Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Linux Software

Where Carmack Goes Next 94

Posted by Hemos from the with-the-end-of-q3 dept.
JayZee writes "The guys at Shugashack have word on John Carmack's plans now that Quake3 is finished. He's going to be looking into cyberspace virtual realities, and even better he's going to be working on open source projects like glx much more! " Well, that's a nice mix: Free Software projects and realizing cyberspace realities, combined with a man who can make them happen.
This discussion has been archived. No new comments can be posted.

Where Carmack Goes Next More

Where Carmack Goes Next

Comments Filter:
  • That DOOM admin stuff was cool. It is sort of exciting to me to think that we could be moving to a new paradigm in computing...that a GUI would not be a mere /facade/ to the computer, but that actual actions in the virtual environment translate directly into actions on OS objects. (Plus I watched Matrix twice last weekend)

  • Woo! (Score:2)

    by rde ( 17364 )
    I think the cheap little web cams have some interesting possibilities

    It's innocuous little sentences like that that result in major paradigm shifts a couple of years down the line. This could be interesting.

    1. Sorry for saying 'paradigm shift'
    2. Am I the only one that had to change the character set so I could read this page?

  • Off the rails at last! (Score:2)

    by Anonymous Coward
    "He's going to be looking into cyberspace virtual realities, "

    This usually indicates the end of a productive career (although not, sadly, a lucrative one). I'll be happy to see Carmack buck the trend, but I'm skeptical.

    Virtual Reality appears to kill everything that touches it, from technical standards (VRML) to creative ideas (see n-thousand pointless 3D metaphors for filesystems and the like).

    That's not to say the field is a dead end - I'm aware of some truly useful projects - surgical training tools, and in the UK the fire departments in some cities are equipped with VR software containing detailed building plans. Flight simulators are virtual reality devices that have been around long before the trendy moniker

    But, at the moment useful VR seems to be only done using expensive closed hardware and software.
  • I have always beleived that Carmack would eventually step into the realm of 'virtual worlds'. His engines are simply screaming to be used for all sorts of interesting things, perstent virtual worlds, interfaces...the possibilites are endless. Kudos to him for wanting to try and make things really amazing.

  • Velly interesting! (Score:3)

    by Ratface ( 21117 ) on Tuesday November 23, 1999 @04:02AM (#1510820) Homepage Journal
    This is good news. Since the early 90's, when VR suddenly went from being the latest buzz to being a dirty word, any projects that have tried to investigate new advances in VR tech have either been doomed to budget failure, or have had to cloeak the project in other terms such as 3D visualisation, or such.

    It takes someone like Carmack, who having completed 3 rounds of Quake (if you get my drift), have the time, inclination, clout and reasons to put some heavy duty effort into getting some real cool "3D visualisation" technology into the hands of Joe Schmuck (that's you and me :-).

    .. either that or perhaps the militay will hand us some of it's offshoots :-)

  • Sounds to me like the Metaverse in the making. Snow Crash, here we come. I know that other attempts are underway to make Metaverse-like worlds, but I would give Carmack the highest likelihood of creating something that I would actually want to use. If that's the kind of thing he's talking about.

    --JT
  • "We will be supporting Q3 for quite some time. Any problems we have will get fixed, and some new features m[a]y sneak in. "

    Hehe. Unscheduled upgrades are always a welcome thing.

    --
  • ...in Kansas.

    :>

    - Jeff A. Campbell
    - VelociNews (http://www.velocinews.com [velocinews.com])

  • 'd:>mkdir research' (Score:3)

    by Hrunting ( 2191 ) on Tuesday November 23, 1999 @04:08AM (#1510825) Homepage
    I'm not going to overanalyze Carmack's words and try to figure out what he's going to sink his fingers into next, but I will point out this little line in his plan. A lot of people on Slashdot flame MS as a user environment, as a development environment, and as a networking environment, but Carmack does indeed use NT for the majority of his development (and not because he's locked into it). It just goes to show everyone that they don't need to be tied to one OS just because they use one OS.

    Mark troll, flamebait, whatever.
  • Im with you that this would be very cool and everything but I do think that where carmack is headed is different. This sort of VR involves a (in my view) a 3d model of yourself that you can use to interact with other ppl or programs etc. The point is that you can have whatever maodel you like, and appear however you want. In my opinion webcams take the hole point away from the net in that you are no longer anonymous and cannot create your own image.

  • way to go carmack! (Score:4)

    by md_doc ( 8431 ) <matt@webhoste r s . com> on Tuesday November 23, 1999 @04:13AM (#1510828) Homepage
    First off for those people thinking carmack is selling out and not going to create another game... your on drugs. Carmack is one of the fastest and best programmers out there. Did everyone forget that GLquake was created in one day on a bet that he could not port quake over to using OpenGL in one day?

    Carmack just sees things different then everyone else and he wants to see if its possible to do some pretty sweat stuff. The really awesome thing that no one is even noticing is that carmack and id have million of "followers" if you will. With carmack doing some side work on opensoure projects this means only good news for the world of opensource because now the people who are hacking quake and quake2 and quake3 might think about getting involved in some opensource stuff which will be nice. Also note that carmack loves to do research and loves to just go and hide away from everything and everyone and just code for days and weeks even on things he finds interesting... its definitly fun to do if you can find the time to get away.

    Good luck with research carmack can't wait to see what you come up with. It always ends up being exciting.

    --MD--

  • More Open Source than we give him credit (Score:3)

    by Shaheen ( 313 ) on Tuesday November 23, 1999 @04:16AM (#1510829) Homepage
    I think that Carmack and company are actually more of an open source company than we give them credit. No, it may not be directly dealing with source code, but I think it deals more with knowledge with this company.

    id Software has always pushed the limits of 3D software and hardware acceleration, and they've also given a *lot* back to the community. Yes, they make money off of licensing their rendering engine, but (from what I hear), the id engineers are always willing to talk to you about how they accomplished certain tasks.

    Like a true geek, Carmack is proud to show off his algorithms to the rest of us - and he's not worried that someone's gonna go off and copy it and patent it and make him pay for his own inventions (like some patent-grabbing idiots out there). (BTW - if id patented their rendering algorithms, how much do you think the value would increase in today's market? Two fold?)

    id may not be Open Source in terms of giving all their source code away, but they give a lot of *knowledge* back to the community that keeps them swimming in dollar bills :) That's why I'm willing to put the $40 into QIII, because Carmack and company is a software firm I can trust to put something back into the community after they have made their millions.

    I guess I was rambling a bit much there, but that's just how I feel...
  • Maybe he shoudl check out cybertown.com. THey are an interactive site, that seems to be interested in doing some pretty cool stuff. They are in SF, CA somewhere.

    send flames > /dev/null

  • Quite right. You should use the right OS for the job. I'm a big GNU/Linux advocate, but if Linux is going to win in the mainstream, it will have to win by being *actually* better rather than simply having great potential.

    Despite a lot of philosophical/technical problems with Windows, I still use it as my primary development environment.

    It's got my favourite Dev. tools (Delphi, KAWA), favourite spreadsheet (Excel) and I actually like Internet Explorer for it's speed and user interface. Even if the GUI sucks, it's still relatively consistent and well supported.

    But once those tools or equivalent are running on Linux, it will be a different story....
  • To add to the usefulness of VR, it does have some very good uses. I saw a show the other day on the Dicovery Channel talking about Zoophobia (fear of animals). And how one group was using VR to desensitize the patient to the encounter of spiders. Before the desensitizing the patient could not get closer than 6' from a spider without feeling major anxiety. But after a several hour session of VR where the patient would first push the VR spider into a sink and wash them down the drain to finally touching a fake spider that was hooked into the VR world and actually hold it and inspect it. After the 2 hour session the patient was actually able to touch a real spider and has not had a bout of Zoophobia since.

    Also the Government uses VR to build there ships in VR from the ground up to include every single nut and bolt to make sure that the design is viable. Hopefully John can bring this technology into the everyday usage. With someone with this much knowledge and clout if anyone can do it I believe John can.

  • Some careful words on fame (Score:3)

    by Junks Jerzey ( 54586 ) on Tuesday November 23, 1999 @04:39AM (#1510836)
    Let me add some thoughts for those people who may not be heavily involved in the game industry (or, more specifically, the 3D engine side of the game industry). I'm going to say this in a positive way, without any complaining. Do you know:

    1. Who wrote the incredible 3D engine for Descent 3, released a few months ago? (This game is flashier than Quake III, technology wise).
    2. For that matter, who wrote the 3D engines for Descent 1 and Descent 2?
    3. Who wrote the ultra-fast engine for the soon to be released Slave Zero (a Windows demo was released this past summer).
    4. Who programmed 1989's multi-processor 3D arcade game, Hard Drivin', which was so groundbreaking it even made it into the color plate section of many graphics texts?
    5. Who programmed the 3D Hard Drivin' inspired games Stunts and Stunt Driver, which were released for the PC before Wolftenstein 3D and Ultima Underworld? (These were "drive anyway, do anything" games, not "follow the track" games).
    6. Who conceived and wrote the engines for the following popular racing games: The Need for Speed, Daytona, Ridge Racer, San Francisco Rush, and Hydro Thunder?

    What's interesting is that none of these people are one hit wonders. All of them have stayed in the game industry and made huge contributions. But who do you hear about? Carmack and Sweeney. They may be the golden boys of publicity, but they're not alone in terms of technical prowess by any means.
  • It could be a lot of fun if you e.g. could communicate with the computer through gestures. It would probably not do much to spreadsheets, but gaming and chatting might benefit.
  • After Quake, the next project was supposed to be Trinity, which I thought was to be a departure from the run and gun games id's been working on lately. Two more Quakes later, no Trinity. I'm kind of wondering if the research Carmack's putting in now is proof-of-concept research...

    Anybody heard anything about this?
  • After 'n' years of pushing the envelope, he'd be mad if he didn't take a break!

    Anyway, I bet he comes up with some pretty neat stuff, and then decides to start work on another game, possibly based on that research.

    It's a damn good way to ponder without the stress of everything resulting directly in a game.

    Even if he doesn't, it's not as if he won't contribute significantly to whatever field of 3d wossnames he decides to pursue.

    The rest of 'id' are going to have enough on their hands exploiting the Q3 engine before they need him to pull another hat out of a rabbit.
  • JC had a info about this a long time ago in his finger.

    Trinity was just a name that represented the various things on which he was doing research. It got stuck as his next-generation engine.

    It was not practical to complete all aspects of his research (Trinity) to release their next game.
  • In a previous interview (alas, I can't remember which one), Carmack said that after q3 he wanted to work on a 3D persistant online environment, along with whatever project 'id Software' is working on.
    So it wouldn't be like all those silly VRML filesystems, or whatever. Perhaps a much more advanced FPS in a global, persistant world.
  • Bit of a history note for folks: Carmack has been pushing for a massive virtual world system since development on Quake commenced. Even at the time (as Michael Abrash wrote), he was planning on building the metaverse described in Neal Stephenson's Snowcrash (if you have not read it, do so). Technology is finally progressing to the point where a massive, persistent environment can be created. Carmack probably has the greatest chance of pulling it off since he has the experience, and the time to do so. Here's hoping...
  • Gaming is probably the place where this is least true. Carmack has spent years in perfecting a virtual world. It currently has limited forms of interaction, and is short on non-visual/audio feedback, but it's pretty good.

    Now if he was to put his experience to work creating a game like Ultima Online, that would rock. All the crisp visuals and responsive controls we expect from Quake, in a persistant massively multiplayer environment.

    It'd be interesting also, to be able to, in a virtual world, link the users gestures and facial expressions to their character. It wouldn't even have to look like you. But, wouldn't something like MS Comic chat be more interesting with high-poly nicely skinned avatars whose facial expressions mimicked those of the user (as seen via the camera on the computer) and whose lips matched the words they were 'saying'?

    Many people might be more comfortable with the 'avatar' concept if it actually worked the way we're used to in the real world. Currently, an avatar in chat is just a picture that flashes next to your words. Big deal. But, if you were in something like a fancy lounge, with avatars that could sit down at couches, sip coffee, etc, and saw this 3d world through their eyes and had your words, either typed or spoken, said via the character...

    Then instead of chat being nothing but a bunch of words scrolling down a page, you'd be able to wander around the room(s), hearing other people's words via 3d positional audio, when you were near them. You'd find interesting discussions by looking for large groups of people and listening in, or looking for people you recognize, etc.

    It's not something *I* currently want, but I can see it being a killer app.
  • bThis will never happen. There is no good maping between OS objects and
    physical ones, or "virtual" representations of physical ones. If
    virtual reality is ever used for a main user interface, it will bear
    the same kind of relationships to underlying concepts as the GUI does
    now. That doom-admin thing was amusing, but things like that will
    never be useful.

    You should watch less movies
  • id may not be Open Source in terms of giving all their source code away, but they give a lot of *knowledge* back to the community that keeps them swimming in dollar bills :) Plus, they give all their source code away. :)

    id has released the source to Wolf3D and Doom, and would have already released the source to Quake, had they not promised licensees of the engine that they would not do so until their products shipped. So, when Daikatana ships, um, any day now, we can expect to see Quake source as well.

  • Wow. If I searched around my acres of 5.25" disks at home, I think I would still find my old copy of Stunt Driver. That game was extremely cool. It used polygon-based 3d graphics to let you do loop-d-loops and corkscrews and stuff. It was also completely customizable, so you could make the breaks better, or the road slicker. This was a truely 360-degree game.
  • Carmack has said that he read Snow Crash and said "I can do 90% of that." Maybe now he will :) Hey John, can I have access to those trapdoors?
  • Consider this.... The software must analyze the video from the webcam with probably neural net technology or something akin... While this is going on the system must render the 3d world on the users screen... Then bear in mind that webcams are low res... The bottom line : this is not feasible. It would be easier, less computationaly intensive and cheap to just use ultrasound with sensors attached to hands elbows and whatnot...
  • He should do something USEFUL.

    Like write and GPL a cross-platform, high performance, highly abstracted high-level, 3d API (like QD3D was supposed to be).

    Then invent a mind-control ray, and use it on all game developers in the world to force them to abandon DirectX, and switch to the superior technology he wrote (because the fact that it IS superior, means that nobody will use it).

    Then people will start writing games in the same way Id did Quake3, simultaneous cross-platform development. The way it should be. The way they do it in Heaven.

    I wish I had a nickel for every time someone said "Information wants to be free".
  • I have always thought that if we were going to get good virtual realities like the ones we see in Snow Crash and Neuromancer the engine would come from games, not from weird VR experiments. VR experiments have given us VRML which isn't totally terrible but the way it's built isn't workable for large scale net environments. An internet ready engine like Q3 would be perfect. The net code in q3 is awesome and if anybody can hook it up to a nice interface for wandering around a world instead of just shooting at people it's Carmack.
  • I agree that an OSes and filesystems just aren't a very good fit for virtual reality, they are abstractions by nature.

    To imagine a real application you have to think about e-commerce.

    Suppose businesses could design a 3d level that presented their products to customers, who could wander around thru rooms, looking at shelves, watching simulated demos, and adding items to their shopping basket. The potential for competitive creativity funded by huge advertising budgets is fairly interesting here. You may rue the commercialism, but it's a real application that will happen no later than the widespread adoption of broadband in my opinion.
  • Often very true,
    Most VR projects are either, plain silly or way too over kill, but still some of us are doing some interesting stuff.
    Last week we attended a seminar were some VR people clamed that to do anything 3D on the net you would need 75 Bit, QOS and no latency. This is why the VR community needs people like carmack.
    We are currently working on a free, OS platform and we think that VR does have a future but not as something you put on your head. but something you can run like a game and that is free , OS, multiplatfom, lightweight. and built by a community.
    check out http://www.obsession.se/verse
    We are still very early in development.
  • And maby then I can finally get to "Be John Malchavich" (sporry for the mispelling). HEHE Have you ever seen that movie? its great!
  • > 5. Who programmed the 3D Hard Drivin' inspired games Stunts
    > 6. Who conceived and wrote the engines for the following popular racing games: The Need for Speed,


    The Driving Team. Ah, the memories.

    "Stunts", and "The Need for Speed" was developed by EA Canada.

    I was lucky enough to work on the Need For Speed PSX port when I was a co-op student. (The first one back in 95/96)

    Just a few of the names I remember off the top of my head...

    Hanno - Producer
    Wei Teh - Lead programmer
    Brad Gour - track guru
    Dave Lucas - physics guy
    Daniel Teh - replay / interface

    If I saw the team pic in the manual I could tell you who worked on what.


    To bring this thread back on topic, as have you mentioned, it's a whole TEAM effort to get a great game out the door.

    I don't remember Sweeney or Carmack wanting any of the lime light, instead they want to be "left alone" to do research and programm, but everyone keeps making them into "super-programmers-stars" (which no doubt about it, they ARE talented.)

    To the contrary, they are some of the more publically modest programmers !

    Michael Abrash has a good quote on game development, which Carmack has said as well: 99% perspiration, 1% inspiration.

    Seems to sum it up.

    IAAGD (I am a game developer)

    Cheers
  • One of the Miller? brothers is also working on VR worlds-type stuff, along with Richard Vander Wende, the man responsible for most of Riven's 'feel'. I would love to see Carmacks awesome talents + The Cyan teams awesom talents blended. Riven comes close to being a VR world, but lacks Quake's immersivity I would love to see the two combined.

  • Re:More Open Source than we give him credit (Score:5)

    by John Carmack ( 101025 ) on Tuesday November 23, 1999 @11:20AM (#1510863)
    There have been a few things that didn't have prior art that probably could have been gotten past a patent examiner -- constant Z perspective texturing in DOOM, surface caching in Q1, and the overbright gamma table stuff to trade range for precision in Q3, for example. The patent issue came up at Id a few times until a made it perfectly clear that if the company pursued any patents, I would leave. John Carmack

  • Re:Off the rails at last! (Score:5)

    by John Carmack ( 101025 ) on Tuesday November 23, 1999 @11:34AM (#1510864)
    Hey, >I didn't say "virtual reality"... I tend to agree with your assessment. VR is a term loaded with high-enthusiasm / low-results connotations. We have worked with a few VR companies in the past, and I have always found them to not have finishing ability. So much of the VR world (and much other academic style research) is high concept, but sketchy on the details. Most VR experiences are heavy on the "You are in a virtual world!!!!", but don't spend too much time on exactly what you are supposed to be doing there. Can you poke and prod to find interesting things? What happens when someone pushes you? Can you dodge something effectively? Are the controls linear, or integrated over time? etc. I think that one of my strengths is a blend of idealism and pragmatism that has resulted in good results over the years. In any case, of the half dozen things I listed, I am clearly not going to be able to do all of them, so it may be a moot point... John Carmack
  • Actually, what I found interesting in that line was the `mkdir'. I'm pretty certain that most dos users used `md' instead of `mkdir' (infact, it was a long time before I knew about `mkdir'), and while MS supports both `md' and `mkdir', unix (ignoring aliases' only has `mkdir', I was suprised to see John use `mkdir'. I guess he was either being clear (`mkdir' is easier to guess the meaning of than `md') or he is in the habit of using mkdir from his unix experience (and I don't mean just his recent Linux use (glx)).

    I don't flame MS as a user environment: it's actually not too bad when it's not crashing; but as an admin environment, it sucks (95 and NT4 that is, 3.x wasn't too bad). For a couple of years I actually liked using Windows because it gave me a (pre-emptive) multitasking environment for my dos development. Yes, those dos boxes in windows 3.x were actually preemptivly task switched. The real windows OS was a 32 bit preemptive multitasking `os' (as such) with the windows everyone knew and loved running in one virtual machine (as a DPMI client) and each dos box in its own vm. This actually worked quit well so long as the dos programs were well behaved: security was almost non-existant.

    Windoes is ok, I guess, but it doesn't meet my needs (and I don't like it), so I will never willingly go back to it. The only time I ever use it now is to fill out my timesheet, and that's on someone elses computer:).

  • A lot of this stuff has already been done by the quake mod scene. I remember one project that would link servers together in a web. Enter certain teleporters, and you connect to a different server. Another project would autodownload maps that the client didn't have (didn't quakeworld do that too?). That's two thirds off what you need right there. Add persistent objects, and a global database to keep track of them, and you've got a crude Metaverse.
  • Just in case anyone is having trouble getting to the page mentioned in the article, q3arena.com [q3arena.com] caches ID's .plan updates as well. John Carmack [q3arena.com], Graeme Devine [q3arena.com], Xian [q3arena.com] and the rest of the crew are over there.
  • Actually there is a whole genre of games that does it. The only one I recall is mind storm(?) that came with my sister's quickcam. it worked pretty ok with my bt848 card as well.. nice.. cute :) good for exerciing.. bascially you controlled a space ship with your whole body.. move your body from side to side.. or flap your arms (ahem wings) .. and so on
    --

  • Very cool. I always thought that VRML was invented by people who assumed a world of fat pipes and weak clients (and supported in that assumption by SGI). Quake was made for the real world of thin pipes and ever more powerful clients. VRML's very broken method of handling Levels of Detail is a great example of that: send all the models in one big file - real bright.

    I've been raving about this for a while [internet.com]. Doom and Quake have been the best possible proving ground for system that could be adopted to nearly any situation. If Carmak makes a commercial interactive system that requires one large initial download for the "browser", but has small, easily downloadable enviorments (automatically, so the user is unaware of it even happening) he stands a very good chance of creating the Next Big Thing. Remember, before Jeff Hawkins came up with the Pilot, everyone said that the PDA was dead and that the Newton had buried it. I'm sure one compelling product can undo the years of bad ideas by other people.

    John, please consider B-Rep geometry with streamed branching and primarily procedural textures with few bitmaps. And a global "real estate" system, so all of the worlds can link together. This could finally be the successor to the web.

  • Somewhat ironically, a lot of the shoulders upon which Carmack stand currently work for Microsoft's R&D.

    So very true. Bill apparently gave Nathan a blank check to hire the best people in a number of different fields. Whatever his other sins, Billy did manage to nearly re-assemble the graphics group of NYIT. He has Alvy Ray Smith, Jim Blinn and Andrew Glassner (if you don't know these names, you don't know CG) all in one department. Interestingly, the rest of the CG pioneers seem to be working for Paul Allan's Interval Research - Dick Shoup (who wrote the first paint program with Alvy) and Alias' Gavin Miller.

  • Of course Abrash used to work at id. I used to talk mabrash@idsoftware.com with him. Wanna say 1995?

    Pan
  • A MOO with a 3d engine? You may want to give EverQuest a try, unless you have a social life and plan to keep it, that is.

  • I guess this is a bit off-topic, but while we're talking about driving games does anyone know what Sean(sp?) Southern is up to these days? I had so much fun playing the old Lotus and Supercar games, and I last remember seeing his name in the credits for a cool rally game on the PC a few years back.
    --
    make clean; make love --without-war
  • Funny, I flame MS mostly as a user environment. My productivity drops by three quarters when I move from my 6-year-old SGI Indy to my Pentium II 300 running NT, mostly because I don't have immediate access to my customized scripts, aliases and hotkeys.

    I know that much of this stuff can be had in the form of UNIX tools for NT, but I cannot be arsed to get them. My goal is to get off NT as soon as I possibly can.
  • I agree with a lot of the comments that John Carmack and a few others have the best position from which to develop useful VR environments that do something someone wants done. I'd be really be interested in how/where we can get involved - from a couple of perspectives.

    Would the work be inside of ID or as part of a new startup? I don't think ID is publicly traded so I can't invest in them and get filthy rich (PS: my corporate nanny firewalls their site and tunneling out and lynxing from elsewhere gives me "Eeek, No frames support. Please upgrade your web browser" --- don't do that).

    I can't code at John's level, but what could less genius hackers contribute to a project like this? I'll finish learning OpenGL and even....c++ if needed. Hey, if ID were hiring I'd even go back into software QA (Oh, God! did I say that out loud!?).

    Think of an open source component to the project. I can see a medium hard time convincing IDers and VC of a total open source vision for a VR project,but open sourcing big chunks and getting the community involved would make easy some things
    that are otherwise hard.

    Hey, he may actually read this. The "John Carmack" that made a couple of posts a half hour ago doesn't have a PGP key in his user page, but it does sound like him.

    garyr
  • Not really. I used to have a demo called EyeCatcher or something like that running on my seriously under-powered SGI Indy. Its motion analysis was good enough to let you control your character on the screen, and probably the entire computing power of an SGI Indy wouldn't ammount to a hill of beams when lost in the sloshing masses of power of modern PCs.

    I don't think using a camera as an interface would cause any serious overhead.
  • its a good thing smart people like carmack actually TRY these things rather than just assuming its impossible :)

    and no, you dont need neural nets necessarily. You dont even know what hes trying to do!!!

    Imagine a setup where the web cam looks at your face, if you smile, the texture on your model smiles, something like that. That is certainly doable.

  • he says:

    I have two rendering technologies that I intend to write research engines for.

    This seems very interesting. John, if you are reading, will you throw us any scraps about these babies, are they speed optimizations, quality optimizations, both, or other? Also, it seems like a lot of people are speaking for you in this story thread, are they even close to what you are thinking?

  • It would be easier, less computationaly intensive and cheap to just use ultrasound with sensors attached to hands elbows and whatnot...

    Dunno about you, but I'm not gonna be attaching a sensor to my whatnot, no matter how much cheaper it makes things!


  • can we say 'vaporware'?

    Man, you have a long memory, whoever you are. I salute ya. I was going to post a remark about Trinity myself and you beat me to it. Grrrrr! hehe
  • (PS: my corporate nanny firewalls their site and tunneling out and lynxing from elsewhere gives me "Eeek, No frames support. Please upgrade your web browser" --- don't do that).

    Actually, it's good advice...more recent versions of Lynx list the links inside of the frame, as well as the idiotic message the too-lazy-to-write-HTML-4 webmaster put there instead of doing the right thing. Not that it applies in this case, as http://www.idsoftware.com/ [idsoftware.com] doesn't have any frames on it, and it perfectly navigable with Lynx. Of course, the webmaster should have put in alt=" " on all the non-linked gifs. But for the most part, the Id team seems to understand the value of not turning anyone away...or at least that some people are going to get Quake for a new platform even before they get a web browser. I mean, there have priorities.

  • This man by himself is responsible for the outcry of violence in todays youth.

    Read some history - up until a hundred or so years ago, one of the most popular forms of public entertainment was executions. Hanging, pressing to death with stones, eviseration, beheading, burning - can you imagine the sound of someone being burned at the stake? The smell?

    According to your logic, all of the kids who saw this - and it was considered a family outing - should have grown up and turned into serial killers or worse.

    The truth of course is this: violent crimes in the US have declined. The worst school killing was not Columbine, but in the 1930s in Minnesota...what video games did that guy play? What has changed is the reporting. 24-hour news sources have to keep the pipe full, so things that might have not made the national press before now remain in the public eye for weeks.

  • You mean you guys don't put set prompt="d:>" in your .cshrc?

    JMC

  • Actually, it's good advice...more recent versions of Lynx list the links inside of the frame, as well as the idiotic message the too-lazy-to-write-HTML-4 webmaster put there instead of doing the right thing

    true, but... I couldn't telnet to home because
    corporate firewalls ssh (I'm going to get mad at
    them one of these days) so I had to telnet to
    one of my domains at pair.com and use their
    installed lynx to take a quick peek at idsoftware
    and see if I can buy stock. I can't upgrade their
    lynx - or rather an email would probably get them
    to do it (they are pretty responsive) but I just
    wanted to peek at the "corporate info" page and
    it's not worth the hassle. I'm lazy, sue me.

    The sort of "unusual but it happens" scenario that
    nobody thinks about when they build pages that
    just flat break if your browser doesn't support
    feature 'X'. Not that I haven't done it myself
    after banging my head on the desk and chanting
    'No No No' at guys in ties hasn't worked...

    this has nothing to do with The Topic at Hand though.


    garyr
  • Why is it that whenever people are confronted with a problem that can't be solved with a perl one-liner they immediatly think "it must need a neural net." WTF? It's always either "neural net" or "quantum computer" or "fuzzy logic."

    Ryan
  • I HATE using Windows, it's totally counter-intuitive to me now that I've been using Linux exclusively at home for upwards of a year now. But I must say that Visual Studio is a pretty sweet development environment, which is why Carmack chooses to continue to use it. I primarily use InterDev at work (we're all NT on the web side, which is where I do my thing), its youth definitely shows (it's only at ver. 2), and it's got some pretty ugly bugs, but overall, it's got some great features that I don't think any IDE in Linux has yet.
  • on the subject of virtual worlds when is a major FPS game going to model natural environments instead of re-hashing interiors. Now i realise that most are leveraging on their current expertise in modelling architecture/code etc., but you can only go so far with interiors. When is the bar going to be lifted? Exteriors was one of the first things I looked for in Doom. Fighting imps in the open. Nicely done by the way. But that was '93. When I checked out Q3 screen shots..., hmmm not much improvement from what I could see.

    I dont think I'm being overly critical here. The interiors have been done to death (and done extremly well at that). Time for the radical leap from order, precision of geometric lines to fractal geometry. Now is a good time to experiment with exteriors, L-Systems, fractals and parametric equations... all that messy math...but I will bow to coding god who transforms virtual reality from the currently manufactured look to a natural one.

  • Re:Hmmm (Score:1)

    by ed__ ( 23481 )
    because neural nets are the second best way to do *anything*. followed closely by genetic algorithms.
  • I think the publicity around Carmack comes from the impression that he has a hand in almost every area of development of id's games. Most of the games you mention come across as team collaborations.


    I can only answer a few of your questions, and guess at the others - for 1 and 2, the work was done by a programming team. There was an excellent article covering the development of Descent 3, with some details about D1/2, at GamaSutra [gamasutra.com]


    3. I don't know much about Slave Zero.


    A team from Atari programmed that engine, but perhaps more remarkable were the Atari ST/Amiga ports, which used the same processor (68000), albeit only one of them. These ports were done by a German fellow (Juergen Dietrich?) in a short space of time, and ran remarkably similarly to the original. An urban legend floating around says the same guy had ported the Star Wars coin-op to these 16-bit platforms from memory!


    Number 5 is easy - Geoff Crammond of course! He had also programmed the classic "Revs" on the C64, and went on to program F1 Grand Prix for Microprose, along with it's sequel. Not sure about the game "Stunts" though.


    I assume #6 were team collaborations at EA, Sega, Namco and Atari. Could you enlighten me?


    Daniel.

  • Magnetic Fields (Shaun Southern's game development studio) do the popular Rally Championship series. The latest one is due this Christmas.

    Daniel.
  • The makers of the driving games you mention refined an existing game genre. Carmack et al re-defined a genre all their own, and escalated the capabilities of this genre with a speed and enthusiasm that matched that of their audience. This was a first, and made a lot of gamers out of typically non-gamers (like myself).

    The driving games aren't hailed as loudly, because they didn't make the impression that Wulf/Doom/Quake did. Driving games appeal to a narrower audience, regardless of tech innovation. Sad but true.

  • Again, before you go spouting off about what tech would be best for VR, you have to consider what you want the VR to do for you. I mean, if the focus is on interactivity, what exactly is the point of a large outdoor area? Sure it looks pretty and lives up to an ideal of a "virtual world", but pragmatically, you want the things you interact with to be at your fingertips, not thousands of virtual miles away. What interactive VR really needs is a good interface for immediate surroundings- traveling to different locations need not be a "see building in distance, walk for several hours to enter building" but rather "think of building, select it, teleport to location." A massive RPG, on the other hand, where aimless wandering and exploring is the whole point, would demand a more expansive terrian engine.
  • Actually, if you're looking for a well implemented third-person view game, Heretic II from Raven Software, using the Quake 2 engine is the best I've seen.

    Tomb Raider's camera shakes around too much for an action game, making it hard to see where you're going. If you're running next to a wall for instance, and turn away from the wall, your viewpoint tried to maintain a constant distance from you, can't because of the wall, and shifts off to an awkward angle. In Heretic II, the camera is linked much more tightly the direction the character looks, and it always gives a usable view, even in the middle of jumping and fighting.

    But, yes, an Avatar for something like a chat room would probably shift to third person view when you're talking, and to first if you were going to navigate the world.

    The Tomb Raider engine isn't much of an example of anything except the power of pixelated tits to sell an obsolete engine with a repetetive game.

  • Well the MWI (mouse-window-icon) isn't that great either. Really, how much stuff do you actually DO on a desktop? The more stuff you try to fit into the metaphor, the more metaphor shear you get.

    Yes, DOOM admin was amusing in a "useless" sort of way, but there are some /real/ issues involved in 3D interfaces. For instance, in a 2D interface everything has the same visual priority. Screen space is precious, and useless things take up just as much space as useful things. Also, in a 2D environment, there is a limited number of ways you can "drill-down" to get to info. Everything must be encapsulated in a higher level of abstraction ("My Computer"->"Hard Drive C:"->"Folder 1", etc.). The hierarchy in the interface is entirely linear. It might be usefull to have a hyperbolic sort of hierarchy and interface, whereby more "important" things get a larger focus. This /can/ be done on a pure 2D interface, but since 2D is only 2D attempting to show multi-dimensional hierarchies gets very ugly. These are all valid human-interface problems. Scientists work and write papers on this stuff. Although I'm not suggesting it is a "cure-all", we should be moving in new directions now that we can. Who knows...if it hasn't ever been done before how do we know it's not better?

Slashdot Top Deals

He has not acquired a fortune; the fortune has acquired him. -- Bion

Close