1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Steam Half Life: Alyx

Discussion in 'Gaming' started by RedFlames, 18 Nov 2019.

  1. d_stilgar

    d_stilgar Old School Modder

    Joined:
    11 Feb 2010
    Posts:
    1,041
    Likes Received:
    145
    Yes, the 2700k will be fast enough. The game is very well optimized and looks great even at low settings. It'll also suggest settings for you upon launch that should ensure a smooth experience.

    Also, the Quest is great for what it is, but if I were trying to go budget VR, I would either go used Vive w/ Index controller upgrade or Windows Mixed Reality (such as Odyssey+) which go on sale almost monthly.
     
  2. Vault-Tec

    Vault-Tec Green Plastic Watering Can

    Joined:
    30 Aug 2015
    Posts:
    14,886
    Likes Received:
    3,667
    Not here in the UK they don't. I had my eye on a few sets for nearly two years and they are grossly over priced here. Even the crappy Dell one is £400.
     
  3. wyx087

    wyx087 Homeworld 3 is happening!!

    Joined:
    15 Aug 2007
    Posts:
    11,987
    Likes Received:
    706
    PC VR gaming isn't the only thing I want a VR headset for.

    I like the idea of Quest where it's a VR headset that can do PC gaming (possibly even wirelessl!) as well as stand alone for times when I just want to watch a video, or play basic non-graphic intensive activities. The reason I didn't get into VR was because I don't want an expensive monitor tethered to the PC for just one purpose. I'd have gotten on Quest earlier if I knew it can play PC games.
     
  4. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    Rift S and Quest controllers (identical and interchangeable) come with hand straps to prevent nearby displays catching Wiimote Syndrome.
     
  5. Vault-Tec

    Vault-Tec Green Plastic Watering Can

    Joined:
    30 Aug 2015
    Posts:
    14,886
    Likes Received:
    3,667
    Hmm. That'll teach me to listen to a Youtuber. Oh well, at least I will have spares :)
     
  6. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,889
    Likes Received:
    1,156
    Yeah so get this - when I first got my Vive, I left the missus in charge of choosing the first game to play, and she picked a downhill skiing game.

    I felt so ill I actually whited out and couldn't stand up, it was ****ing horrible.
    I had the same idea, I wanted to try VR porn, VR movies, VR youtube, 360 videos - all that jazz. Tried them all on the Vive and found them weird and underwhelming because of the resolution of the screens and the oft-remarked-upon 'screen door' effect. Especially when going from a 4k and/or OLED TV down to the Vive headset, it's like, "wow, I can reach out and touch the pixels".

    For gaming it's not an issue because you're involved, and your brain kinda switches into active mode. For media consumption, I found it really distracting and ugly.

    Some of the other headsets have higher pixel density than the Vive now, but I'm not sure they'll be high enough yet to overcome this problem. Moderns TVs and monitors have gotten us used to pretty liquid images and going back to jaggies for media consumption is a nope for me.

    On a separate note since I mentioned it: approach VR porn and other airquotes "immersive" experience-focused videos with extreme caution. They made me feel off - not motion sick, but like it was just goddamn weird. Where normal media taking place on a screen is something we're used to suspending disbelief for, VR immersive videos just feel 'off' somehow. I couldn't suspend disbelief, found it massively distracting to be 'inside' a real video. Like when people who've never seen theatre before try it and can't stop thinking "that's just a guy in a costume and I could reach out and trip him up", I couldn't stop thinking "wow this is awkward, the person holding this huge 360' camera must be breaking his neck right now".

    So many things about VR are hard to anticipate accurately. You gotta try it I guess. I know loads of people who love immersive video, VR porn, and even that weird 'cinema' mode where you watch films in a virtual cinema (seriously why), but I just don't get on with them.

    On an Alyx-related note, why are devs ramping up their VR requirements so ridiculously high? I'm still on 3rd gen i5 and it's taken everything I've thrown at it. Which is hardly surprising considering the performance jump across the subsequent 5 years is like...50% (where by Moore's law it should be about 300% by now). Devs are clearly still aiming for the ~3470k target, because everything works on mine seamlessly, including VR and Modern Warfare 2019, but the requirements are all consistently telling me I shouldn't be able to run anything any more.

    Are they just in Intel's pocket and trying to stimulate consumption of new hardware, or what?
     
    Last edited: 7 Apr 2020
    wyx087 and Pete J like this.
  7. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    '360 video' and '180 stereo video' are absolute trash. Worthless garbage whose association with VR lessens it as a medium.
    The primary issue is that they lack any sort of accommodation for head motion. No head motion = near zero feeling of presence, in addition to rapid onset nausea. And that's in a best-case with obscene resolution, obscene bitrates, and display-matched framerates. Add in the realities of encoding video for a normal PC (let alone a mobile device) to handle and you have a steaming pile of crap.

    Because performance for flatscreen gaming != performance for VR gaming.
    For flatscreen gaming, you can be doing multiple things in parallel, like having one frame postprocessing while the next frame is rendering and the frame before that is getting the geometry ready to pass to the GPU, and the frame before that is doing the physics simulations. Several frames of latency is not only tolerable but almost unnoticeable for most parts of the pipeline, and GPUs are designed around doing a bunch of stuff in parallel whenever possible.
    VR demands low latencies uber alles. You have a 20ms budget from starting your frame to having it displaying on the HMD (often called the 'motion to photons budget', as it starts wehn you head moves and ends when you see the finished frame), and going beyond that means you've got a frame miss and need to be rewarping a past frame to replace it (in reality, fixed latency overheads like frame scanout and the time to get that rewarped frame ready mean you have ~17ms of time for Oculus OVR API games, and ~11ms for SteamVR API games, withi nwhich to get all your rendering done). That means that CPUs and GPUs that run faster (rather than being wider') are dramatically more effective for VR performance even if that has little effect on 'normal' gaming performance.
    That 20ms threshold is the maximum time that is acceptable for comfortable viewing. You gain comfort benefits for getting things done faster, but going above that is a guaranteed unpleasant experience. This is why everyone now uses universal post-render rewarping for every frame (so it is already accounted for in your fixed latency budget) and everyone has a frame-synthesis technique to accommodate frame misses (because if a frame takes more than 20ms to be ready it may as well not have been rendered at all). Frame synthesis is not perfect, so you can have a very dramatic change in perceived quality fro ma very small change in performance: if you're handling 19.9ms per frame you're golden, if you're getting 20.1ms per frame you're stuck in constant frame synthesis.
    The other problem is that everyone is used to looking at framerates. You have a 90Hz HMD, you are getting 90FPS, so everything is good, right? The problem is framerates tell you nothing whatsoever about latency so are utterly worthless for gauging VR performance.
     
    Last edited: 9 Apr 2020
    Pete J, wyx087 and boiled_elephant like this.
  8. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,889
    Likes Received:
    1,156
    That was a very interesting infodump, thank you!
     
  9. wyx087

    wyx087 Homeworld 3 is happening!!

    Joined:
    15 Aug 2007
    Posts:
    11,987
    Likes Received:
    706
    That's interesting, in your opinion VR as a medium are only really good for games?

    I have to admit, reason for Quest was 3 fold:
    - possibility of able to consume media on plane or the like, to drawn out other distractions.
    - "immersive media" :naughty:
    - possibility of wireless VR gaming without spending more for wireless adaptor. Link cable was my fall back options
    I know all 3 are in the category of "try it first".

    On the other hand, if I get Rift S, I'd at least guarentee a good budget/mid level HL Alyx experience and able to try other two and decide if it was worth buying when second or third generation stand-alone PC-linking headset later. Decisions, decisions.....

    I've only briefly tried VR simulator on Vive, it was great; a ultra-cheap cardboard smartphone holder, that was utter rubbish. Mainly due to poor controls and low resolution of a iphone plus, as well as the cheap lens.
     
  10. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,889
    Likes Received:
    1,156
    Personally I'd focus the budget and the expectations on the gaming end of it, yeah. A lot of VR gaming is really good and I'm enjoying it in small doses - still not the kind of thing I can play for hours on end, personally, 'cos of comfort, heat, and general disorientation. But video and media? Eeeeeh...

    I was honestly really psyched for VR porn, I've always loved porn (who doesn't, right) and it seemed like the next logical step. But for all the technical reasons edzieba mentioned, immersive video is reeaally bad and weird, and for other psychological/personal reasons I found porn in particular to be super bad. Generally any video where there's people, interactions, social settings, things close to the camera - even just a 360 video of some people on a rollercoaster or whatever - it feels silly, and the camera tech does not trick your brain the way a game engine does. I couldn't turn off the part of my brain that knew I was sat alone in the dark, feeling like I was uncomfortably close to total strangers whilst simultaneously knowing full well that it wasn't real.

    This might be different for other people, it might just be because I'm quite introverted and cagey in general. But your thing about planes brings up another point, I find it really odd to sit and use a VR headset with other people around. I guess I knew too many ****heads at school, but all I can think the whole time is "someone's probably ****ing around with me 2 inches from my face right now and I don't know it". Unless you're a super chill person it's quite discomforting to know that there are people very close to you but be unable to see or hear them. Playing standing games in a group with VR is a blast - you take turns, laugh at each other, and it goes down really well. Probably my favourite thing to do with it is introduce people to Gorn, Space Pirate Trainer or experience games like The Blu. But using sat-down entertainment for long periods with other people around feels uncomfortable to me. Me and the missus will happily sit playing games on our rigs all evening - different games, even - but I hate sitting and using the VR for something like Elite Dangerous when she's there, because it feels antisocial and I'm very aware that she can just see me, sat there like a neo-noir tech zombie.

    The sheer number of 1st person pronouns in this post sums up how deeply personal and individual all this is, though. My experiences won't necessarily be your experiences. And that's kind of the problem: VR is a hugely individual thing. This, to be briefly not-off-topic, is why I resent Alyx as a Half-Life instalment, because so many people will fundamentally just not like or even be able to use VR, even if they want to and have the cash. It's as personal as whether you enjoy rollercoasters, whether you can watch horror films, or what flavours in food make you gip.

    So Rule 1 of VR, whatever you're gonna buy, find some way to try it first.

    edit - oh you literally acknowledged this already in your post. But yeah, a thousandfold yeah, try it first.

    edit 2 - oh, and the immersive video thing has one notable exception, which is super high quality documentary/scenic style video. Slowly gliding along a river on a boat, plane flight over the Grand Canyon, all that ****. It's got a sort of Planet Earth-esque appeal that doesn't rely on immersion, so as long as the purpose is purely to go "ooh that's pretty" and see amazing sights, as a passive observer, there is some value in that - providing the source quality is really good, to overcome aforementioned technical issues.
     
    Last edited: 8 Apr 2020
    wyx087 likes this.
  11. wyx087

    wyx087 Homeworld 3 is happening!!

    Joined:
    15 Aug 2007
    Posts:
    11,987
    Likes Received:
    706
    Ok. Thank you for the valuable information. Unfortunately it's kindda difficult trying out VR experience at the moment :duh:

    Yes, I think this is the way to go for now. The Quest is far from perfect, very much a first generation product (front heavy, slow-ish processor). I can try everything with PC tethered headset first, if I like other aspects of VR, I can then upgrade to 3rd or even later generation standalone PC-link-capable headsets.

    What would be a reasonable price for second hand HTC Vive? Looking on eBay, seems £150 for headset with strap, 2x £90 for base station........ plus old not-ideal-for-Alyx controllers, it's going to be over £400, price of Rift S, might as well wait for Rift S to be in stock.

    edit: Or Oculus Rift cv1? Just under £350-400, less display resolution than Rift S (so screen door effect?) but other than that, seems compatible in usability (don't mind base stations) and readily available.
     
    Last edited: 9 Apr 2020
  12. Pete J

    Pete J Employed scum

    Joined:
    28 Sep 2009
    Posts:
    7,226
    Likes Received:
    1,784
    One thing I'll say is that VR is AMAZING for aiding CAD. Only got to try it once, but to actually be able to walk around a design is incredible.

    Combine this with VR conferencing and someone could make a lot of money...
     
    d_stilgar likes this.
  13. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    IrisVR are probably doing quite well at the moment.
     
  14. Parge

    Parge the worst Super Moderator

    Joined:
    16 Jul 2010
    Posts:
    13,022
    Likes Received:
    618
    Just a big FYI: Quest Link is in beta - and it is definitely NOT ready for full release. It works like 25% of the time. I'm using the Anker cable (that Oculus Recommend) and its been a frustrating experience.
     
    wyx087 likes this.
  15. sandys

    sandys Multimodder

    Joined:
    26 Mar 2006
    Posts:
    4,919
    Likes Received:
    724
    I wonder if this will come out on PSVR
     
  16. d_stilgar

    d_stilgar Old School Modder

    Joined:
    11 Feb 2010
    Posts:
    1,041
    Likes Received:
    145
    Responding to a lot of things here.

    VR Video, 360, 180 -
    I think we're too early days to write off 360 or 180 video. Early movies were just stage plays with a camera aimed straight on. Or forced perspective visual gags, or editing tricks. The innovative early silent films watch more like technical demo reels. We didn't have the film language we know now and I think it's okay to wait and see what good might still come, what techniques translate to 360, which don't, what new things will be developed.
    That said, there's the technical limitation of not having six degrees of freedom (6 DOF) that I don't see being overcome any time soon. If you've done any 6 DOF gaming, going to a fixed head position is awful and makes you more sick than if you've never done 6 DOF.
    But I'll point out one positive that may get better with time. 3D movies are almost watchable. The resolution on HMDs isn't there yet, but I've tried a few 3D movies on my Vive and Index. It's fun to sit in a VR theater. You still get 6 DOF. The 3D effect is better than anything else I've ever done. The fact that 3D movies are already half resolution means that you don't feel the hit as hard when viewing on a (comparatively) low res HMD.
    I think the 6 DOF issue could be overcome with lightfield video, but the bandwidth for that would be insane. If you haven't experienced a lightfield yet, I'd try Google's free "Welcome to Lightfields" app, which is very impressive. It gives limited space to move your head within a captured "image" with parallax maintained. It's impressive and not a 3D model. The technical challenge to transitioning that to VR video would be the fact that the video stream would effectively be 128+ 4k videos all being streamed at once. Good luck with that, even with h.265.

    PC Spec Requirements for VR -
    As has been pointed out, you're aiming for 90fps minimum (Valve index goes up to 144hz) for two 1440x1600 screens, nonstop. That's effectively two 1080p screens (the pixel count is identical). That's 11ms per frame for 90hz, 7ms per frame for 144hz for two screens and if you get dropped frames, you get motion sickness. It's very demanding.
    So yeah, the big bottleneck for the foreseeable future will be in PC hardware. Adding resolution or increasing frame rate for VR screens is much easier than getting hardware to magically be able to push more frames for more pixels. I would say we're at a happy medium between HMD specs, computer hardware, and how devs have optimized their games.

    What VR hardware to buy -
    This issue is tougher for a lot of reasons. The best answer is going to depend on local prices, second-hand availability, and what experiences you want to have. I'll just say that the Index controllers are incredible as is the off-ear spatial audio on the Index HMD. The haptics of the controllers were put to very good use in HL:A. The way the haptics combine with some implied hand positions makes the level of "hand presence" very high. The off-ear headphones make it both easier to put the HMD on and take off. Having off-ear headphones also means the sound is attenuated by the shape of your ear, subtly changing the sound in a way that is unique to each of us but is how we're used to hearing the world already. The way you hear things makes the sound less "in your head" and, again, increases presence in the world in a subtle, but profound way.
    But barring the most expensive version of VR, I'd recommend anything but the original Vive wands, which won't cut it. The good thing is that, if you have an original Vive, you can just upgrade to the Index controllers. $400 is a good used price for a complete Vive kit. The $250 for the Index controllers is a nice way to upgrade as well if you want to upgrade as you go.
    I would also not recommend the Vive Cosmos. There are too many issues with tracking and the price is too high.
    I've heard mixed reviews of the Quest for desktop VR. You'll have the Quest OS, then the Oculus layer when linked to your PC, then the SteamVR layer. It's a lot of overhead to then try to get down to the 14ms required per frame for the 72hz refresh rate of the screens. With updates between all three softwares, hopefully things get better, but it also means any one layer could get an update that slows things down everywhere.

    VR for CAD -
    A couple years ago I was making VR building walkthroughs at work. It was a relatively easy process and everyone we showed it to was blown away. Most people can't imagine a 2D floor plan as a 3D space. I was able to show design options to clients without them ever leaving VR, switching between one remodel concept and another without them ever leaving their (virtual) kitchen. They loved it.

    Wireless VR -
    Again at work, we got the TPCast to make the Vive run wirelessly. We found it was more annoying than keeping the wires. Part of the issue was that there were still wires going down to the battery, which was worn on a belt. Maybe for long play sessions at home, this would be fine, but the trouble of getting everything on and off feels more like you're making a commitment, so short play sessions were more likely to just get skipped altogether.
    Sharing this with other people was also more annoying because we had to be even more awkwardly intimate with people as we helped with the battery belt. So what was supposed to resolve a tripping hazard issue with the HMD cable ended up being more weird.
    Add to that the lag and other intermittent issues that were hard to diagnose as either tracking or wireless lag and we ended up removing it.
    If I wanted wireless VR, I'd go with the Quest or other all-in-one solutions.

    Edits: Some clarifying language, grammar and spelling problems. 2AM me should not be typing.
     
    Last edited: 13 Apr 2020
    adidan, Pete J, wyx087 and 1 other person like this.
  17. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,889
    Likes Received:
    1,156
    That's a top post and a good summary of the state of affairs. Got me thinking about the film/media side of things.

    I wonder whether VR will diverge into more gaming-specialized and media-specialized hardware. The idea of watching 3D movies in VR is fantastic, because there's no other way to get domestic 3D, and probably never will be again. 3D TVs are dead and they're officially not coming back. 3D blu-ray has pretty much gone now. It flashed in the pan and that was it. Which sucks, because for all its faults as a piece of storytelling and writing, Gravity in 3D at the cinema was probably one of the most gripping and amazing experiences of my life. I actually couldn't believe how well it worked, how good it looked, and how much 3D transformed the experience. I've tried to show people the film on blu-ray at home since; even on a 55" OLED, it's fairly underwhelming. That film, more than any other, belongs in 3D. Avatar was the other big gamechanger - I went back to see it two more times, because it was so gorgeous in 3D. The film itself is okay, passable, enjoyable. But the 3D in and of itself was just a thing you wanted to keep seeing over and over again. Cinematic 3D, executed properly, is a different quality of experience to regular cinema - not the same, not entirely overlapping or compatible, but worth having for its own sake.

    But here's the rub. To get an IMAX sort of experience in VR, you need equivalent of HD or better resolution on a virtual screen, at a virtual distance from your point of view in VR, occupying only part of your virtual field of view. Probably about 70 degrees? Which means the total resolution of the headset would need to be more than double that to make it comfortable and for the pixel density to work out right. The virtual cinema screen, occupying about half of the headset's screens, needs to look as sharp as a real TV or cinema screen. To look as good as a big 3D TV in a living room, the headset would have to be at least...4k in each eye, just to ballpark it?

    That's actually plausible. It doesn't take long for screen tech to develop, it's only ever been market interest and investment that limited screen panels. Maybe we can get 4k OLED in each eye.

    So then you've got your amazing HD headset that can do VR cinema properly. You can finally watch 3D films as you saw them at the cinema. But - but.

    Do you want to use that headset for VR gaming too?

    It's a pair of 4k monitors. You want to drive them at at least 90hz, more if possible. How far into the future does that delay us, in terms of the development curve of graphical processing power?

    We can currently manage about 60fps on a single 4k screen, ish, in a typical triple-A game in max settings. With bleeding-edge GPUs that cost about £800. Assuming they follow a normal sort of progression, that level of GPU power might be £400 by next year, and by 2022 we should have cards that can manage 90fps on a single 4k screen, for about £800-1000.

    Fast forward another two years, 2024. 90fps 4k is now £300-400. Bleeding edge can do about 200fps for £800-1000. But how would this translate to the dual 4k requirement of our imaginary VR headset? Performance doesn't scale linearly with resolution. If a card can do 200fps at 4k, can it do 100fps across two 4k displays? Given all the other latency, processing and synchronizing considerations described by edzieba and d_stilgar, it seems unlikely. Probably more like 50fps.

    Fast forward another two years, 2026. 100fps across two 4k displays is now £800-1000.

    By 2028, that processing power has come down to about £300-400. Normal humans can now start saving their pennies.

    The problem is that this continuous GPU growth is only going to happen with stable price curves if there is ongoing demand for the bleeding edge, which can only come from CAD/workstations/mining, gaming, or VR gaming. VR gaming is not a money pot because adoption is slow, because it's expensive, because adoption is slow, because it's expensive...; non-VR gaming has kind of plateau'd and you can grab a £200 card that does everything non-VR, so the mass of normal gamers aren't knocking at Nvidia's door for an £800 card; CAD/workstations aren't high turnover enough to push an entire market (hence the horrible Quadro prices of yore) and mining is too wildcard to even predict right now, but it's not exactly exploding as fast as it briefly did.

    So there isn't a huge surge of demand for GPU advances in the next few years. They'd happen anyway, inevitably, but they happen faster or slower depending on what's coming out. New releases aren't Crysis-ing the industry. I can run top-end 2020 releases flawlessly on my mid-range 2016 card. And if nothing pushes these new super GPUs into existence, there's no point in the VR boys releasing super high-resolution headsets in the meantime, because the market for them will be absolutely tiny now - most people who want VR already have it, and those who have it won't spend another £1200 just to play Budget Cuts and Beat Saber in slightly higher resolution. And if they're not releasing new VR headsets, why develop bigger, chunkier, more ambitious games for VR? And if nobody's developing bigger, chunkier games for VR, why focus GPU R&D in that direction?

    Stalemate.

    Which, appropriately enough, brings us meanderingly back on-topic for the thread: Alyx. I still haven't played it, but UpIsNotJump informs me that it's absolutely fantastic:


    Valve are so big, rich, carefree, nonconventional and...well, eccentric, that they can actually just go "yeah...there are no big mainstream full triple-A games in VR, so let's just make one." And godwilling, their having just done so might - might give other, more cagey developers the courage to try making a big mainstream VR title of their own. WWII Call of Duty VR? Why not? Tomb Raider 2 & 3 are already VR, spurred on by the conventional 3rd-person game Hellblade: Senua's Sacrifice working amazingly and surprisingly well in a VR port.

    And if big devs start taking VR seriously and investing in it, suddenly people will want higher resolution headsets, and it'll be worth upgrading, and the growing market will bring down the costs of the midrange VR headsets...and then economies of scale take care of the rest. More VR headsets, more GPUs, more games, more headsets, more GPUs.

    So it all depends on whether any other devs take note of Alyx (and, to a lesser extent, Tomb Raider) and start doing stuff - proper stuff, not paddling pool games - in VR.

    If they don't, though, I confidently predict that VR is going to go the way of 3D TVs. If you can't develop and sustain a demand for something, it vanishes. Niche nerd communities buying a few thousand of something doth not an industry make.
     
    Pete J and wyx087 like this.
  18. d_stilgar

    d_stilgar Old School Modder

    Joined:
    11 Feb 2010
    Posts:
    1,041
    Likes Received:
    145
    I don't personally think we'll see a divide in the hardware. I've tried the Pimax HMD with dual 4k displays and it's fine. There are some issues with the FOV and lenses, but I think many of the complaints from people on that front are exaggerated. That said, different people are more sensitive to different things and it's possible that it is "that bad" for those who have complained about it. I guess I'm just saying that there's a whole range of HMDs already with different strengths and weaknesses, so people can decide what they want now and will continue to be able to in the future, but there's not a strong enough difference between what would make for good movie experiences and good gaming experiences to see a hardware differentiation, and the movie side of it is so content-limited that there's not going to be a market for specialized software in the foreseeable future, if ever.

    On the subject of IMAX, I'm reminded of the mostly-defunct IMAX Dome/Omnimax screens that were a huge dome that would wrap around you. I actually think that 180 degree 3D video on a virtual dome could be okay. It would still be a virtual theater with some position tracking in that regard and the virtual theater could help ground people while in VR, which should overcome at least some of the of the motion sickness issues.

    The issue I see with HMDs and virtual theaters is less about resolution, but pixel mapping. On a TV, you can have 1:1 pixel mapping for as perfect an image as you're going to get at home. On an HMD, you're always going to have interpolation. I think that will be a bigger issue moving forward than the HMD resolution itself. The good part of it is that the HMDs we have now are actually pretty close to the native resolution of the 3D frames, since most 3D movies tend to be released in 1080p side-by-side, meaning each eye only gets 960x1080, and with the interpolation of the pixels, it actually looks a lot closer to "native res" on the HMD than you'd think. So you have a drawback in not having 1:1 pixel mapping, but the virtual screen can be HUGE and the 3D effect is essentially perfect since you're just getting one image per eye and not having to filter out frames via flickering or polarized lenses, both of which darken the image and sometimes result in some ghosting.

    And the content/hardware gap has always been there. The thing with 3D is that a big push was made and essentially flopped. 3D is still being released to Blu-ray, but it's a niche market. I completely agree, however, that some movies are definitely enhanced by it. I remember seeing The Walk (2015) in 3D and loving it. Guardians of the Galaxy was awesome in 3D. The Martian didn't suffer for the 3D at all. Its implementation in films has gone far beyond the old tropes of Pinocchio's nose growing into the audience. It's matured and companies have learned to apply a subtle hand when doing 3D conversions. The only hope we can have now, is that we keep getting 3D movies in theaters and that we eventually get 3D HMD support via Netlfix, Disney+ and the rest.

    On the hardware/content gap, that's always been there. Consoles try super hard to avoid it, but we all remember the day-one release library of many consoles being ten games or fewer. For a brand new device like VR (or any other hardware), you'll always get the hardware in a big wave of availability, then a valley where it looks like no software will ever be there to support it (often called the "trough of disillusionment") and then the software catches up, but only if the hardware side can hold out long enough for that catch up to happen. For VR, we've had years since the 2016 launch of the original Vive and Oculus HMDs, but we've had a steady stream of decent games and dedicated adopters. This is anecdotal, but most of my friends who've adopted VR haven't gone back to what they now call "pancake games." The VR stuff, even when it's not great, is still more compelling than AAA stuff released for monitors.

    I personally have a more holistic approach. The first half of Half-life:Alyx had me wishing it had just been made for WASD and mouse. I actually think an overhaul of the game could convert it in a decent enough way, but it would have to overhaul all of the guns and firefights to revert back to something more akin to HL2. Scrounging for resin with a mouse would be dumb and the gun fights would be so easy with a mouse that it would be extremely boring. So I still think there's a place for mouse and keyboard, and a controller on your couch, and a VR HMD with motion controls. They all have a place. I also don't think VR is going anywhere. Small, interesting, good experiences will have a strong (if small) market for forever, so unless every single HMD manufacturer decides to stop making the hardware, the market will survive on the software side for sure.
     
    boiled_elephant likes this.
  19. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,889
    Likes Received:
    1,156
    What about games development in VR? This is something I know very little about. Is it fundamentally harder, more expensive, or more time-consuming to develop VR games than conventional ones?

    I keep wondering why devs don't get in there and try more stuff, particularly the triple-A studios who clearly, easily could. Is it just lack of imagination? Lack of a culture of experienced VR devs to do the work? I feel like my hand-wavey explanation that they're all "scared" of it is missing some finer details.
     
  20. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    It's not fundamentally different or harder at a technical level, but it changes a lot of decisions that would be made otherwise for 'flat' gaming.
    For example:
    Objects need to be the right size. For flat screen gaming, this matters very little: what looks good in screen may have nothing to do with how big something is in reality. You may make a lightbulb the size of a baseball so it can hang bare from a tall ceiling and look good in screen. In VR, that lightbulb just looks comically obviously wrong.
    Then there's lighting and texturing. Basically little to no baked lighting or shading works in VR properly, because anything that looks even vaguely interesting will be observed close up from multiple angles. Even distant environmental objects immediately look wrong if using baked lighting as you move around a scene and watch the shadows and highlights stay painted on (indeed, 'painted on' is the look a lot of shading and texturing tricks take on in VR).
    Control and navigation also need more than the cursory "throw in classic WASD+M or two-stick traversal" nod than they normally do. Unless you just want to ignore the large portion of the population who have not yet (or outright cannot) acclimatise to Sim Sicknes, you need teleport-movement to be implemented, and that has a whole host of gameplay design impacts you need to take into account (e.g. if you are ever going to be chased, if you need to 'jump' onto or over obscacles or be blocked into an area by the ubiquitous Shin-High Wall of Total Impass, etc).
    Interfaces and UI need to be in world-space. Slapping indicators at the corners of a virtual 'helmet' locked to your face sucks in games as much as it sucks in real-life (Heads Up Dispalys exist for a good reason!).
    Most of this is obvious once you have a HMD on (when developing, working on a flat screen for prototyping and then occasionally donning a HMD does not work, you need to view as much as possible in VR as often as possible, right down to individual artists), and most of this has been worked out over the last few years of early development experimentation in the DK1/DK2 era. Of course, everyone likes to ignore all this data and advice and make the same basic mistakes over and over again.

    As for AAA devs: $#!^£ expensive, yo. AAA games are generic for the same reason Hollywood blockbusters are: that expense (high tens to hundreds of millions requires low ROI risks. Couple that with that even with 'smash hit everyone with a HMD must own it' sale volumes that are still a fraction of a potential single-console-platform sales volume, there just isn't the market to even vaguely pretend to guarantee a return on investment by a publisher.
    That leaves it down to the handful of studios indie enough to self-fund and self-publish, but flush with enough cash that they can basically work on a loss-making project (e.g. Valve). Or to taking a wad of development cash from Oculus or Sony to make a game for them.
     
    Pete J, d_stilgar and boiled_elephant like this.

Share This Page