Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Games

John Carmack Left id Software Because He Couldn't Do VR Work There 146

An anonymous reader writes John Carmack left id Software last year, more than 20 years after he founded the company. There was a lot of speculation as to why, and now an interview at USA Today provides an explanation. Carmack had become Chief Technical Officer for Oculus VR a few months prior, and he was excited about bringing virtual reality gaming into the mainstream. Unfortunately, he couldn't get id Software's parent company, Zenimax, onboard. He'd hoped they would 'allow games he worked on to appear on the Oculus Rift headset. Had the deal been consummated, Wolfenstein: The New Order — an upcoming sequel to Wolfenstein 3D, an early id release — could have been part of the Oculus' tech demonstration that earned raves and awards at the recent Consumer Electronic Show.' Carmack said, 'But they couldn't come together on that which made me really sad. It was just unfortunate. When it became clear that I wasn't going to have the opportunity to do any work on VR while at id software, I decided to not renew my contract.'"
This discussion has been archived. No new comments can be posted.

John Carmack Left id Software Because He Couldn't Do VR Work There

Comments Filter:
  • Best of luck, John (Score:5, Insightful)

    by Anonymous Coward on Tuesday February 04, 2014 @08:13PM (#46157605)

    If I had to make a list of people in the gaming industry who could make VR gaming a reality, John Carmack would be at the top of the list.

    Good luck, John! We're all rooting for you.

    • by Anonymous Coward on Tuesday February 04, 2014 @08:27PM (#46157707)

      John Carmack + Gabe Newell + Oculus Rift = HL3

    • by Anonymous Coward

      He made 3D gaming a reality, which is like VR without a helmet, and that's already better than VR.

    • by Anonymous Coward
      Michael Abrash would be up there on my list. He has written some very nice posts on his blog [valvesoftware.com] that are simple to understand, and illustrate some obstacles we'll need to overcome to create the best HMDs we can.
    • What we need to do is get Carmack together with some military historians and have them get their grognard on.
    • The world needs Oculus Rift more than it needs another Wolfenstein game. And your comment is true, John Carmack still does have the imagination in technical matters to be up to this task, especially if he is passionate about it like he is.

      Carmack said long ago he planned to actively develop well into his 70s and I believe it.
    • Yeah he did some great programming back in the mid 90s.

      • by Zmobie ( 2478450 )

        He does great programming now. One of his panels at Quakecon last year he talked about doing Wolfenstein in friggin Haskell just as a research project....

    • by Zmobie ( 2478450 )

      Honestly, I wasn't that sold on anyone being able to do decent VR, but hearing him at Quakecon last year he was very excited about it and had some very interesting discussions and ideas. I can't wait to see some demos that the general public can play around with because with him working on it I have hope it has a real chance.

  • Someone with his talent could do so many more interesting and innovating things... Oculus is just one of them, who knows what the future holds?
  • by PhrostyMcByte ( 589271 ) <phrosty@gmail.com> on Tuesday February 04, 2014 @08:31PM (#46157737) Homepage

    Zenimax not wanting their prized programmer to spend a lot of his time working on promotional material for his other business seems reasonable. I don't fault them for it, nor do I fault him for leaving to work on another passion.

    Two things had become constants at id: the lack of interesting games, and the boundary-pushing tech. Lets be honest, the only thing at id that kept it notable was Carmack. And I say that with a crushed, broken heart, as one who's run a TF server, mastered the trick jumps, and played thousands of rounds well after Quake was out of its prime.

    Carmack leaving id for Oculus will free him from the constraints of a big business and allow him to inject some of that coding genius into yet another promising, young, experimental industry. This is exactly where we need him, and where he'll be able to thrive.

    • Thing is (Score:5, Insightful)

      by Sycraft-fu ( 314770 ) on Tuesday February 04, 2014 @09:28PM (#46158161)

      Their tech really didn't push boundaries that much, at least not usefully, in recent years. The measure of pushing forward with game engines isn't coming up with something new that doesn't work all that well on modern tech, it is coming up with new methods to make things look more real with existing tech. To make things work better, faster, etc.

      So sure, the whole iDTech 5 "megatexture" thing sounds cool... But when you see it in practice it is less impressive than procedural techniques from other engines. On top of that, it requires server class hardware to build maps, whereas other engines feature tools that work on regular systems. Same kind of deal with iDTech 4's lighting model. Ya everything comes from a real light source is neat, but lacking radiosity or other kind of global illumination it ended up only working well at being dark and having extremely hard shadows. Other engines gave much more realistic looking lighting, even if the math was technically less correct.

      To me, it seems like they've been too interested in playing around, and not in delivering useful products. Not that playing around isn't fine, but if you are going to make and sell games and game engines, you need to focus on delivering a good product.

      Hence why iDTech 4 and 5 saw next to no licenses but Unreal Engine 3 saw hundreds. It had good tools, a good workflow, and looked damn good.

      It's sad too because clever tricks to make things look better, even if it wasn't the "right" way of doing things is what made iD famous. Doom was a sea of compromise. It didn't actually have a 3d map, just height information, did clever tricks with the limited pallet to get distance fade, used shortcuts to make the math work fast enough on systems with no coprocessor and so on. Net effect was it looked better than people thought you could make a game look on the hardware of the time.

      Now we have things like Rage. iD can crow on all they like about the technology, doesn't change the fact that Frostbite 2 (Battlefield 3) looks WAY better in actual operation and scales better too.

      • by 0123456 ( 636235 )

        So sure, the whole iDTech 5 "megatexture" thing sounds cool... But when you see it in practice it is less impressive than procedural techniques from other engines.

        I wouldn't exactly call 'textures are all white when you turn, until they load in a few seconds later' merely 'less impressive' than other game engines.

        • Well, if you have a properly CUDA-enabled system you don't see that part...

          • Perhaps, but what did occur on every system was the most horrifically jarring texture popping. Looking away from a surface and then looking back (without changing position) was enough to flush the texture and have it reloaded so it takes a second or two to hit maximum resolution. That is a terrible way to handle texturing.

            Our eyes are most sensitive to motion. Geometry-pop is bad because a few pixels change in a way that is inconsistent with the model. Texture-pop means that lots of pixels change in a way t

            • by am 2k ( 217885 )

              What probably also didn't help was that it's a first-person shooter, which is a genre where you tend to turn quickly a lot.

              I guess it would have been fine if the caching would have been less aggressive on the view-dependency. I don't understand how id thought that this level of quality was acceptable for release.

              • I don't understand how id thought that this level of quality was acceptable for release.

                Nor did I.

                That said, as a longtime admirer of his pioneering work, I'm pleased he's on to a new and exciting challenge. Thanks for the memories, John.

        • Lol, this is soooooo true. No matter how great the rest of RAGE was, this was bad enough that the game was unplayable.
      • Re:Thing is (Score:5, Interesting)

        by PhrostyMcByte ( 589271 ) <phrosty@gmail.com> on Tuesday February 04, 2014 @10:46PM (#46158709) Homepage

        Their tech really didn't push boundaries that much, at least not usefully, in recent years.

        The distinction to make is that it was poorly applied. That doesn't mean it wasn't there. id tech 4 and 5 were examples of id taking Carmack's latest idea and running with it full stop, even if the tech wasn't ready.

        Other developers eschewed these technologies in favor of older ones, because they had the focus to pick tech they could apply immediately and successfully to fulfill their vision. id didn't have this focus, and the games clearly suffered as they made the games to suit the technology. The so-called "tech-demo" syndrome that everybody uses to describe the latest id games.

        Eventually those technologies made it into other games. Per-pixel shading is all over the place now, but still alongside lightmaps. Megatexturing is so compelling that support for it is built into the latest graphics standards, so that games can use it properly and without putting in the monumental effort that Carmack did.

        You can't say that he wasn't pushing boundaries. Come on. It's all right there. The games were failures, and other engines look better in many aspects, but the tech was there and it was ahead of its time.

        • by epyT-R ( 613989 )

          I wouldn't call doom 1/2 and the first three quakes failures..

        • by Xest ( 935314 )

          "The so-called "tech-demo" syndrome that everybody uses to describe the latest id games."

          It's not just the latest games, it started with Quake 3. Why do you think Quake 3 was the first game devoid of any proper single player? Not because they wanted a multiplayer only game, but because it was the quickest and cheapest tech demo they could build to get at the real money - their engine licensing fees.

          Quake 2 was really the last game from id that was about the games, rather than about the engines, and that was

          • Very good points.

            I notice you don't mention Willits in your pantheon of id talents. Wisely, I'd say. His design stewardship of Rage -- and its resulting perfect banality -- was as much a lodestone around its neck as was one more advanced yet performance-lacking Carmack technology. Put together, the mixture was lethal.

            There was an id fantasy game in the works, briefly, a decade or more ago; it was killed off in favor of another FPS retread. I wonder what might have been if a) the faction that wanted to k

      • I was a big fan of doom but have been hooked on WoT for the last couple of years, their engine is good but nothing special these days. What makes WoT a beutiful game to look at is the art, good art art is more about fooling the eye than it is about faithfully reproducing details.
      • I think Rage is one of the best shooters of recent years. I don't get why it's not well received. It's fluid and much more interesting and involved, and offers more gameplay than any Call of Duty. AI and character animation are also some of the best, if not the best, I have seen.

      • by DrXym ( 126579 )
        I really wanted to like Rage but it spent about an hour playing it and realised it just wasn't an interesting story (generic apocalyptic fare a la Fallout or Borderlands), and the engine for all its claims was delivering a largely linear FPS experience with a bit of driving thrown in. I think Fallout 3 / Vegas were better games within a similar setting.
    • by hermitdev ( 2792385 ) on Tuesday February 04, 2014 @09:30PM (#46158181)

      Zenimax doesn't need Carmack. Zenimax probably doesn't want Carmack. Zenimax is about pumping out products. Look at the poor state Fallout 3/New Vegas were released in, as well as Skyrim. These are some of their premier products and the released them so buggy as to be near unplayable (Fallout 3 was the best of the lot, New Vegas on the 360 would routinely hang after 15 minutes). Carmack is too much of a perfectionist to fit into such a culture. He's fine delaying a product for years if it's not ready (at least technically, let's face it, he's not about the content/design/story).

      Both parties, Zenimax and Carmack, are probably inwardly happier for the separation.

      I like Zenimax games, the stories, but they've been lacking quality of engineering. I had hoped that with the acquisition of id that the quality of engineering might have rubbed off, and with Carmack's departure, I'm disheartened about it.

      • >>Look at the poor state Fallout 3/New Vegas were released in, as well as Skyrim.

        Which were better than Morrowind, which was better than Daggerfall. But that's specifically Bethesda Softworks (except FNV, which was made by Obsidian), a company with only a passing interest in QAing their code.

        Obsidian actually did spend a lot of effort testing FNV (a friend of mine is near the top on the credits list), but it still had a ton of game breaking bugs at first. In part, it's because the Gamebryo engine is a

        • Id's engines were and are much less buggy, but even still Rage was unplayable at release and I haven't tried it since.

          I bought it at (pre)release on PC and never had a problem. Played it all the way though and never had any issues either. Done a few reinstalls as I updated my PC, copied savegames between PC's in different places I was commuting between and never had any issue there either, I wish more games were like that.

          The question is, if you actually paid for a copy at release why have you never tried it again since. Since it was DRM'd on steam (even if you bought a retail copy like me) you could not actually have sold

          • I got a copy for the Xbox360, and it was basically unplayable.

            I didn't buy it for the PC due to the bad experience I had on the console.

            (And don't call people pirates unless you know for sure. Check my Steam profile, same username as here, to see how many id games I own.)

        • Odd, I didn't have any problems with Rage on the 360, nor much later on the PC.

          One thing that annoyed me about FNV (and Skyrim to a smaller extent) was that my follower would disappear for around 30 minutes (if I stayed overland) after I took a route they refused to follow. If I'd fast travel or enter a dungeon, boom, they'd reappear. It was hard to tell if they were lost of you didn't realize you accidentally killed them. Lots of frustration.

    • Re: (Score:3, Funny)

      by Darinbob ( 1142669 )

      Third thing that was a constant at id: Carmack's ego.

      • What about his super ego?

        • Re: (Score:2, Funny)

          by Anonymous Coward

          What about his super ego?

          That left id years ago to found Ion Storm.

      • by epyT-R ( 613989 )

        It usually takes a strong ego to cut new paths. Insecure types rarely step outside the box because they're unwilling to challenge groupthink.

      • Third thing that was a constant at id: Carmack's ego.

        At least he never promised to make everybody his bitch and then released Daikatana. That event alone pretty much sheltered all the lesser hubris of the classic shooter era...

        • Daikatana was more fun than Doom 3.

          • In the spirit of fairness, I did note that Carmack was coming out of the monster closet by leaving Id. It's almost certainly for the best that he's going somewhere where he can do more focus on tech. It's been quite a while since Id's engine prowess has been enough to carry their comparatively limited progress in game design. Doesn't make the original Doom and Quake any less classic; but even as early as Quake III arena vs. UT(No location-based damage vs. has a sniper rifle) things were getting a trifle awk
      • by Teancum ( 67324 )

        One thing I know about John Carmack is that he gets things done and knows what he is talking about. That he decided to jump ship and go to Oculus VR shows that he is willing to in this case take a pay cut and really does believe in the technology he is working for (although I think Mr. Carmack also got a sweet deal with likely stock options worth quite a bit of money potentially in the future if it works out).

        If anything, his ego is smaller than the actual performance he can deliver. Hopefully he can get

    • by Zmobie ( 2478450 )

      His rendering engines for Id were actually quite advanced. He did a panel on graphics lighting at Quakecon and it was damn interesting. I don't care much for graphics programming (business logic is my favored area), but there was some really cool stuff he was bringing up throughout that panel. Rage had some great engine innovations and work, it just wasn't near as good on the storyboard side imho.

      I do agree though that I think the parent companies were really holding him back. All in all I hate to see C

    • >Two things had become constants at id: the lack of interesting games, and the boundary-pushing tech. Lets be honest, the only thing at id that kept it notable was Carmack. And I say that with a crushed, broken heart, as one who's run a TF server, mastered the trick jumps, and played thousands of rounds well after Quake was out of its prime.

      Indeed. What was remarkable about Quake and Quakeworld was not the single player game (though lord knows I've played it through enough times by myself and in co-op) o

      • by Anonymous Coward

        QuakeC is a terrible hack, which is why they dropped it in Quake 2, but it had several important advantages: almost anyone could pick up the source code and mod it (leading to Team Fortress and then CustomTF), and since it was all run within a sandbox, you could download executable code from the internet and run it on your server without risking compromising your server.

        And more importantly, the mod doesn't have to be ported.
        I used the standard Team Fortress mod with the Amiga port of Quake. Different CPU and different graphics system but it still just works.
        I am pretty sure that all obscure Quake mods works just fine on ARM ports of quake too.

        • Yeah, good point. That was another reason why Quake 2 was a step back.

          Still, I wish Carmack had build a few more sanity checks into his code. Overflowing silently was just a pain in the ass to deal with on CustomTF.

    • Zenimax not wanting their prized programmer to spend a lot of his time working on promotional material for his other business seems reasonable. [...] Two things had become constants at id: the lack of interesting games

      This is why it doesn't seem reasonable. iD games are just tech demos for their engines these days anyway. Without Carmack, those engines will be less impressive. They'll have to start actually making games. The Rift and iD could have been good for each other. Now they won't be.

      • Zenimax announced [ign.com] they would not be licencing the rage engine to outside studios. They also said they wouldn't be using it for Elder Scrolls titles. Very strange they didn't just give John a blank sheet to come up with an open world engine to beat Gamebryo. I guess there must have been disgreements over direction. Big missed opportunity for Zenimax, and the community. Imagine an open world tech built to Carmack quality standards, with Zenimax content.

        • Zenimax announced they would not be licencing the rage engine to outside studios.

          That is daft. Engine licensing kept iD going through times of no creativity.

  • by Anonymous Coward

    This is like the owners of Lockheed Martin vetoing a project that is military in nature and might result in an aeroplane design.

  • Would it be fair to describe leaving Id as 'coming out of the monster closet'? Or have they done something worthwhile recently?
  • John Carmack used to post on /. semi regularly, it would be interesting to see if he chimes in here.

    But really, it makes sense, making 1 game VR enabled is different than making VR a reality, and it sounds like Carmak wants to do the latter rather than the former.

    If I was zenimax I would be worried about Carmack making his next game too dependent on VR tech, which would lock out a lot of the market who won't have an oculus rift right away, and if Oculus rift wants VR support for their new experimental hardw

  • I lknew about him leaving like over month ago...
  • All that attention and hype for idSoftware's new game all wasted down the drain. Zenimax, why?

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...