Slashdot Mirror


Demoscene: 64k Intros At Revision Demoparty

An anonymous reader writes "Last week-end at Revision demoparty, demosceners have pushed further the limits of what can be done in a single 64kb executable file. Using extensive procedural techniques and compression, Gaia Machina (video capture) and F — Felix's Workshop (video capture) are realtime animations, featuring high quality rendering, sound, 3D models, and textures."

141 of 207 comments (clear)

  1. Re:Am I supposed to have heard about this before?? by Anonymous Coward · · Score: 5, Insightful

    Get off my lawn!

  2. The reason you haven't heard about it by tepples · · Score: 4, Interesting
    I think I know why North American readers may never heard of it. (USA and Canada represent well over two-thirds of the population of industrialized anglophone countries.) From the article:

    in Saarbrücken, Germany

    For some reason they never have demo parties like this in North America. Why is that?

    1. Re:The reason you haven't heard about it by beltsbear · · Score: 2

      Well not really. We had demo parties in the USA demoing the Amiga demo's from Germany.

    2. Re:The reason you haven't heard about it by slyrat · · Score: 2

      I think I know why North American readers may never heard of it. (USA and Canada represent well over two-thirds of the population of industrialized anglophone countries.) From the article:

      in Saarbrücken, Germany

      For some reason they never have demo parties like this in North America. Why is that?

      In Atlanta, Ga a friend of mine did a cheese and demo party every year and showed a bunch of demoscene stuff from the past year. There was also a bunch of cheese tasting involved. That is the closest to a demoscene I've experienced in the states.

    3. Re:The reason you haven't heard about it by X3J11 · · Score: 5, Informative

      I think I know why North American readers may never heard of it. (USA and Canada represent well over two-thirds of the population of industrialized anglophone countries.) From the article:

      in Saarbrücken, Germany

      For some reason they never have demo parties like this in North America. Why is that?

      I'm Canadian, and I know of the demoscene and all the related terminology the GP doesn't. While I never had the opportunity to attend a party, I loved watching what some of these groups put out. The Amiga era was a few years before my time; I got into it in the mid-90's when names like Renaissance, Future Crew and Triton were the big guns.

      There are caps of some of the big name demos from back in the day available on YouTube, with Future Crew's probably being the most popular. I still remember the awe I experienced when I first saw Triton's (later Starbreeze Studios) Into the Shadows demo. I purchased their subsequent game, Enclave, because of that demo (and they were the guys who wrote FastTracker 2).

      For those who don't know what the big deal is, way back when PC hardware was pretty crappy these groups were putting out some of the most demanding and advanced programs, stuff that put a lot of what the game companies were pushing to shame. The aforementioned Into the Shadows demo was released in 1995. This is long before 3D accelerators and hardware floating-point math were standard. It really was impressive at the time, and it was being done by groups of kids.

    4. Re:The reason you haven't heard about it by Kozz · · Score: 1

      I was wondering why it's "demoscene" and not "demo scene". I was reading the former as if it were an ancient middle-eastern city mentioned in the Bible:
      "deh-MOS-keh-nee".

      --
      I only post comments when someone on the internet is wrong.
    5. Re:The reason you haven't heard about it by gloom · · Score: 2

      I'd say that the closest the US got to having "a demoparty like this" (meaning: with such good releases and turnout) was NVScene in 2008 which I helped organize. The event was documented in this now-severely-outdated blog, if you're interested in catching up: http://demotrip.blogspot.com/

    6. Re:The reason you haven't heard about it by mrhight · · Score: 3, Funny

      I still have floppies with demos from Future Crew. The GP probably doesn't know what floppies are either. ;)

    7. Re:The reason you haven't heard about it by tepples · · Score: 3, Insightful

      If a program doesn't need a hardware upgrade to run it then it can't possibly be any good.

      Then explain the DS outselling a PSP with a CPU clocked over three times faster.

    8. Re:The reason you haven't heard about it by Anonymous Coward · · Score: 1

      Probably because the only good demoscene group from the USA was Renaissance and they've been gone for a long time now.

    9. Re:The reason you haven't heard about it by sqldr · · Score: 5, Interesting

      For some reason they never have demo parties like this in North America. Why is that?

      Hello, I'm meaty. I coded the synth for Gaia Machina. There were some Americans there. he was saying to me that you're all up for it, but spending months/years working on it isn't an american thing. We spent 3 years rewriting our engine since "ephemera" which pretty much maxed out our last engine. I'm British, the rest of the group is Swedish. There ARE American groups who do stuff. And Canadian - Northern Dragons kick ass. We may even visit an american party with a release just to kick your backsides and get you to do some fucking work :-) There's loads of tracker musicians in America, but you never hear music done in tracker that took more than 2 days to write! You guys have the talent, you have the place, just pull your fingers out and get working!

      PS. Linux port of gaia machina is coming. It already compiles ok. We need to check it first.

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    10. Re:The reason you haven't heard about it by ZahrGnosis · · Score: 3, Interesting

      Out of points or I'd mod you up for mentioning Future Crew. I still have a 3.5" floating around with some great old demos from that era.

      I'm not surprised anyone hasn't heard about the demoscene any more, but it's easy to figure describe: fit the most impressive graphics and sound you can on a very small memory footprint -- 64kb is the common limit now apparently, but I seem to recall some good "anything you can fit on a disk" rules and some impressive 4k demos.

      Other than raw coolness, the point is to "do more with less" -- push creativity, efficiency, and algorithm design by artificially limiting resources. It's very impressive stuff, and having been out of touch for a long time, I'm AMAZED at the quality these vids are putting out. Has anyone been able to run the .exe's to verify? My rig keeps failing them -- I imagine it requires specific hardware and software versions.

    11. Re:The reason you haven't heard about it by sqldr · · Score: 2

      While I never had the opportunity to attend a party

      Get your ass to assembly 2012, helsinki, finland. I'll see you there :P

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    12. Re:The reason you haven't heard about it by sourcerror · · Score: 2

      For some reason they never have demo parties like this in North America. Why is that?

      Because you can't get rich doing it.

    13. Re:The reason you haven't heard about it by RyuuzakiTetsuya · · Score: 1

      While demos from that era looked nice, they had the problem of not having to worry about AI, an interface, asynchronous audio and video...

      Granted if these demo sceners were to build a 3d engine it would probably kick id tech and unreal engine's ass. ;). Although I would watch a John Carmack demo.

      --
      Non impediti ratione cogitationus.
    14. Re:The reason you haven't heard about it by SiChemist · · Score: 1

      The first one Gaia Machina ran (slowly looked like about 10-15 fps) under wine.

    15. Re:The reason you haven't heard about it by SiChemist · · Score: 2

      Gaia Machina ran under wine but was pretty slow (1024x768, windowed, looked like around 10-15 fps). Would love to see it running on my Linux box native.

    16. Re:The reason you haven't heard about it by Albert+Sandberg · · Score: 1

      Many people from the demoscene go on to work for game companies later.

      For instance, Alex Evans (known as Statix, famous demo scener from the 90's) went on to work for lionhead studios (peter le minoux founded, famous from Popolous on the atari/amiga) and created the popular game Black and White... then he went on to found Media Molecule who created little big planet.

      I am a demoscener myself and I did some programming on a Texas Hold'em online game in the early days of poker online, but I went off to work with something else completely after that.

      Game programming and demo programming are in many aspects the same thing. Even if a demo isn't user controlled, it is controlled somewhere along the path and some kind of interactive tools are often used.

    17. Re:The reason you haven't heard about it by sqldr · · Score: 2

      I'm surprised it ran in wine.. we did some REALLY dirty shit to keep the size down, including using windows system calls to avoid linking against stdlib :-) linux version is similar.. we call kernel traps.

      Have 15k of music.. this got rejected for not being jolly enough for the nature theme: http://spacepigs.com/meaty-alternature.exe

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    18. Re:The reason you haven't heard about it by sqldr · · Score: 1

      PPS. I'm open-sourcing the synth in a year's time. You don't get it just yet.. we're not giving our secrets away so soon!

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    19. Re:The reason you haven't heard about it by Adriano+ML · · Score: 1

      Could you specify which GPU are you testing with the linux port? Just curious :)

    20. Re:The reason you haven't heard about it by _Shad0w_ · · Score: 1

      Purple Motion released a CD sometime back, I bought a copy of it for happy fuzzy memories (and because it was good). I think Skaven is kind of still about.

      I think ScreamTracker 3 was FC's last release, if you don't count the compilation they did last year. They fizzled out because a lot of the older members of the group hit national service age (most - if not all - of them were Finish). Futuremark, Bitboys and I think Remedy all had FC members as part of their start-up teams.

      --

      Yeah, I had a sig once; I got bored of it.

    21. Re:The reason you haven't heard about it by JigJag · · Score: 1

      It's Peter Molyneux, not le minoux.

      --
      "The hallmark of humanity is the ability to move beyond sensory inputs" - Mary Helen Immordino-Yang
    22. Re:The reason you haven't heard about it by Ihmhi · · Score: 2

      For those who don't know what the big deal is, way back when PC hardware was pretty crappy these groups were putting out some of the most demanding and advanced programs, stuff that put a lot of what the game companies were pushing to shame. The aforementioned Into the Shadows demo was released in 1995. This is long before 3D accelerators and hardware floating-point math were standard. It really was impressive at the time, and it was being done by groups of kids.

      These kinds of people need to get together and make games, or even better a game engine. Inefficiency is rampant in the programming world today and is especially prevalent in the gaming world. Hell, even our OSes have these kinds of problems. I tried an Amiga at a friend's house, and the OS was snappier and more responsive than any GUI from the past 10 years that I've used.

    23. Re:The reason you haven't heard about it by monkeyhybrid · · Score: 3, Funny

      "demo scene" requires an extra byte compared to "demoscene" ;)

    24. Re:The reason you haven't heard about it by bazald · · Score: 1

      Reread what he wrote. Assuming what you wrote is true, a program running on a DS would need a hardware upgrade much more than one running on a PSP. Therefore it would be much more likely to be good on a DS. QED.

      --
      Insert self-referential sig here.
    25. Re:The reason you haven't heard about it by moogaloonie · · Score: 1

      "For some reason they never have demo parties like this in North America. Why is that?" I will speculate based on my memory of the Amiga era. We'd meet occasionally in person and someone would show up with a bunch of cracked games and other stuff pulled from a BBS... The intros and cracktros had to fit on the floppy with the game itself, which was likely already using most of the available diskspace. It seemed that most of Europe had pretty low-end machines, mostly stock Amiga 500s, and that was what the crackers wrote to. In the US, many more Amiga users were into video and/or 3D. Versus Europe, US users tended to have expanded desktop systems with upgraded CPUs, FPUs, and extra memory. It was well known that demos and cracked software often wouldn't run on our upgraded systems, or there would be NTSC/PAL related timing problems. (Xenophobe for the A500 wouldn't work if you had the 512k expansion as it changed the memory addresses) For this reason above all others, US Amiga users valued OS compliant, multitasking applications with style-guide complaint interfaces over metal-banging demos/games that often required trial-and-error with disabling CPU caches, burst modes etc. just to run. Our contribution was artists like Eric Schwartz creating playable system-friendly animations in programs like MovieSetter.

    26. Re:The reason you haven't heard about it by noodler · · Score: 1

      Bah, you and your fancy VGA's... :P
      Demos are at least 10 years older than 1995.

    27. Re:The reason you haven't heard about it by noodler · · Score: 1

      ""demo scene" requires an extra byte compared to "demoscene" ;)"

      That is why the people in the scene generally refer to it simply as 'the scene' or just 'scene'.

    28. Re:The reason you haven't heard about it by kmike · · Score: 1

      While demos from that era looked nice, they had the problem of not having to worry about AI, an interface, asynchronous audio and video...

      I present to you.. .kkrieger!
      A 3D FPS in 96KB.

    29. Re:The reason you haven't heard about it by SiChemist · · Score: 1

      Just goes to show what an awesome job those guys at WineHQ are doing!

  3. Re:Am I supposed to have heard about this before?? by Anonymous Coward · · Score: 1

    They put links in the summary for people like you.

  4. Re:Am I supposed to have heard about this before?? by Albanach · · Score: 1

    Seven digit UID? Probably not.

  5. Re:Am I supposed to have heard about this before?? by Nyder · · Score: 5, Funny

    Demoscene? Demoparty? 64kb executable?

    Ya, i wish there was a website that you could like, search for the meaning of stuff and maybe websites about it and crap.

    --
    Be seeing you...
  6. If you have to ask why this is cool... by Dr.+Manhattan · · Score: 5, Insightful

    ...you're not part of the intended audience. Admittedly, there's a lot of necessary hardware support to get these kinds of results, but still... full A/V in a space less than the banner image of most websites. Makes you wonder what could be done with similar techniques and, say, a megabyte of space.

    --
    PHEM - party like it's 1997-2003!
    1. Re:If you have to ask why this is cool... by Anonymous Coward · · Score: 1

      These demoparties still have 4KB competitions in addition to the 64KB competitions (and unlimited everything), and also retro competitions (best C64/CPC/etc demo, best classic Amiga/ST/etc demo, ...).

      Some 64KB demos are basically now just "best procedural texture generator + best scene generator" demonstrations, but that's no bad thing.

    2. Re:If you have to ask why this is cool... by virgnarus · · Score: 2

      Well, Farbrausch did release .kkrieger , which is a 96k FPS with Doom 3 graphics. Not much in the gameplay department, but it's still a revolutionary example of what can be accomplished within a small filesize.

    3. Re:If you have to ask why this is cool... by TheRaven64 · · Score: 2

      That's 64KB of code and data, so it's still fairly impressive. Even if you're using a completely off-the-shelf 3D engine, fitting all of the geometry and textures for a virtual environment into 64KB is nontrivial.

      --
      I am TheRaven on Soylent News
    4. Re:If you have to ask why this is cool... by Terrasque · · Score: 1

      Cool, can you give us some examples of programs from that area that had this kind of graphics? I'll even let you pick non-realtime stuff :)

      --
      It's The Golden Rule: "He who has the gold makes the rules."
    5. Re:If you have to ask why this is cool... by AmiMoJo · · Score: 1

      Makes you wonder what could be done with similar techniques and, say, a megabyte of space.

      Grab an Amiga (or emulator) and find out. Amiga floppy disks are 880k, and there are plenty of one or two disk demos. The results are damn impressive.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    6. Re:If you have to ask why this is cool... by CaptnMArk · · Score: 1

      Agreed, from someone that has seen a JIT generate >64KB of code for a function call.

    7. Re:If you have to ask why this is cool... by tusam · · Score: 1

      Keep in mind that even your comment is already 368 bytes in size.
      While pushing random pixels on the screen might be possible (depending on the hardware) with pretty much the same amount of code as hello world would take, fitting anything more complex than that in 64B has never been possible.

    8. Re:If you have to ask why this is cool... by Albert+Sandberg · · Score: 1

      Yeah. If you would model a scene of this 64kb intro in maya or 3d studio max, one strawberry would easily take up 65kb of data excluding the texture. That's how densly "packed" a 64kb intro is.

      Of course they don't store the data, they store the steps on how to create the objects and tweaks along the way. Intros generally have a pretty long loading time to create the data.

    9. Re:If you have to ask why this is cool... by sqldr · · Score: 1

      Gaia machina was exaclty 65536 bytes. we did EVERYTHING to keep the size down, including some dirty tricks. The 4k demos are the new vogue in the demoscene, but there's still groups like us going for the 64k. Most amiga games used all the RAM. my synth in Gaia Machina is 8k, but it uses 150MB of ram when it unpacks...

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    10. Re:If you have to ask why this is cool... by MiG82au · · Score: 1

      Haha, obsolete Microsoft system. Plays the latest and greatest games, and lets me do my engineering work; what more could I want? I have Ubuntu too, but Windows is pretty damn good.

  7. Re:Word Salad by Nyder · · Score: 4, Insightful

    I can't seem to make heads or tails of this post. It's techno-babble and word salad. I guess I should remember this feeling when I talk about programming with my non-programming friends.

    Sort of sad as a programmer you have no knowledge of some of programming history.

    --
    Be seeing you...
  8. Re:Am I supposed to have heard about this before?? by mwvdlee · · Score: 1

    Welcome to the world of computer nerds.
    Do not be afraid to ask; once we too were noobs like you.

    --
    Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
  9. Re:Word Salad by Tukz · · Score: 4, Insightful

    Makes me wonder if he really IS a "programmer" and not just a "HTML/CSS" scripter.
    You would be surprised at how many HTML/CSS monkeys, calls themselves "programmers" these days.

    --
    - Don't do what I do, it's probably not healthy nor safe. -
  10. Re:Word Salad by Anonymous Coward · · Score: 1

    it's talking about how programmers manage to get high quality animations and sound from executables limited to 64Kb.

    they use programming tecniques such as procedural generation and compression to achieve this.

    i could simplify it even more for you if you like? maybe i could write it in crayon?

  11. Re:Word Salad by vlm · · Score: 1

    Sort of sad as a programmer you have no knowledge of some of programming history.

    How else would marketing sell the same old idea as something new? That is about 99% of "innovation" in IT.

    --
    "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
  12. Re:Word Salad by Pseudonym · · Score: 1

    Oh, crap. You just called the demoscene "history". That means I'm old.

    --
    sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
  13. Unbelievable... by RedBear · · Score: 1

    I saw a couple of demo files years and years ago. DOS-based stuff. I think they were probably 16k files. I was amazed at how long the animations and music lasted from a 16KB EXE file. The demo just went on and on, for like ten minutes. Had some fairly impressive animations too. But it was all line-based sorts of things, like old screen savers.

    But this... this is insane. I can't even believe what I'm seeing. I'm downloading the 720p version of the first video in MP4 format, and it's 91MB. A 91MB full-motion video rendered from a 64KB demo file. That's just nuts. It frazzles my brain to even think about how this is possible.

    1. Re:Unbelievable... by Anonymous Coward · · Score: 1

      Demo scene is the perfect example of what real programmers can do. There is a lost methodology of programming that came with C++ and higher level programming languages. the methodology of refining code to use less space and do more using advanced maths and utilizing the hardware, including flaws, to extract the most you can from it. The game 'Mayhem in Monsterland' on the C64 was 68K long, the developers found an additional 1K of addressable memory in each of the 4 ROM chips "where someone had left it".

      With the introduction of high level languages came an overhead problem that was easily solved using the Microsoft method, throwing money at it, buying new hardware, and making bulkier and lazier code, after 20 years, code has become so bloated that Demoscene looks like wizardry to a modern software developer.

    2. Re:Unbelievable... by rev0lt · · Score: 1

      Demo scene is the perfect example of what real programmers can do.

      I'd say demoscene is the perfect example of what can be done, software-wise. I'd guess you've never peeked at a demo sourcecode, or you wouln't call them generically "real programmers" (incidentally, there is usually much talent behind the code, but no structure or discipline whatsoever).

    3. Re:Unbelievable... by Internal+Modem · · Score: 1

      The files are larger than 64K...

    4. Re:Unbelievable... by Shinobi · · Score: 2

      Actually, there is structure and discipline. But it's not academic structure or discipline. It's the structure and discipline of hard limitations and necessity. For example in 4, 16 and 64KiB compos, where every single byte matters, pretty much. In the 4KiB compos, a huge focus is on the packers and all the math surrounding that, where the packer is optimized for two things: Size and speed. And since you're not showing just one or two static effects, but animations, preferably with multiple effects, there's a lot of code that needs to go into that limited footprint.

      In fact, I think demos should be required study in comp.sci etc, especially these limited footprint demos, specifically to teach the role of creativity in problem solving, instead of just wasting resources to sidestep them.

      A prime example is that the demoscene, if we look at it as an entity, pushed the boundaries of real-time procedural content generation before the games industry or academia did, and still continue to do so. The ones in the game industry who do look into procedural content generation often have a background in the demoscene

    5. Re:Unbelievable... by Albert+Sandberg · · Score: 1

      I wouldn't say that's the case of everybody. Yes, some people think that it's not important to make your code nice, but when I code (and many with me) we do it with pride and produce libraries well suited for public release if we wanted to.

      There is a very high bar that we set onto ourselves.

      The back side of the coin is that sometimes, we make demos really fast, and the actual _demo code_ isn't so well structured, but you can be sure the libraries behind it is.

    6. Re:Unbelievable... by rev0lt · · Score: 5, Interesting

      During the nineties, I've done a lot of scene-related assembly code (mostly graphics and infrastructure - extenders, memory managers, etc), so I'm well aware of the limitations imposed and the usual workarounds. Most sourcecode I've seen is an ugly mess, even if it works. There are some true clever and elegant algorithms in the lower categories (128/256b and 4k), but what I've learned since then is that computer processing power is cheap, but software lasts. One of my favourite demos of all time is Heaven7, mostly because it implements a realtime raytracer than runs smoothly on a P200. But probably many of the optimizations that allowed it to run smoothly in such a limited processor aren't valid for a P3. Or a P4 (handcode optimization for any of the P4 lines is a _nightmare_). So, if it was a maintained application, the highly optimized algorithms would probably have to be rewritten, to take advantage of the latest features, and keeping "pushing the boundaries".

      As an example (and more in line with the nineties), a lot of effort was put into highly optimized bresenham line algorithms, because traditional implementations implied a div operation per pixel, and integer division was awfuly slow (like 40 cycles on a 486). So, even if bresenham's requires extra instructions, it would still be a lot faster. Well, on a pentium, not only the div instruction took 1 clock cycle (like most of the other instructions), some instructions could be paired for execution (the processor had 2 execution pipes), so the bresenham implementation is usually a lot slower than using the div instruction. If you had to write maintainable software, would you worry about implementation details that could double your development time, but be obsolete on the next processor to be released? I guess not.

      Demos are a kind of development on its own. They require no user interaction, no external data pulling (other than already packaged resources), have no error checking whatsoever and usually are buggy as hell - slightly different hardware may give you completely different results, or just don't run at all. So, that's why I don't like to call scene coders "real programmers". They are more of a class of "code artists", and yes, they should deserve more merit than "regular" coders, not only because their algorithmic skills, but also because of their creative way of implementing them.

    7. Re:Unbelievable... by rev0lt · · Score: 1

      I guess I was misinterpreted. I regard scene programmers more like "code artists". The software product itself is unique, in the sense that is not usually maintained or ported. Choosing long-term algorithms and performance compromises is usually not a problem, since it is made to be awesome with a specific combination of hardware. The algorithms have to be fast enough to implement the desired funcionality on a specific point in time (the party or the release date) and usually on a specific processor, not on hardware that will exist 6 months from now.

    8. Re:Unbelievable... by MiG82au · · Score: 1

      Look harder... GaiaMachina.exe 64.0 KB (65,536 bytes)

    9. Re:Unbelievable... by Qwertie · · Score: 1

      Interestingly, optimizations that are "no longer valid" on new x86 CPUs are typically still valid on ARM CPUs, so don't count out those old techniques yet. The ability of x86 desktop CPUs to do fast floating point and division depends on having millions of transistors, dedicated to those tasks, running at high clock speeds. Since ARM is intended for low-power markets, it does not have an integer divide instruction, some ARM CPUs don't have a floating-point unit, and those that do floating point can't do FP as fast as they can do integer math. Similarly, very deep instruction pipelines are not available in ARM. Consequently, the performance characteristics of ARM CPUs are more similar to the oldest Pentiums than to the newest ones. And of course, although demos targeted at old CPUs don't use new CPUs in the most ideal way, they still run much faster on new CPUs than they did on the CPUs for which they were intended. Thus, the old demos should still work perfectly well.

    10. Re:Unbelievable... by rev0lt · · Score: 1
      Modern ARM (v7) cpus already have an integer divide instruction (and many other low power architectures already have them, including modern versions of traditional 8-bit microcontrollers and microprocessors, such as Z80). Most of the optimization tricks you'd use on x86 aren't very good on a RISC platform (I have no experience with ARM; but I've played with Intel RISC processors). Versions with SMID support are widely available, and it is a far more important feature than FP for most intensive algorithms. Also, most devices already have some form of GPU, so I don't expect to find OpenGL compatibility, as an example, implemented in software.

      I honestly do not know how deep pipelines relate positively with performance (during the P4 GHz wars, AMD used a much shorter pipeline, so it could get the same performance at a fraction of the frequency - and that's why a single core of a modern [pentium-m descendant such as core 2] 2.0Ghz processor blows out of the water any >3.0GHz Prescott).

      And of course, although demos targeted at old CPUs don't use new CPUs in the most ideal way, they still run much faster on new CPUs than they did on the CPUs for which they were intended.

      Yes, but all the cleverness of the algorithm optimization is somewhat lost, isn't it? If the result isn't faster than the reference implementation, why waste time optimizing it at that level? And even if you do waste that time and go the extra mile, how do you feel everytime you need to basicly rewrite all your libraries because the rules of the game have changed?

      Thus, the old demos should still work perfectly well.

      Except many don't. Because some rely on instruction timing for music sync, some rely on "padding" instructions for delay, some rely on vertical retrace to sync the graphics, as an example of many hardware artifacts that were actually used as a feature. Probably since DX/OpenGL demos compatibility has improved, but try to run an old DOS demo on a modern computer. Or one that relies, as an example, on a processor "defect" such as the 386 "unreal" mode.

    11. Re:Unbelievable... by rev0lt · · Score: 1

      replace SMID with SIMD :P

    12. Re:Unbelievable... by rev0lt · · Score: 1

      Read what you quoted. Then read again. got it? People used bresenham because traditional algoritms implied a div.

    13. Re:Unbelievable... by rev0lt · · Score: 1

      Tiny corrections :D
      - Traditional line drawing uses a div per pixel. Check pseudocode at the wikipedia page (http://en.wikipedia.org/wiki/Line_drawing_algorithm). There are some "hybrid" algorithms that implement one division per line, but usually are a variation of bresenham's algorithm with better precision.
      - I'm not familiarized with Heaven7's internals, but it is somewhat obvious that uses tricks. As most other rtr implementations out there. And some non-realtime raytracers. If you can simplify global illumination (using "post-production" effects), scene complexity (by camera angles and removal of invisible objects), and recursion (limit interpolation and reflection/refraction), you still can get quite impressive results, even with "old" cpus, by today's standards.

    14. Re:Unbelievable... by timq · · Score: 1

      From the original post in question:

      ... a lot of effort was put into highly optimized bresenham line algorithms, because traditional implementations implied a div operation per pixel, ...

      Which is worded poorly enough to be taken to mean, "traditional bresenham implementations require a division operation at each pixel".

      I also posit that the quoted post is wrong about the division instruction: DIV still takes a lot of cycles to execute*, that's just the nature of the maths involved. Of course modern processors will try to do some clever code reordering that may make slow instructions appear to be executing a lot quicker, but the instruction dependency chain does not always afford this opportunity.

      * cf for example http://www.agner.org/optimize/instruction_tables.pdf

    15. Re:Unbelievable... by rev0lt · · Score: 1

      English isn't my native language, but it seems the problem lies with people thinking that breseham's implementation is "traditional". It's not. To be blunt, the base algorithm isn't even integer based and envolves a div (to calculate the slope), so, by all metrics, the original problem is ignorance. The integer bresenham algorithm is quite popular, but it amazes me that people think is a "traditional" approach, instead of solving a 1st degree equation. (Wtf are people learning in schools nowadays?).

      Regarding your div remarks - yes, division is a complex process, but easily solved by throwing more transistors at it. The dependency chain between instructions is only relevant to high-penalty cpus (such as P4), or when you have multiple execution pipes, but no out-of-order execution support. Well, the Pentium mechanism is pretty simple, if instructions can be paired on the U and V pipes, they will, if not, tough luck - they'll be executed sequentially. That is the motive I used it - it is simple to understand. Pentium Pro and later processors do feature multiple "all purpose" pipes, but also feature out-of-order execution. Not that matters much, because the point I made was precisely about the futility of single-processor optimization - what is clever on one cpu release may be utterly stupid on another (and almost everything was stupid on the P4 aberration).

    16. Re:Unbelievable... by robsku · · Score: 1
      --
      In capitalist USA corporations control the government.
  14. Re:Not that impressive by black6host · · Score: 1

    For PC based demos Win 7 is used. (Linux also available) The latest version of Direct X and .Net are present and available for use. Other OSs and hardware will have their own support packages as well. So yes, third party libraries are available but that does not make the outcomes any less impressive.

    Here is a link to where the info above came from with additional details:

    http://www.revision-party.net/compos/pc/

  15. Re:Not that impressive by Anonymous Coward · · Score: 1

    Yes it does make the outcome less impressive. It's easy to make small executables when you can shove off the hard work to the 100s of megs of libraries you use. Real demosceners were using their own rendering code in a tight executable not using bloated rendering frameworks like Direct3D. Lame.

  16. Damn you kids, get off my lawn. by tgd · · Score: 4, Interesting

    /. is really making me feel old these days -- I was writing demos in the early 90's. I don't know if its my overall grumpy old-man mentality or not, but as impressive as these are, they're powered by a crap-ton of software running behind them. There's not 64k of assembly pumping bytes into a framebuffer and twiddling the PC speaker port to synthesize digital audio.

    One thing I couldn't find in there (and I've been out of the scene for a LONG time, so I don't know how this works on new-fangled fancy computers...) -- do these write directly to the video hardware? Or do they use OS services like DirectX11, etc? When they say 64k, is it a 64k executable using up another dozen meg of OS DLLs?

    I have to give it to them, they are very impressive. But are people still getting down and counting clock cycles?

    Anyway, for you youngins, this might explain the demoscene a bit better: http://www.youtube.com/watch?v=iRkZcTg1JWU

    1. Re:Damn you kids, get off my lawn. by Terrasque · · Score: 1

      do these write directly to the video hardware?

      No, that would be impossible on modern computers (unless you would only run the demo on one gfx card - and maybe even only on one specific firmware version at that). The closest I know of is the 4k demo Elevated by RGBA and TBC. Everything except audio is done by shaders on the graphics card. Even the camera movements are done there (as the cpu don't even know how the terrain looks).

      IIRC they started with an OpenGL initialization to get the shader code to the GPU, but later switched to DirectX, because that used 100 less bytes in the finished product.

      But people still make C64 and Amiga demos, and lft's Craft and Phasor is even more extreme, as he does everything himself :)

      --
      It's The Golden Rule: "He who has the gold makes the rules."
    2. Re:Damn you kids, get off my lawn. by virgnarus · · Score: 1

      Unless they use their own driver, I can't see how they can write directly to the video hardware without doing so through passing it via DirectX11 APIs. Nowadays it's all a bunch of API calls.

      I think it also goes to saying that there's not going to be much involvement with reducing clock cycles per instruction because nowadays CPUs are very well capable. Now the focus on how to efficiently utilize APIs to do your bidding and how to use tricks to compress textures and code as much as possible. So yes, while reducing the code to as minimal as possible would entail some optimization for clock cycles, a lot of the management is no longer done directly via the internal code but through external DLLs supplied by the OS.

    3. Re:Damn you kids, get off my lawn. by Hatta · · Score: 3, Interesting

      Or do they use OS services like DirectX11, etc? When they say 64k, is it a 64k executable using up another dozen meg of OS DLLs?

      That was exactly my question. If it's not 64K running on bare metal, it's cheating.

      --
      Give me Classic Slashdot or give me death!
    4. Re:Damn you kids, get off my lawn. by tgd · · Score: 1

      do these write directly to the video hardware?

      No, that would be impossible on modern computers (unless you would only run the demo on one gfx card - and maybe even only on one specific firmware version at that).

      Admittedly its been a while, but the VGA interfaces are still on those cards. If you needed higher rez, the VESA interfaces were pretty broadly supported ten years ago. I assume they're still buried down deep on those cards. But I grant you that utilizing more than just the framebuffer, you need to code to specific cards. But during the mid 90's, that was pretty common -- you saw lots of demos coding to specific cards like the early ATI Rage or (much more commonly) the 3DFX VooDoo cards. Same with audio hardware.

      So you *can* do it on modern hardware, you just have to want to take on that challenge.

    5. Re:Damn you kids, get off my lawn. by Ben+Hutchings · · Score: 1

      There's not 64k of assembly pumping bytes into a framebuffer and twiddling the PC speaker port to synthesize digital audio.

      Of course. But all the creative work is squeezed into 64K.

      One thing I couldn't find in there (and I've been out of the scene for a LONG time, so I don't know how this works on new-fangled fancy computers...) -- do these write directly to the video hardware? Or do they use OS services like DirectX11, etc?

      They use DirectX, because that is the only way to support a reasonable range of hardware. (Also, you can't hit the hardware without installing a new driver or exploiting a kernel bug. Neither of which is very friendly.)

      But are people still getting down and counting clock cycles?

      Cycle counts aren't even documented today. Now it's all about avoiding cache misses and cache invalidation.

    6. Re:Damn you kids, get off my lawn. by Hatta · · Score: 3, Interesting

      Yes, the executable. Running on what OS, with god knows how many megabytes of libraries? That's not a 64K demo.

      --
      Give me Classic Slashdot or give me death!
    7. Re:Damn you kids, get off my lawn. by AmiMoJo · · Score: 1

      The rule is that the executable must be 64k and not use an data files. DLLs are allowed, however, and all of these demos use Direct X for video and sound.

      The objects and textures are procedurally generated with a bunch of pixel shaders thrown in. Often they take minutes to load even on a high end machine as they do all the generation. No-one counts clock cycles because the target hardware is abstracted anyway.

      A long way from the way things used to be, hitting the bare metal. In fact the original limit was 40k because that was what would fit on one track of an Amiga floppy disk and thus what most crackers used for their "cracktros", little demos at the start of cracked games. Back then the crack usually includes some cheats too. Somewhere along the line the PC guys decided 40k wasn't enough and moved to 64k.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    8. Re:Damn you kids, get off my lawn. by Narishma · · Score: 1

      Since modern operating systems won't let you directly access hardware, the only way to do what you want is to bypass the OS, and thus write drivers for any hardware you may want to use. And the resulting demo will only run on your particular machine, or one extremely similar to it. Good luck with that.

      As for counting cycles, I don't think it's even possible in modern CPUs due to how they run instructions out of order.

      --
      Mada mada dane.
    9. Re:Damn you kids, get off my lawn. by Narishma · · Score: 1

      They may have moved to 64k because that was the maximum size of an executable .COM file on CP/M and MS-DOS.

      --
      Mada mada dane.
    10. Re:Damn you kids, get off my lawn. by sqldr · · Score: 1

      do these write directly to the video hardware?

      I'm meaty. I coded the synth in gaia mahina. We link against opengl, but we had to do some real dirty shit to throw out stdlib and don't get me started on the manifest. There were oldskool demos at the party on A500s and C64s, but don't get us wrong.. we had to do some crazy shit to get it down to 64kb. It was exactly 65536 bytes after some work. Try getting something like that down to 64k and you will see.

      PS. have 15k of muisc: http://spacepigs.com/meaty-alternature.mp3

      Same synth

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    11. Re:Damn you kids, get off my lawn. by sqldr · · Score: 1
      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    12. Re:Damn you kids, get off my lawn. by Man+Eating+Duck · · Score: 1

      That was exactly my question. If it's not 64K running on bare metal, it's cheating.

      Running on bare metal is hardly feasible these days, as hardware is so varied. You need the driver APIs to get anything done. Or you can watch Amiga-demos from the same compo (some incredibly good stuff there as well), but remember that they're also "cheating", employing Denise, Paula and Agnus as they are. For shame, they probably even use the blitter instead of doing it in CPU!

      And - did you actually *see* "Gaia Machina"? As there is no way they can pack all that geometry pre-made in 64KB, far less textures, they have to generate it procedurally. The result is, if not photo-realistic, at least very good-looking strawberries, apples, leaves, plants, stones and so on. Same goes for animation and geometry morphing. Yes, they get a lot of 3d functionality from DirectX, but as the starting line is moved, so are the goal posts.

      If this is not impressive to you, you're incredibly jaded, or I could give you the benefit of the doubt and say that you probably didn't really consider it. Keep in mind that this is likely the *best 64KB demo in the world* right now. Be sure to let us know when you're the best in the world at something.

      --
      Are you a grammar Nazi? I'm trying to improve my English; please correct my errors! :)
    13. Re:Damn you kids, get off my lawn. by Hatta · · Score: 1

      Running on bare metal is hardly feasible these days, as hardware is so varied.

      A tiny demo that runs on exactly one piece of hardware is more impressive than a 100mb demo that runs on everything.

      Or you can watch Amiga-demos from the same compo (some incredibly good stuff there as well), but remember that they're also "cheating", employing Denise, Paula and Agnus as they are. For shame, they probably even use the blitter instead of doing it in CPU!

      That's hardware. Fair game.

      Yes, they get a lot of 3d functionality from DirectX, but as the starting line is moved, so are the goal posts.

      Then include the size of DirectX in the size of the demo.

      If this is not impressive to you, you're incredibly jaded, or I could give you the benefit of the doubt and say that you probably didn't really consider it.

      It is impressive, it's just not a 64KB demo.

      Keep in mind that this is likely the *best 64KB demo in the world* right now.

      I'm not arguing that it's not the best of its class. I'm just aruging that it's not really a 64K demo.

      --
      Give me Classic Slashdot or give me death!
    14. Re:Damn you kids, get off my lawn. by k8to · · Score: 4, Informative

      Newsflash; we haven't had OS-less demos on the PC, ever. Even in msdos days we were using bios routines to load in data, if not to initialize the videocard etc.

      --
      -josh
    15. Re:Damn you kids, get off my lawn. by k8to · · Score: 1

      You're out of date.

      The idea of demos and intros shifts over time. Originally, intros weren't even designed with any sort of 'competition' in mind at all, so the whole idea of 'cheating vs 'not cheating' couldn't even apply.

      Nowadays, the idea of "showing lots of polygons texture mapped etc" isn't very interesting, because the hardware can this so well that there's nothing interesting to show by being clever. So the demoscene has moved on: Demos are typically more about doing something visually creative, wheras 64kb and 4kb intros are more about how much you can express in that amount of space. You have to somehow create your music, textures, objects, camera paths, and so on all in that space. If you think this is not a technical challenge, then you're deeply wrong.

      --
      -josh
    16. Re:Damn you kids, get off my lawn. by scot4875 · · Score: 1

      So let's see your modern 64K demos that that run on bare modern hardware. Bonus points if they're at least as impressive as the best ones from the early-mid 90s.

      Put up or shut up.

      --Jeremy

      --
      Jesus was a liberal
    17. Re:Damn you kids, get off my lawn. by sexconker · · Score: 1

      Or do they use OS services like DirectX11, etc? When they say 64k, is it a 64k executable using up another dozen meg of OS DLLs?

      That was exactly my question. If it's not 64K running on bare metal, it's cheating.

      What a clown.

    18. Re:Damn you kids, get off my lawn. by Terrasque · · Score: 1

      "No true Scotsman" fallacy? :)

      Of course, unless you've created your own demo platform (all the way from mining your own minerals... Oh, and making your own pick to mine the minerals.. And ...), you're just using what others have done for you already.

      You have to set a limit somewhere. You just seem gruff because you've arbitrary set the limit a different place than the competition holders. Most of the old DOS demos used BIOS routines, after all.

      The compo rules for win 64kb :

      Windows:
      * The compo machine will be installed with Windows 7 64bit, however all entries have to be supplied in 32bit exe format and be able to run in a 32bit environment!

      * We will provide the then-latest DirectX distributables.

      * The "Media" and "Music samples" directories will be deleted from the compo machine.

      * .net Entries : We will install the then-current version and patches of the latest .net runtime.

      * And in case someone might still have this mad idea : We will not install any drivers or virtual machines to provide or improve DOS compatibility.

      64k intro:

      * Maximum file size is 65536 bytes for the executable. All other files in the archive will be deleted before showing the entry in the competition.

      * Maximum running time: 8 minutes (including loading/precalc).

      * We will not install any additional Runtimes, SDKs, Codecs, Drivers etc on the compo machine. This means that, among others, msvcr70.dll, msvcr71.dll and msvcr80.dll will not be available.

      * You may not use the contents of the Windows "Media" or "Music Samples" directories. These directories will be deleted on the compo machine.

      * Java entries are not allowed

      * .net entries are allowed when they follow the rules listed above.

      Pretty reasonable for a windows demo IMHO. You get stock windows, latest directx distributables (otherwise you wouldn't be reasonably able to do anything with the gfx card in the machine), and that's it. Under windows (NT kernel series) you don't have any direct access to the hardware, unless you write a custom driver (which they will of course refuse to install) or find a kernel exploit. This is what a modern demo is. Using a modern computer. There are of course also compos for older machines, if that better tickles your fancy.

      You of course also got other wild stuff.

      --
      It's The Golden Rule: "He who has the gold makes the rules."
    19. Re:Damn you kids, get off my lawn. by sqldr · · Score: 2

      we crammed a LOT of code into 64k. when it decompresses, it explodes into about 800 megs of calculations before it starts, but you have a point, and I'd like to see you prove it by releasing a 64k demo that beats one of ours..

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    20. Re:Damn you kids, get off my lawn. by sqldr · · Score: 1

      the release version was on windows, but the linux version is coming. we promise it soon. As for libraries, we used a lot of assembler to throw out stdlib and the windows manifest.

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    21. Re:Damn you kids, get off my lawn. by bored · · Score: 1

      (per writing to the hardware)

      No, that would be impossible on modern computers (unless you would only run the demo on one gfx card - and maybe even only on one specific firmware version at that).

      If by "modern" you mean linux, there is fbdev which has a semi standard kernel interface for providing drivers that export the framebuffer. You mmap the local fbdev device into your process. Its fairly cross platform and cross video card (for many older ones it uses the VESA API). Sure it won't let you twiddle the PCI BAR's or individual card registers, but if you want to do that you can hack the appropriate fbdev driver. Most of them are just a thousand lines or so (been there done that, don't ask). It just depends on how much you really want to tweak the hardware. Most it seems just want a framebuffer and hardware blit support (or line drawing/etc).

      Sitting on that, you can have SDL or DirectFB, which attempt to provide more than a basic framebuffer and mode setting API in a cross platform manner. The primary goal seems to be some basic hardware acceleration. The nice thing about them, is that if a particular piece of hardware doesn't support some feature they emulate it with software. In the case of some old sparc video cards that apparently goes as far as emulating the framebuffer.

    22. Re:Damn you kids, get off my lawn. by viperidaenz · · Score: 1

      But even then you're sitting on top of the entire linux kernel, the video drivers AND the framebuffer API. Not really much different from DirectX.

    23. Re:Damn you kids, get off my lawn. by bored · · Score: 1

      yes... same with vesa, and another poster pointing out that even in the DOS days people depended on the DOS for file io, and setting up the initial execution environment.

      That said, the closest thing in windows (that i'm aware of) to the fbdev in linux is the deprecated directdraw hardware surface (which is a darn lot like SDL!). As I understand it, directdraw is completely emulated in software since vista. Even back in the w2k days it didn't give you full access to the hardware framebuffer, instead giving you a mapping to what was effectively a mapped surface that the hardware placed into the actual framebuffer. Basically, it behaved more like mapping a texture to a polygon being flushed to the actual hardware framebuffer than full framebuffer access.

      Not that you really want to access the framebuffer over PCIe anyway, its a lot faster to use one of the dozen 3d tricks than access it directly. Hence direct2d.

    24. Re:Damn you kids, get off my lawn. by viperidaenz · · Score: 1

      You don't want direct access to the hardware framebuffer, regardless if its over PCIe or not, unless you're a video driver.

    25. Re:Damn you kids, get off my lawn. by noodler · · Score: 1

      "I'm not arguing that it's not the best of its class. I'm just aruging that it's not really a 64K demo."

      You just have to make the switch and see the available system api's as hardware that you program against.
      And then the whole trickbox approach becomes valid again.
      Because you can do DX calls the 'normal' way or you can optimize the shit out of them and put more usefull code in the same space.
      So while the actual things that you tinker with now are software interfaces for the perifferals as opposed to hardware you still get to be smart-ass and do cool stuff with the limitation.

      For me, it is not so much about tickling hardware as it is about doing as much as possible in as little as possible space.
      I mean, even a VGA interface is 'cheating' when compared to programming for the 2600's stella, and the 2600 needed serious tricks every time you wanted to show anything on screen, not just demos.

      Every generation has to deal with a different set of platformity and each generation has to figure out what the challenge is for the available platforms.
      Since 3D hardware is ubiquitous but highly complex and depending on drivers to actually work (the drivers are a big part of the actual design of those components) it's just natural to use the provided interface (the driver as opposed to the bus) to do your tricks on.

    26. Re:Damn you kids, get off my lawn. by bored · · Score: 1

      I'm not sure I agree with this 100%. In general no, most apps don't need or shouldn't have access to an actual framebuffer. Thats assuming something that even makes sense as a framebuffer is available.

      On the other hand, the topic of mirror drivers comes up fairly frequently on ntdev, because windows lacks the ability to access the framebuffer directly. Sometimes people with legitimate needs want to know what parts of the screen are being updated, etc..

      So, there are reasons for lower level access to the video hardware. The nvidia stance that its should all be black magic outside of D3D or whatnot is complete BS. I've personally made utilities that hack some portion of a video card, because the vendor didn't provide support for some feature or another. That was back in the bad old days when you could call up the vendor of the video chipset and receive a 800 page book in the mail detailing the register set.

    27. Re:Damn you kids, get off my lawn. by TarMil · · Score: 1

      The closest I know of is the 4k demo Elevated by RGBA and TBC. Everything except audio is done by shaders on the graphics card. Even the camera movements are done there (as the cpu don't even know how the terrain looks).

      Actually, most current 4ks work that way. Rendering is done by just drawing a quad the size of the screen and doing all the work in the shader. The most common use is to draw fractal shapes that would take way too many vertices if rendered classically (an example from Revision: Hartverdrahtet).

  17. Wow! by DJRikki · · Score: 1

    I remember the Fishtro from Future Crew exactly 20 years ago and it was 220k! (http://www.pouet.net/prod.php?which=1283) this is breathtaking stuff. Magic!

  18. Just don't use more than 10 EXEs by Anonymous Coward · · Score: 1

    Because 640K ought to be enough for anyone.

  19. Personal favourite by cockroach2 · · Score: 1

    My personal favourite is still farbrausch's "fr-08 - the product": http://www.youtube.com/watch?v=1dcrV_7JpXQ

    It was amazing "back then" and even today I still think it's highly awesome. All of that in 64k.

    1. Re:Personal favourite by cockroach2 · · Score: 1

      Seriously?

    2. Re:Personal favourite by Man+Eating+Duck · · Score: 1

      My personal favourite is still farbrausch's "fr-08 - the product": http://www.youtube.com/watch?v=1dcrV_7JpXQ

      It was amazing "back then" and even today I still think it's highly awesome. All of that in 64k.

      Yeah, that's a classic, complete with awful scene poetry and all :)

      I have a couple of friends from Andromeda (Hyde and Archmage), they came in second to farbrausch's "debris" at Breakpoint 2007. Archmage simply stated that "losing only to farbrausch is still a victory".

      That said, to me Gaia Machine surpasses the .product in technical quality and polish. Also check out some procedural 4k images, the best ones really boggle the mind.

      --
      Are you a grammar Nazi? I'm trying to improve my English; please correct my errors! :)
  20. Re:Word Salad by Anonymous Coward · · Score: 1

    And also what the hell happened that the first two comments are expressing confusion over what this story is about?

    I was never involved in demoscene stuff (despite having an Amiga during the early 90s), but I certainly knew about its existence. Was it really so niche after all?

  21. Re:Am I supposed to have heard about this before?? by discord5 · · Score: 1

    Demoscene? Demoparty? 64kb executable?

    Here is a nice article on the subject. Now hand in your geek card.

  22. Re:Not that impressive by Sarten-X · · Score: 1

    Demoscene was not always about just making the most impressive output from a given piece of technology. As I recall, demoscene evolved from game pirates marking their cracked games with elaborate intros using only the available space unused by the game. I also recall hearing that some teams would occasionally use routines from the actual game as part of their animations. With that history in mind, how can one object to using libraries now? The original spirit of the art, where your own code had to fit into a small space, is perfectly intact. Today, you can expect DirectX and OpenGL just as much as you could expect a clock in older hardware.

    --
    You do not have a moral or legal right to do absolutely anything you want.
  23. Re:Not that impressive by dmacleod808 · · Score: 1

    The point is the libraries exist, programming does not need to be a 16GB Blue ray, code is so bloated and inefficient for the amount of processing power available.

    --
    There Can Be Only One...
  24. And? by SmallFurryCreature · · Score: 1

    And people before that did it without DOS. Go all the way back and people where flipping switches to input their code. And they found it boring to rewrite the same stuff over and over again and so created common libraries that soon became an OS and everything else.

    If you make a cake from scratch, do you grown your own wheat? Then you are using the library of nature/god! Slacker! I create my own universe for every sandwhich, Big Bang all the way or you are just a faker!

    Personally I think 64k is to limitting, it was nice some years ago but today, it just isn't realistic anymore, not when your average PC has 8GB of memory PLUS video memory. Go crazy, go a full MB! Make that floppy work!

    I can appreciate the skill but it is like seeing someone make a nice statue with a flint... nice... now here is a steel chisel. Enjoy!

    --

    MMO Quests are like orgasms:

    You may solo them, I prefer them in a group.

    1. Re:And? by tgd · · Score: 1

      I can appreciate the skill but it is like seeing someone make a nice statue with a flint... nice... now here is a steel chisel. Enjoy!

      That skill is the point of demos, in the demoscene sense of the word.

  25. Re:Not that impressive by msauve · · Score: 1

    third party libraries are available

    Oh, then the summary did both of those, and more, in only 698 bytes.

    --
    "National Security is the chief cause of national insecurity." - Celine's First Law
  26. Re:Mod -1 by Georules · · Score: 1

    The terminology is not obscure. You are ignorant.

  27. Re:Word Salad by virgnarus · · Score: 1

    So instead of trying to make sense of it all by researching on the topic, you immediately start complaining because it assumed you had prior knowledge? I'd hate to see how you've reacted to the math and science articles here a /. .

    What ever did happen to people involving themselves in /. because of an inquisitive nature?

  28. Yeah yeah old man by SmallFurryCreature · · Score: 3, Insightful

    But in your days it was easy, you could count the clock cycles on the fingers of one hand and if you wanted a bit flipped you just climbed inside the computer with a hammer!

    Anyway, you weren't all that impressive, you relied on a blacksmith for a hammer and a miner for the coal to fire your machine. You were just the slave master benefiting from the slave labor of others.

    --

    MMO Quests are like orgasms:

    You may solo them, I prefer them in a group.

    1. Re:Yeah yeah old man by sqldr · · Score: 1

      craft by lft was on bare hardware :-D

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
  29. Re:Not that impressive by Terrasque · · Score: 1

    No, real demosceners make their own demo hardware. Those so-called "sceners" that use pre-assembled electronics are just lame.

    Sorry, but just accept that the world have moved on. And you should also understand that these modern demos are just as impressive as ye olde ones.

    And for that matter, people are still making C64 demos. Like Edge of Disgrace from 2008.

    --
    It's The Golden Rule: "He who has the gold makes the rules."
  30. and the super Nintendo had 3d by issicus · · Score: 1

    64k? at least it will fit on my ssd..

  31. Wortbildung by tepples · · Score: 3, Informative

    I was wondering why it's "demoscene" and not "demo scene"

    Because English puts spaces in its compound words more often than German does.

  32. Re:Am I supposed to have heard about this before?? by SCHecklerX · · Score: 2

    To be worthy of posting on slashdot? Yes.

    Here's the blurb I posted on my facebook page:

    I forgot to mention. This stuff is also all rendered in real time. It's not a movie. The music is also composed/tracked. It is not a recording. Here's another impressive entry in the 64K competition: http://www.youtube.com/watch?v=6CiF034IhgY&hd=1 It's mind blowing where these guys have gone over the years. I thought the demoscene would have died as computers became more powerful and anyone could create effects without having to be an artistic assembly programming god. I apparently, and thankfully, was wrong :-)

  33. Re:Am I supposed to have heard about this before?? by Johann+Lau · · Score: 1
  34. Re:Am I supposed to have heard about this before?? by wisnoskij · · Score: 1

    If a modern day game company made that it would be a 500MB cutscene.

    --
    Troll is not a replacement for I disagree.
  35. Re:Word Salad by sqldr · · Score: 2

    You would be surprised at how many HTML/CSS monkeys, calls themselves "programmers" these days.

    one. I met him once in a pub. he's a right cunt.

    --
    I wrote my first program at the age of six, and I still can't work out how this website works.
  36. Re:Files are larger than 64K by MiG82au · · Score: 1

    Why the fuck don't you actually open them instead of talking shit? Do you have an uncontrollable tendency to do so?

  37. Re:Am I supposed to have heard about this before?? by HyperQuantum · · Score: 1

    There is one such website. Google for it.

    --
    I am not really here right now.
  38. Re:Word Salad by Anonymous Coward · · Score: 1

    It's specific technical language. None of it is babble or word salad. All of the words have meaning, whether or not you understand.

  39. Re:Not that impressive by Goaway · · Score: 1

    The "libraries" only ever abstract away the 3d hardware. If you limited yourself to a single kind of hardware and it was documented, you could do pretty much the same things in mostly the same space.

  40. I am completely humbled by iMadeGhostzilla · · Score: 1

    Just downloaded the Felix demo and ran it on my PC. Even if it were 64MB I'd still be very moved by the skills and heart put into the artistic side of the demo. At 64KB I've been sitting in silence for 15 minutes now.

    1. Re:I am completely humbled by sqldr · · Score: 2

      oh you bastard! I helped code gaia machina! download the other one! we won ffs!

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    2. Re:I am completely humbled by iMadeGhostzilla · · Score: 1

      isn't life funny, you know you did an incredible work and you actually won, by the vote of your peers, and then some random guy (me) comes along and says how he likes the other one, and disappointment kicks in. :-) I downloaded gaia first but it crashed. I have downloaded a few demos in the last few years and every single one crashed, except for Felix. I believe, but I need to see the code running for humbleness to rise.

      Btw I liked Felix Workshop in particular b/c I'm partial to music visualization.

    3. Re:I am completely humbled by sqldr · · Score: 1

      linux version is coming.. windows being windows, we have no idea why it doesn't work for some people. linux version will :-)

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
    4. Re:I am completely humbled by iMadeGhostzilla · · Score: 1

      that's fine, I personally wouldn't expect ultratight code to work on platforms that deviate from strict specs. you guys put in amazing work, nevertheless. that level of control over a computer I had only for a 1000 times less complex machine, ZX Spectrum, 1000 years ago. :-)

  41. Re:Not that impressive by _Shad0w_ · · Score: 1

    Some of the demo coders from years ago went on to start companies that created mass market hardware based 3D rendering. I doubt they'd be mocking at all.

    --

    Yeah, I had a sig once; I got bored of it.

  42. Re:Word Salad by jgrahn · · Score: 1

    And also what the hell happened that the first two comments are expressing confusion over what this story is about?

    I was never involved in demoscene stuff (despite having an Amiga during the early 90s), but I certainly knew about its existence. Was it really so niche after all?

    Yes, I think it was. There's Unix, there's PC/Mac, and there's the various home computers. If you were in one of these camps, chances are you knew nothing about what was going on in the two others.

  43. Re:nVidia comparison by viperidaenz · · Score: 1

    Because the nvidia demos are full of highly detailed textures. These demos contain highly compressed textures.

  44. kb != KiB by pgn674 · · Score: 1

    The summary says 64kb (kilobits), or 64,000 bits. That's 7.8125 KiB (kibibytes), or 8,000 bytes. The competition is actually for 65,536 bytes, or 64 KiB.

  45. yea and by Osgeld · · Score: 1

    they have been doing it for two decades ... you should forward this story to my mom, she might be astounded by it

    1. Re:yea and by DJRikki · · Score: 1

      Yea and nVidia have been bringing out new video cards for over a decade ... still see the geekdom getting boners for the latest and greatest.

  46. Enhanced patdown by tepples · · Score: 1

    Then perhaps I was unclear: How should someone leave the United States without being groped?

    1. Re:Enhanced patdown by sqldr · · Score: 1

      wear body armour

      --
      I wrote my first program at the age of six, and I still can't work out how this website works.
  47. Graphics workload by tepples · · Score: 1

    These kinds of people need to get together and make games, or even better a game engine.

    You mean like .kkrieger?

    But seriously, a 20 hour game is a lot more work for a demogroup's graphician than a 20 minute demo. There's an amount of work beyond which a graphician has the patience to work for free.

  48. Re:Not that impressive by noodler · · Score: 1

    'Lame.'

    The reality is that even the initialisation code of modern GPU's is several MB in size.
    And the hardware is propietary.
    And it is so incredibly complex that you would need a serious team for a long time to make a driver for that specific piece of graphics hardware.
    Writing your own GPU drivers for the purpose of demos goes beyond the purpose of writing demo's.
    Mind you, where would you get the specifications of the diverse GPU's from?
    It used to be relatively easy in ye old days where almost all available hardware was either documented or simple enough to experiment your way through.
    Nowadays even the software interfaces to the drivers are dragons. You don't want to know what is actually happening on the bus level.

    I mean, you're expecting demo guys to write their own GPU drivers while the OS community is struggeling with this despite a ;arge ammount of serious people working on the issue.
    Go ask these guys if they would whip up a driver for the sake of a demo.
    Then ask them if it would fit in any size below 1MB.
    Note the strange looks and insecure laughs.

    That is just the reality of computing these days.
    I first noticed when i saw java vm bytecode demo's 10 years ago. The world had changed.

  49. Irony by thegarbz · · Score: 1

    I purchased their subsequent game, Enclave, because of that demo

    That's quite ironic given how the most exposure many people have to a demo scene is the executable that reads the NFO file that comes with a lot of pirated software.