Slashdot Mirror


User: wonkey_monkey

wonkey_monkey's activity in the archive.

Stories
0
Comments
8,419
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 8,419

  1. I demand my own personal emojis. I'm triggered by emojis that differ from me by gender, race, eye colour and baldness pattern.

  2. Re:my nexus player does this on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Is it the Nexus player doing this or my TV?

    It probably is the TV. It may have different settings for different sources. Look for "judder reduction" or something like that.

  3. Re:Some graphics card already do on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Not what the submitter is talking about.

    There's "motion compensat[ed interpolation]" and there's "compensating for player motion." Two entirely different things.

  4. Re:Because it works by delaying frames on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Everyone is conflating two entirely different concepts.

    First, there's motion compensation - perhaps better termed "motion compensated interpolation." That's what the GP and the submitter are talking about. For that you need the future frame.

    But there's also what Nvidia call Asynchronous Timewarp, which compensates for the motion of the player (but otherwise has nothing to do with the other kind of motion compensation) by warping a single frame, either because it has the extra time to do so before display is required, or because the new frame is not yet ready, so gets dropped and replaced with a warped copy of the previous frame to mitigate stutter.

  5. Re:Because it works by delaying frames on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Yes, it really would be noticeable.

    This is why modern TVs have game modes to turn off motion compensation and other processing. Otherwise there's too much latency and gameplay suffers.

    Even with game mode, I still find playing on a CRT or other truly-dumb monitor to be much more responsive for gameplay.

  6. Re:Also because it is artifact-prone on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    It doesn't help that the technique is also *very* prone to artifacts

    But the submitter said "no visible stutter or strobing whatsoever"!

    Do you mean to say he may be mistaken?! I just don't know what to believe any more.

  7. They already know the camera position for the future frame

    How? Player input and/or head position can change before the future frame is to be rendered.

  8. Re:Because there is no such thing as magic on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    And another is that there is no difference in focus in a VR world. Your eyes will strain to make sense of that, just as they do when you wear someone else's glasses.

  9. VR can't happen with current or foreseeable technology for precisely the reasons you mention.

    Err, no. The reasons mentioned have nothing to do with why VR "can't happen." The motion compensation trick is entirely unnecessary if the GPU can render frames fast enough. And it runs counter to the aims of VR, anyway.

    In case you hadn't noticed, VR is already happening. It might fizzle out, it might not. It's too early to say.

    It sounds like you just have a beef with VR, saw an opporunity to air said beef, and took it without really reading what was being said.

  10. Although the broadcast image was interpolated

    Interlaced.

  11. They also do a warp if (on hopefully rare occasions) the GPU can't get the new frame ready in time, repeating the previous frame but warped to match the new head position.

  12. Pretty much this.

    No, not this at all. A lot of people are making this mistake. It's not what the submitter is talking about. He's talking about what is usually called "motion compensation" but should, in this case, probably be called "motion compensated interpolation."

    It has nothing to do with the motion of the viewer's head - which is what Async Timewarp is all about - but is about the "motion" between two frames of video, and using the detected motion to create an interpolated frame between each "real" frame to increase framerate.

  13. Re:Asynchronous Timewarp on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Asynchronous Timewarp is something different than what the submitter is talking about.

    The submitter is talking about interpolating between rendered frames to create smoother video - i.e. only rendering every other frame fully, then just interpolating between them to fill the gap.

    The obvious problem is that it means you've added an extra frame's delay to everything. Also, motion compensated interpolation isn't perfect.

  14. Re:THEY ALREADY DO on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    Time-warp is not motion compensation, not in the sense used in the question.

    It does compensate for motion, yes, but "motion compensation" is a different thing in video terms (it should probably be called "motion compensated interpolation.")

  15. Because you can't interpolate into the future on Ask Slashdot: Why Don't Graphics Cards For VR Use Real-Time Motion Compensation? · · Score: 1

    You need to know what the next frame will be before you can interpolate towards it. This means delaying everything, That's a big no-no in VR.

    Also, motion compensation is not perfect. That's why all these "60fps!" trailers people like to put on YouTube look quite underwhelming.

    Real 48/60fps video looks a lot nicer.

  16. That's not motion compensation - not the kind the submitter is talking about, anyway.

  17. Re:Huffman alternative on Dropbox Open Sources New Lossless Middle-Out Image Compression Algorithm (dropbox.com) · · Score: 1

    First, the claim that it reproduces the original file "bit-for-bit perfectly".

    By "original file," they mean a JPEG.

  18. How about BPG? Looks better than JPEG2000 to me.

  19. Re:She "may" become the new prime minister on Theresa May Becomes UK's 'Spy Queen' and New Prime Minister (arstechnica.co.uk) · · Score: 1

    I know, it's like lolcats are writing for the news.

    Theresa may has become PM?

  20. If it's a statement, then someone must have said it.

  21. If you only expect to need such a tower for a few days every year - or even a month - it could easily be far more cost effective to use drones. They can be up and down in minutes. You couldn't say the same for a 50 or 100m tower. You also couldn't fix a broken one in the middle of a big event, whereas you could just send up a backup drone.

  22. Re:So much coverage! on Why So Much Coverage Of Amazon Prime Day? The Incentives, Of Course (theguardian.com) · · Score: 2

    A lot less than they'd have got if a Prime Day had been posted before Prime Day.

  23. Yeah, and except for the thing of them having different meanings, you're absolute right!

  24. That is absolutely true on Pokemon Go Was Never Able To Read Your Email (gizmodo.com) · · Score: 1

    Pokemon Go Was Never Able To Read Your Email

    It certainly wasn't. I've never installed it.

  25. Now release Wii-U Zelda dammit.