I am aware of the following computer systems that can play back and (sometimes) record audio data, with their characteristics. Note that for most systems you can also buy "professional" sampling hardware, which supports much better quality, e.g. >= 44.1 k 16 bits stereo. The characteristics listed here are a rough estimate of the capabilities of the basic hardware only (and even here I am on thin ice, with systems becoming ever more powerful).
machine bits max sampling rate #output channels Mac (all types) 8 22k 1 Mac (newer ones) 16 64k 4(128) Apple IIgs 8 32k / >70k 16(st) PC/soundblaster pro 8 (22k st, 44.1k mo 1(st) & compatibles PC/soundblaster 16 16 44.1k 1(st) & later & compatibles Atari ST 8 22k 1 Atari STE,TT 8 50k 2 Atari Falcon 030 16 50k 8(st) Amiga 8 varies above 29k 4(st) Sun Sparc u-law 8k 1 Sun Sparcst. 10 u-law,8,16 48k 1(st) NeXT u-law,8,16 44.1k 1(st) SGI Indigo 8,16 48k 4(st) SGI Indigo2,Indy 8,16 48k 16(st,4-channel) Acorn Archimedes ~u-law ~180k 8(st) Sony NWS-3xxx u,A,8,16 8-37.8k 1(st) Sony NWS-5xxx u,A,8,16 8-48k 1(st) VAXstation 4000 u-law 8k 1 DEC 3000 u-law 8k 1 DEC 5000/20-25 u-law 8k 1 Tandy 1000/*L* 8 >=44k 1 Tandy 2500 8 >=44k 1 HP9000/705,710,425e u,A-law,16 8k 1 HP9000/715,725,735 u,A-law,16 48k 1(st) HP9000/755 option: u,A-law,16 48k 1(st) NCD MCX terminal u,A,8,16 52k 1(st)
4(st) means "four voices, stereo"; sampling rates xx/yy are different recording/playback rates; *L* is any type with 'L' in it.
All these machines can play back sound without additional hardware, although the needed software is not always standard; also, some machines need external hardware to record sound (or to record at higher quality, like the NeXT, whose built-in sampling hardware only does 8000 samples/sec in u-law). Please don't send me details on optional or 3rd party hardware, there is too much and it is really beyond the scope of this FAQ. In particular, there is a separate newsgroup devoted to PC sound cards: comp.sys.ibm.pc.soundcard, which includes FAQ of its own (also posted to comp.answers and news.answers).
The new VAXstation 4000 (VLC and model 60) series lets you PLAY audio (.au) files, and the package DECsound will let you do the recording. In fact, DECsound is given away free with Motif 1.1 and supports the VAXstation, Sun SPARCstation, DECvoice, and DECaudio devices. Sun sound files work without change. The Alpha systems also have DECsound bundled with Motif. Also, the DEC2000/300 (aka DECpc AXP 150) can use a Microsoft Sound Card, with AudioFile (see below) for sound.
The SGI Personal IRIS 4D/30 and 4D/35 have the same capabilities as the Indigo. The audio board was optional on the 4D/30. The Indigo2 and Indy features are a superset of the Indigo features.
The new Apple Macs have more powerful audio hardware; the latest models have built-in microphones.
Software exists for the PC that can play sound on its 1-bit speaker using pulse width modulation; Older Soundblaster boards record at rates up to 13 k and plays back up to 22 k (weird combination, but that's the way it is). Newer ones can record at 22k and play at 44k.
Here's some info about the newest Atari machine, the Falcon030. This machine has stereo 16 bit CODECs and a 32 MHz Motorola 56001 that can handle 8 channels of 16 bit audio, up to 50 khz/channel with simultaneous playback and record. The Falcon DMA sound engine is also compatible with the 8 bit stereo DMA used on the STe and TT. All of these systems use signed data.
On the NeXT, the Motorola 56001 DSP chip is programmable and you can (in principle) do what you want. The SGI Indigo uses the same DSP chip but it can't be programmed by users -- SGI prefers to offer it as a shared system resource to multiple applications, thus enabling developers to program audio with their Audio Library and avoid code modifications for execution on future machines with different audio hardware, i.e. a different DSP. For example, the Indigo2 and Indy do not have a DSP chip.
The Amiga also has a 6-bit volume, which can be used to produce something like a 14-bit output for each voice. The hardware can also use one of each voice-pair to modulate the other in FM (period) or AM (volume, 6-bits).
The Acorn Archimedes uses a variation on u-law with the bit order reversed and the sign bit in bit 0. Being a 'minority' architecture, Arc owners are quite adept at converting sound/image formats from other machines, and it is unlikely that you'll ever encounter sound in one of the Arc's own formats (there are several).
Tandy notes (Jeffrey L. Hayes <firstname.lastname@example.org>): The maximum sampling rate for output is at least 44k. (I don't know the maximum rates; I have recorded at 22k and played at 44k. Higher rates are probably possible.) There is one output channel, not three. The belief that there are 3 channels probably stems from the fact that Music.pdm, bundled with these machines, can create 3- channel music modules (analogous to Amiga .mod's). Music.pdm probably does that because it is designed to work with the Tandy's 3-voice tone generator circuitry (compatible with the Texas Instruments SN76496 in the IBM PC-Jr) if there is insufficient RAM to load sound samples. The Tandy chip is able to record at lower rates than it is able to play back, as is the Soundblaster (i.e., the divider used to program the chip to record is lower than that used to program the chip to play back). The Tandy DAC can go faster than the original Soundblaster, however.
The NCD MCX terminal has audio integrated with its X server. The NCDAudio server is an extension of the X server, working together with it, with stress on the networking capability of sound transmission. The NCDAudio API provides format handling (ULAW8, Linear Unsig 8, Linear Sig 8, Linear Sig 16 MSB, Linear Unsig 16 MSB), flowing (to the server, from the server, to the i/o, from the i/o), wave form generators (Square, Sine, Saw, Constant) and the capability of area broadcast using UDP. Provision for manipulating data files (SND, WAV, VOC & AU) is also provided.
CD-I machines form a special category. The following formats are used: