Slashdot Mirror

Directsound was killed off in windows vista. Openal was supposed to replace it but that's pretty much a dead stick too, though some engines still support it. The current standard in windows, xbox, and phone is xaudio2 which does support multichannel with a low latency software mixer. However, there is no hardware acceleration for it. Some titles do their own mixing and dsp in-game with libraries like FMOD and just push the output to xaudio. Titles typically expose multiple speaker modes so adding multichannel isn't that big of a deal.

https://en.wikipedia.org/wiki/...
http://www.fmod.org/

I'm not saying they can't do a good game, it's just that they're not quite there yet.

I vote with my wallet and commend them for their lack of DRM in the past. Not to mention they take care of their customers.

For cross platform I'd add Bioware and Blizzard to that. Not Bethesda, they do good games, but they seriously need more solid QA. It's sad to think that their least buggy console game on the PS3 was Oblivion

Simply criminal. A vanilla experience robbed of Bethesda's strength: mods. Bioware was once a powerhouse, Mass Effect 3? Blizzard releases are strong.

As much as that feature is appealing to some do you think that they should dedicate dev time to duplicate functionality already in other applications or the OS?

Games do this all the time, take widgets for example. Typically the engines feature media playback functionality or a sound library is licensed, and judging by the glut of media players out there, they're not exactly taxing to write. The Xbox has games that do exactly this.

• Mac OS X
• iPhone
• GNU/Linux (both OSS and ALSA backends)
• BSD
• Solaris
• IRIX
• Microsoft Windows
• Xbox
• Xbox 360
• MorphOS

FMOD is also available on most game and PC platforms and has very reasonable licensing fees.

Neither of these audio solutions (or "stacks" as you prefer to call them) are obsolete or "suck".

Your mod file example is great. It's a now archaic format which is no longer in vogue.

Fortunately the LPT port nonsense was a hardware specific nonsense completely independent of the format, and open source code which can reproduce the music on any modern platform is readily available. http://mikmod.raphnet.net/ http://www.modplug.com/ (sourcecode for modplug is availble, believe it or not)

There are also closed solutions: http://www.un4seen.com/

And there are industry standard sound libraries that do the job fine: http://www.fmod.org/

So, it seems that these antiquated technologies (mod dates to 1987) tend to get supported just fine.

Holy crap, I remember those! Anyone have something that'll still play them?

Of course. The format is still in widespread use, although typically more than 4 channels are employed these days (and they aren't hard-coded to left and right channels, as with the Amiga). There are many Pocket PC, Gameboy and Cellphone games that use tracker style playback (most completely MOD compatible) to save storage space.

MikMod, fmod and Hekkus are three different libraries currently used by game developers for mod playback. However since flash storage has increased dramatically over the last couple years, more and more developers are using mp3 format. So that may finally put an end to the use of MODs.

Dan East

There is a very simple man in the middle attack. MusicForMasses.exe sends unencrypted mp3 data to fmod.dll for decoding and playback. Just create your own fmod.dll and snatch the plaintext.

All you need is the fmod 3.75 programmers API from http://www.fmod.org/

I think the big question is, how can we get small game studios back? Is it really not possible for a small team to make commercial games?

I believe that the problem smaller studios face can be overcome with some lateral thinking. The problem is two-fold: production costs and marketing costs are too high to allow indies to compete on equal footing with the big boys. The solution, then, is to not compete on equal footing.

Don't: Try to copy a game that took 60 people 3 years to create.
Do: Draw from an existing genre, but come up with a unique twist -- something meaty that doesn't exist elsewhere.

Don't: Compete with larger productions on the same style of graphics.
Do: Come up with a unique look; it's easier to wow people with a fresh style. (Though Monolith is not a small studio, Tron 2.0 was the opposite of the hyper-realism trend, and set itself apart on appearance, among other things.)

Don't: Try to out-advertise Activision, Microsoft, or Infogr- er- Atari. A small studio's meager advertising budget should be used towards development.
Do: Make as much use of word-of-mouth marketing as is humanly possible. It's easier to connect with your individual players because... well... there are fewer of them.

Don't: Re-invent the wheel. id Software must create its own 3D engine from scratch; you don't (necessarily) have to.
Do: Make as much use of middleware as possible. You don't need to be an artist to create skycubes. You don't need to know DirectX or OpenGL intimately to create an engine. You don't need to write your audio engine from scratch.

And I deeply believe better games would be coming out of a smaller and more laid back studio...

I like the cut of your jib. I hope you're right.

________________________
Inago Rage - A first-person shooter where you fight in arenas of your own creation.

Onboard sound sucks for the same reasons as onboard video; the CPU gets too busy, the sound STOPS (on my brandnew laptop its doing just that now), and you get noise from transistors or capasitors near the sound chip.

Wrong. On-board video doesn't stop when the CPU gets busy. You don't get noisy audio from on-board sound if you use the digital outputs. Sound always has the highest priority in a game too, seeing as there is a low-latency buffer that needs to constantly be refreshed. (Ever noticed how the frame rate drops, yet the sound continues on?) If the CPU gets too busy with something else to service the sound buffer then it doesn't matter if it's on-board or add-on sound because the sound will fuck up either way. Oh yeah... if the xbox is a cut-down PC (as everyone here reckons) then why doesn't its on-board audio fuck up or sound shit?

I hate what creative did too, but this move by id will probably make the game run better, since the sound processing will be moved from the CPU to the sound card, leaving more CPU time for the physics engine.

Yeah, the physics engine that's locked to 60Hz. Have you seen that G4TV video on Doom 3 that was linked to from here the other day? JC talks about audio in it... how games have always used a few % of CPU time for it, even as CPU power has increased - hence why modern games have awesome sound.

Really, you know this how? Somehow i think you're talking out your ass...and don't forget there could be an over all performance gain even if thats true, just b/c the CPU has more time to do other things then process sound.

Somehow I think you don't know what you're talking about. Do you code games for a living? I know I do. Doom 3 uses fmod for audio. (Well, I know that leaked version did.) The guy who wrote fmod knows his audio shit way more than you and I, which is why fmod is used in loads of games and you aren't the writer of what is arguably the world's best cross-platform audio API. Doom 3 also uses OGG encoded audio, so there's a CPU-side bit of decoding to do. Mixing audio is far less CPU intensive than decoding OGG, so why bother with setting up the audio hardware for the mixing? I could go on, but really there's no point.

Independent studios want to create wonderful, experimental titles, but are, in part, held back by business requirements. As businesses, our first priority is to become profitable, and the least-risky way to do this is to create more traditional offerings. (The same is true for large development houses.) Fortunately for us, better middleware tools and increased publicity can free us of this constraint. The former will allow us to experiment and develop easier; the latter will allow us to reach an audience now reserved for the large publishers. As these conditions improve, you'll see independents take more risks.

Middleware comprises the audio libraries, AI plugins, and 3D engines such as Torque, Conitec A6, and FMOD. These tidbits are the lifeblood of independents. Without them, we'd have to code everything from scratch, and you'd see even more Tetris clones than you do now -- little innovation. With them, we're freed from the low-level stuff. We can create games that look and sound good enough to attract consumers. As middleware improves, it'll become even easier to experiment and innovate.

Publicity is trickier -- while events such as the Independent Games Festival allow us to bend the ears of larger publications, it's still the big studios that are going to command the previews and exclusives. Having approached a number of print publications, I've found that it can be difficult to secure a sizable preview for our game, even though I think folks might like to hear about where we're innovating. But even this is improving; sites like The Adrenaline Vault are particularly indie-friendly, often posting press releases from smaller development studios.

I think, then, that it's only a matter of time before the smaller studios attempt experimental titles in substantial numbers. Many will be terrible; some will be great fun. But as it becomes easier to experiment, you bet we'll be doing more of it, simply because we can.

Oh well, I'll try again. Uh. fmod, for sound and music. Though you'll owe them money if you want to make any.

I'm hoping that Electronic Arts's transition from boutique software house to publishing juggernaut leaves room for -- well -- other boutique software houses. Many here must recall the early days of EA. They published titles that their small teams were passionate about; and while I've enjoyed many of EA's recent, grander offerings, they appeal to me in a much different way.

For the time being, the advent of a middleware industry is making it easier -- not harder -- for smaller studios to produce good-looking titles with depth. Consider that there are many audio libraries, 3D engines, and AI middleware libraries which are quite reasonably priced. Smaller studios seem to go strange and wonderful directions with these; (if you haven't already, try some of the Indie Game Jam titles, which make use of a simple, standardized physics engine).

I labor under the impression that the gaming public has a desire for boutique products; if I'm wrong, I don't mind taking my licks and moving to something more productive.

Gamedev.net has a TON of resources relating to game development, including a giant reference section which points to a lot of information about anything game-related.

Related to your question, a good compromise between low-level OpenGL and a full game engine (or even a commercial one here) would be GLUT. Instead of giving a piss-poor explanation of what the GLUT library is all about, I'll just point you to their FAQ. For sound there is OpenAL or the ubiquitous FMOD.

Depending on what you're looking for, a library probably is out there just waiting for you to find it. Instead of asking very general questions like this, I highly recommend you check out a little search engine called Google to find libraries or engines that suit your specific needs. Cheers!

May I suggest that you take a look at:

SDL - 2D and input (granted, it doesn't support as wide a range of input devices, but in most cases keyboard and mouse are all you need
fmod - best sound lib out there, used by loads of commercial games.
HawkNL/HawkVoice - good networking libs

Hope that helps you out. It helped me out.

--

The first thing you'll need to to is learn a programming language and get familiar with it's development tools on your platform of choice, be it Linux, Windows or MacOS. You don't need to master the language as demos do not always need the most advanced features that some languages offer (such as multiple inheritance, etc). At first, writing 'hello world' -programs and such may be quite boring and frustrating but have patience - eventually you'll see that the skills you've learned will prove useful. A final word regarding the language; In the old days all demos were written in pure assembly language but these days C++ is the one most coders use and would be smart choice to learn. A good book on C++ is a way to get started.

Once you're comfortable with the language and it's tools, the next step is to learn more about your computer works. You'll need to know how to control the hardware; graphics card, sound card, timers and such. In modern operating systems this means studying the application programming interfaces (APIs) that the operating system provides. Fortunately, there are plenty of good tutorials that teach how to, for example, display graphics using OpenGL or DirectX. There are also many good libraries (Bass, FMOD, etc) for playing different audio files and you'll find them to be quite useful when adding music to your demo productions. Most people use their own timing system but there are some timeline editors such as Demopaja which may ease the work.

Now that you know the language and how to control your computer, you'll need to know how to code effects. Learning the math behind 3D graphics is quite useful, as is knowing how to manipulate bitmaps with convolution matrices and filters. Don't be afraid of math - it's a very useful tool and a key to understanding how effects really work. The internet is full of information on how to code different effects and functional examples to experiment with. Watch a lot of demos and try to figure out what effects they use and how they are are made. Read magazines, look at pictures in them and try to imagine how they would look animated. Look around when walking in a city or in nature - think of ways how to model things in the real world. Experiment with your code - quite many effects have been found by accident when experimenting. Chat with other demosceners on the net, maybe even meet them in real life at parties. Friendship is an important part of demoscene and discussing ideas with other people will often spring new ones. Code alone doesn't make a demo and you'll need find friends who can make music, paint pictures or create 3d models.

Don't feel overwhelmed by everything that you need to learn and do. If you can find the right frame of mind, you'll endure through the tedious basics of programming - see it as a challenge. After a lot of work, seeing your own effect work is a very gratifying experience. Once you know the basics, the possibilities are endless. Above all, the most important thing is to have fun.

Some links:
NeHe's OpenGL tutorials
Andy Pike's DX8 tutorials
Bass
FMOD
M0ppi Demopaja

New formats are being picked up by software developers (especially console game developers, where RAM and Storage are at a premium).

I'll vouch for that. The games company I share an office with are using Ogg exclusively for the sound and music on their current (PS2/Xbox/PC) game. It's better than MP3, there are no royalty hassles to consider, and it's seamlessly supported by the sound engine they're using.

Or they could use FMOD...

A lot of game developers use this; not open source, if that is an issue for you, though.

-m

Furthermore, in the original Unreal, Epic ditched a weapon at the last minute which would've been excellent: a four-barreled shotgun which fires gatling-gun style. To see the weapon, go into the Mesh Editor in Unrealed and take a look at "QuadShotHeld" and "QuadShotPickup". In Unreal1, you can even summon it (console command: summon quadshot), but it won't work since they didn't put in the code.