have all kinds of customizations on it (like the addition of a GDDR5 memory controller). AIUI GDDR5 memory is much more expensive than DDR3. So it may happen eventually, but I don't think it will be any time soon, and believe me, I really want PCs to be better than consoles in every way.
Post reply on HN