ja26612@uxa.cso.uiuc.edu (*** Jeff ****) (07/15/90)
I posted this before, but got no responses. Is it true that Commodore is
developing a new 16-bit DSP that will be used inside Amigas? I heard Info
and Amazing Computing wrote this in their rumor columns. Is there anyone
from Commodore that can affirm or deny this? How about a "no comment?"
-------------------------------------------------------------------------------
Jeff Axelrod | "If you chose not to decide, you still have made
| a choice" -Neil Peart, Rush ///
ja26612@uxa.cso.uiuc.edu | \XX/ AMIGAUSER
-------------------------------------------------------------------------------Chuck.Phillips@FtCollins.NCR.COM (Chuck.Phillips) (07/15/90)
Jeff> I posted this before, but got no responses. Is it true that Commodore is Jeff> developing a new 16-bit DSP that will be used inside Amigas? Please, oh, please let it be so! A 16 bit DSP chip with at least a near CD sampling rate is essential for studio quality sound. I think the availability of supported 16 bit 44+MHz stereo+ sound would really bolster the Amiga's image as a quality multimedia/desktop-video machine. In fact, if it were possible to add lots of DSP chips with discrete outputs, say 8 or so on expansion cards, the Amiga could become the center of a (relatively) low cost all digital recording studio. (Just add multiple midi ports and a direct to DAT interface... ;^) I realize there are some technical problems to overcome (like _massive_ sustained data transfers), and it may be this could only be featured on the faster 030 models, but please consider it. AmigaOS seems to have a lot of potential for real time applications, and this might just be the application to prove it. Just another $0.02 from, -- Chuck Phillips MS440 NCR Microelectronics Chuck.Phillips%FtCollins.NCR.com Ft. Collins, CO. 80525 uunet!ncrlnk!ncr-mpd!bach!chuckp
psycho@mango.ucsb.edu (Scot Kellan Forbes) (07/15/90)
In article <1990Jul14.173215.25433@ux1.cso.uiuc.edu> ja26612@uxa.cso.uiuc.edu (*** Jeff ****) writes: >I posted this before, but got no responses. Is it true that Commodore is >developing a new 16-bit DSP that will be used inside Amigas? I heard Info >and Amazing Computing wrote this in their rumor columns. Is there anyone >from Commodore that can affirm or deny this? How about a "no comment?" If this sucker is actually on the drawing board (in the CAD program?) how about 8 voices? Perhaps 16? I think Amiga would be very competitive in the music industry if you could use it as a 16-bit sampler as well as a sequencer, perhaps even syncing live video using AmigaVision (it is rumored that AmigaVision will support MIDI in the futuer. Even now, all you would need is a sequencer that supports ARexx). Way back when someone made a comment to the effect of musicians would not buy Amigas for the purpose of sampling sound because they could just as easily buy a dedicated sampling keyboard or module. I can think of at least three reasons why they might Amigas with 16-bit audio and actually use its capabilites: 1) It would probably cost less than most 16-bit samplers. (The upgrade to 16-bit Paula, that is) 2) One less piece of equipment to set up/carry along/program etc... 3) One less piece of equipment to syncronize. ...Just a little fantasy to keep my mind off of things :) -Scot Forbes "Paradise is exactly like where you are right now, only much, much better!" - Laurie Anderson
jk87377@korppi.tut.fi (Kouhia Juhana Krister) (07/16/90)
In article <5958@hub.ucsb.edu> psycho@mango.UUCP (Scot Kellan Forbes) writes: >In article <1990Jul14.173215.25433@ux1.cso.uiuc.edu> ja26612@uxa.cso.uiuc.edu (*** Jeff ****) writes: >>a new 16-bit DSP that will be used inside Amigas > >If this sucker is actually on the drawing board (in the CAD >program?) how about 8 voices? Perhaps 16? I think Amiga would be Unpractical, it's better make it with two (stereo) and use direct sampling from harddisk. How many voices Amiga can handle through DMA? And same time there is a program which handles MIDI data. I like Amigas DA chip, because it read samples automatically. (Without my program.) Have somebody done direct sampling from harddisk yet? Juhana Kouhia jk87377@tut.fi
visinfo@ethz.UUCP (VISINFO c/o Sascha Schnapka) (07/16/90)
In article <1990Jul15.215912.25094@funet.fi> jk87377@korppi.tut.fi (Kouhia Juhana Krister) writes: > >I like Amigas DA chip, because it read samples automatically. >(Without my program.) >Have somebody done direct sampling from harddisk yet? Yes. Chris Weber has written a programm called LargeSoundPlayer which is able to play very long IFF Sound Files directly from hard drive with up to 29 khz sampling rate. The problem is sampling to hard disk. We made some 3.5 MB long files using 4 MB fast ram. Sampling directly to hard disk requires a buffer in the sampler itself, because it is impossible to sample and write to the hard drive at the same time. >Juhana Kouhia >jk87377@tut.fi /* -------------------------- SG (Simeon Graphics) ---------------------- */ /* Peter Simeon, UUCP: | // // */ /* visinfo@bernina.ethz.ch | // Long live the AMIGA! // */ /* BIX: hardwiz | \X/ \X/ */ /* ---------------------------------------------------------------------- */
jk87377@korppi.tut.fi (Kouhia Juhana Krister) (07/17/90)
In article <5119@ethz.UUCP> visinfo@bernina.ethz.ch.UUCP (VISINFO c/o Peter Simeon) writes: > >Yes. Chris Weber has written a programm called LargeSoundPlayer which is able >to play very long IFF Sound Files directly from hard drive with up to 29 khz >sampling rate. The problem is sampling to hard disk. We made some 3.5 MB long >files using 4 MB fast ram. Sampling directly to hard disk requires a buffer >in the sampler itself, because it is impossible to sample and write to the >hard drive at the same time. How difficult it's make such kind of sampler (AD-converter) that uses same technique like as Paula. I mean DMA technique. Is such kind of card exist yet? (8 or 16 bits) I asked a couple weeks ago if there is exists DAT interface card to the Amiga. I think it needs only one dsp-prosessor (read Amigas memory via DMA and send data to/from S/PDIF chips). Sony has a couple chips which handle S/PDIF format: one for decoding and one for encoding. Is it possible read Amigas memory 2*2*44100 bytes per second via DMA? (Stereo, 16 bit, and same time read data from harddisk.) Of course, it's possible to connect S/PDIF connector to amplifier, which has digital interface. Amplifier have AD/DA-converters, so, there's no need to buy them separately. Amplifier costs less than $1000 here, and in USA more less than here. Or am I right? Juhana Kouhia jk87377@tut.fi
vik@lynx.uucp (Vikram Sohal) (07/20/90)
>Have somebody done direct sampling from harddisk yet?
We have done it under LynxOS, our real-time UNIX-compatible OS running
on a (deep intake of breath) 386-AT clone. Using double buffering,
we are able to easily read or write at 176K per second from a SCSI disk and
write/read to a Analog I/O board for CD-quality sound. Using SCSI, we can
actually get up to a 500Khz sampling rate without the use of any special
chips. I imagine that an Amiga version of our software would run even faster
because of the tightly integrated DMA architecture of the Amiga bus.
----------------------------------------------------------------------------
Vic Sohal - vik@lynx.uucp
Lynx Real-Time Systems
----------------------------------------------------------------------------nut@wet.UUCP (adam tilghman) (07/24/90)
>>Have somebody done direct sampling from harddisk yet? > > There are at least two direct-to-disk sample recorders for the Atari ST, and seeing that the ST and the Amiga are not too different, I would assume that it is entirely possible on the Amiga. According to several reviews and ads, these samplers are able to keep up with four-channel CD-quality sound (math works that out to be around 176Khz). But you had better have a _REALLY BIG_ hard disk to do any type of large-scale sampling with that beast (only 400 Seconds == 6.75 minutes on a 1Gb drive!) Adam -- == Adam G. Tilghman - Trendy Quote: "Beware of Greeks bearing Trojans!" == ============ Bang-Path UUCP: {uunet | ucbvax}!unisoft!wet!nut ============ =============== Disclaimer? My employer? What, me work? ================
jk87377@korppi.tut.fi (Kouhia Juhana Krister) (07/24/90)
In article <1383@wet.UUCP> nut@wet.UUCP (adam tilghman) writes: >>>Have somebody done direct sampling from harddisk yet? >> >ads, these samplers are able to keep up with four-channel CD-quality sound >(math works that out to be around 176Khz). But you had better have a >_REALLY BIG_ hard disk to do any type of large-scale sampling with that >beast (only 400 Seconds == 6.75 minutes on a 1Gb drive!) For some reason I don't like quantity "Gb" = "Gbits". Stereo CD-quality sound: 2 bytes * 2 channels * 44100 samples * 10 mins * 60 seconds = 101 MB That's not a huge! And ten minutes is enough. If piece is only 5 mins, there is still 5 mins for instruments and singing voices. Useful program for making a composition is Csound or Cmix. Juhana Kouhia jk87377@tut.fi
root@dialog.stgt.sub.org (Christian Motz) (07/26/90)
In article <1383@wet.UUCP> nut@wet.UUCP (adam tilghman) writes: >>>Have somebody done direct sampling from harddisk yet? >> >> > There are at least two direct-to-disk sample recorders for the Atari ST, >and seeing that the ST and the Amiga are not too different, I would assume >that it is entirely possible on the Amiga. According to several reviews and >ads, these samplers are able to keep up with four-channel CD-quality sound >(math works that out to be around 176Khz). But you had better have a >_REALLY BIG_ hard disk to do any type of large-scale sampling with that >beast (only 400 Seconds == 6.75 minutes on a 1Gb drive!) This is probably the reason why CD's only play 4 Minutes, is it? ;-) Uhmmm, perhaps you should re-check your math. Now let' s see ... we have 44 kHz sampling frequency, which leaves us with 44000 16- Bit words every second. Double that, since we want to have stereo sound, right? Ok, that makes up 176000 bytes per second, or 10560000 bytes per minute. That's almost exactly 10 MBytes per minute, leaving us with about 600 MBytes per hour. Does this value strike you as familiar? Surprise!!!! It's the average capacity of the standard Compact Disc ... -- Christian Motz root@dialog.stgt.sub.org
greg@walt.cc.utexas.edu (Greg Harp) (07/26/90)
In article <2261@dialog.stgt.sub.org> root@dialog.stgt.sub.org (Christian Motz) writes: >In article <1383@wet.UUCP> nut@wet.UUCP (adam tilghman) writes: >>>>Have somebody done direct sampling from harddisk yet? >>> >>> >> There are at least two direct-to-disk sample recorders for the Atari ST, >>and seeing that the ST and the Amiga are not too different, I would assume >>that it is entirely possible on the Amiga. According to several reviews and >>ads, these samplers are able to keep up with four-channel CD-quality sound >>(math works that out to be around 176Khz). But you had better have a >>_REALLY BIG_ hard disk to do any type of large-scale sampling with that >>beast (only 400 Seconds == 6.75 minutes on a 1Gb drive!) I'd be hard-pressed to name any, but I've heard rumors. It should be possible, though. Certainly a harddrive (non-DMA :-) could store the info fast enough. Anyone have bandwidth/specs handy to confirm or deny this? >This is probably the reason why CD's only play 4 Minutes, is it? ;-) >Uhmmm, perhaps you should re-check your math. Now let' s see ... we >have 44 kHz sampling frequency, which leaves us with 44000 16- Bit >words every second. Double that, since we want to have stereo sound, >right? Ok, that makes up 176000 bytes per second, or 10560000 bytes >per minute. That's almost exactly 10 MBytes per minute, leaving us >with about 600 MBytes per hour. Does this value strike you as >familiar? Surprise!!!! It's the average capacity of the standard >Compact Disc ... I'm not gonna go through the math again, but the CD audio standard (Phillips/Sony) provides for 75 minutes of music at 16 bits and 44.1 kHz. Of course, lots of tracking information and even graphics data are stored on these disks, too... However, these are printed, read-only disks. Anyone know what the highest storage acheived on WORM or other writeable CD currently is? >-- >Christian Motz root@dialog.stgt.sub.org ..greg... ___ Disclaimer: The opinions expressed above are not my own, but AMIGA! //// the property of some higher-up power, to which I am only a tool. //// "Welcome, my son. Welcome to the machine." -- Pink Floyd ___ //// "Reality is only a simulation, and it's still in beta testing." -- Me \\\\//// \\XX// Greg Harp greg@ccwf.cc.utexas.edu