[comp.sys.amiga] Amiga DSP Boards.

WHE46@CCVAX.IASTATE.EDU (Marc Barrett) (11/09/90)

 In a message someone commented:

> There are DSP boards available for the Amiga.

   What is the installed base of Amigas with DSP boards?  Far, far less
than the installed base of NeXT systems with DSPs, to be sure.  I doubt
there are 100 Amiga with DPSs world-wide, compared to about 30,000 NeXT
systems with DSPs.

   Another problem with these DSP boards for the Amiga is that they are
totally useless, because they have no software available for them.  If
I were to purchase one of these DSP boards, I would have absolutely no
use for it, because I wouldn't be able to do anything with it.  If I
wanted to use it to demonstrate 16-bit stereo sound, I would be unable
to because I would have to write my own sound composition programs for
the thing, and then compose my own songs.  It would be simpler to just
get a NeXT.

   Another problem is price.  The NeXT systems are quite good deals,
and these Amiga DSP boards are quite expensive.  An Amiga with a DSP
board is more expensive than a NeXT, with less DSP software.

   Before I get torched by all the trigger-happy people around here,
let me clarify myself.  I am NOT bashing the Amiga with this message.
I am bashing the Amiga DSP boards.  I am pointing out that, before
the availability of DSP boards is going to have any impact on the
market as a whole (or the NeXT), there have to be more of these
DSP boards available, they have to all be totally compatible with
each other, and they must drastically come down in price.  Since
the DSP systems are too expensive and have no software (and are
therefore totally useless to most people), these boards might as
well not exist at all.

   I do have a solution for some of the problems with DSP boards
for the Amiga.  There are plenty of programs for the Amiga that
will play sound via MIDI synthesizers.  The SMUS standard is totally
compatible with MIDI, and most Amiga multimedia software (including
AmigaVision) supports MIDI.  The way to make these DSP boards more 
useful is to make them compatible with programs that support MIDI.

   In essence, here is how it could be done.  I would like to see
these Amiga DSP boards modified so that they support 16-bit stereo
sound, and communicate with Amiga software in exactly the same way
that a MIDI synthesizer (connected to the Amiga via a MIDI interface)
does.  The same board should also include a hi-fi 16-bit stereo sound
digitizer, with both the sound synthesis and digitization supported
by the DSP.

   Some people might ask "Why not just get a MIDI synthesizer and
a MIDI interface?"  The answer is that such modified DSP boards
could be more desireable because it would be complete and on a single
card, and would not have the tangled mess of cables and large amount
of space that a synthesizer does.   It would also include both 16-bit
sound synthesis and digitation on a single integrated card.

   If this is done, then these resulting DSP boards would actually
be useful to the average person -- much more useful even than a NeXT
system.  

                                   -MB-

David.Plummer@f70.n140.z1.FIDONET.ORG (David Plummer) (11/18/90)

I would suggest to you that the low, make that very low, proportion of
amigas that are equipped with DSP boards would be indicative of the 
general market's need for such boards.  Given that, it would seem that
the inclusion of said hardware with every NeXT system is a waste, since
95% of NeXT owners therefore pay for hardware that only 5% of them really
need or use.
 
Granted, there are people who need these boards, and they'll either buy
a NeXT or become the 101st Amiga owner to add such a board.  All in all,
I don't feel that Digital Signal Processing is a mainstream need, and in
no way reflect the credibility of a machine or it's users.  
 
Not to mention the fact that attempts to be all things to all people is
rarely successful.  To me, a machine that combines DSP abilities with
onscreen PostScript display seems to have a poor sense of self-purpose.
 
-----------------------------------------------------------------------
=     plummerd@uregina.ca or plumme@hercules.uregina.ca (BITNET)      =
=---------------------------------------------------------------------=
= Amiga__ If I had any good   | Make it possible for programmer's to  =
= __  /// opinions, I'd sell  | write code in English and you'll soon =
= \\\///  them, not give them | discover that programmers cannot      =
=  \XX/   away here for free! | write in English.                     =
-----------------------------------------------------------------------




--  
David Plummer - via FidoNet node 1:140/22
UUCP: ...!herald!weyr!70!David.Plummer
Domain: David.Plummer@f70.n140.z1.FIDONET.ORG
Standard Disclaimers Apply...

jeh@sisd.kodak.com (Ed Hanway) (11/21/90)

David.Plummer@f70.n140.z1.FIDONET.ORG (David Plummer) writes:
>I would suggest to you that the low, make that very low, proportion of
>amigas that are equipped with DSP boards would be indicative of the 
>general market's need for such boards.

DSPs won't catch on until someone develops an application for them.
A DSP+ADC+DAC combination, with the proper software, should have many,
many applications in music, and would probably sell Amigas just like the
MIDI port sold Atari STs.  The problem is that at this point, I don't know
of any software being delivered with DSPs, either for the Amiga or any
other platform, other than "developer's toolkits."

>                                        Given that, it would seem that
>the inclusion of said hardware with every NeXT system is a waste, since
>95% of NeXT owners therefore pay for hardware that only 5% of them really
>need or use.
 
I agree that, presently, most NeXT users do not really use the DSP.
However, I've heard that the Motorola 5600mumble used in the NeXT is quite
inexpensive in quantity, so including the DSP is probably not as big a deal
as NeXT would have us believe.  Besides, just because it isn't useful today
doesn't mean that it can't be useful someday.  How many Amiga owners used
hold-and-modify mode in 1985?  How many use the analog joystick inputs today?

--
Ed Hanway --- uunet!sisd!jeh
Must be 18 or older to play.  Prerecorded for this time zone.
Do not read while operating a motor vehicle or heavy equipment.

rehrauer@apollo.HP.COM (Steve Rehrauer) (11/21/90)

In article <944.27477729@weyr.FIDONET.ORG> David.Plummer@f70.n140.z1.FIDONET.ORG (David Plummer) writes:
>I would suggest to you that the low, make that very low, proportion of
>amigas that are equipped with DSP boards would be indicative of the 
>general market's need for such boards.  Given that, it would seem that
>the inclusion of said hardware with every NeXT system is a waste, since
>95% of NeXT owners therefore pay for hardware that only 5% of them really
>need or use.
> 
>Granted, there are people who need these boards, and they'll either buy
>a NeXT or become the 101st Amiga owner to add such a board.  All in all,
>I don't feel that Digital Signal Processing is a mainstream need, and in
>no way reflect the credibility of a machine or it's users.  

Uhhh, and the first mainframe makers thought there was a worldwide
market for at most, what, 10 machines?  And H.P. underestimated the
market for their first handheld scientific calculator by, what, two
orders of magnitude?

What I'm getting at is that I think it's an enabling technology.
Obviously the general market can exist without it, because they have.
But if people could have STANDARD voice-mail & fax capabilities on their
desk in the same box that crunches their Lotus spreadsheets (i.e.: without
needing to be intimidated by the process of figuring out what optional
'wares to buy), I think a DSP would quickly become a "necessity" in their
minds.  (Or rather, the features that a DSP makes easier and more cost-
effective to provide would become the necessity.)

We seem to be at an odd point in time, in which hardware gets niftier
faster than software can be written to really take advantage of it.
I think there's a number of technologies (CD-ROM, DSPs, video-in-a-
window, to name a few) that we haven't begun to fully exploit, and
not because the hardware is really exhorbitantly priced anymore.

(No "the Amiga will die because it doesn't have XYZ" message is implied.)
--
The goons are riding motorcycles, but WE'VE  | (Steve) rehrauer@apollo.hp.com
got a whole big metal car! This will be like | The Apollo Systems Division of
stepping on ants... -- Freelance Police      |       Hewlett-Packard

rick@ameristar (Rick Spanbauer) (11/21/90)

In article <1990Nov20.161034.25281@sisd.kodak.com> jeh@sisd.kodak.com (Ed Hanway) writes:
>DSPs won't catch on until someone develops an application for them.
>A DSP+ADC+DAC combination, with the proper software, should have many,
>many applications in music, and would probably sell Amigas just like the
>MIDI port sold Atari STs.  The problem is that at this point, I don't know
>of any software being delivered with DSPs, either for the Amiga or any
>other platform, other than "developer's toolkits."

In my opinion, DSP is an area where CBM needs to dive in and set a few 
standards, probably by producing a piece of hardware.  As things stand 
there are too many degrees of freedom standing in the way of a defacto 
standard:

	* Which DSP?  The major players (AT&T, Motorola, TI, Analog Devices)
		      all have differing word lengths, instruction sets,
		      architectures, chip resources, etc.

	o Standard I/O resources?  Should the board have 14, 16, or 18 bit
		      a/d, d/a?  What sampling rate (variable)?  What
		      pre/post filtering?  A telephone interface for modems?

	o IPC model?  Shared memory?  Mailboxes?  Serial interface?

	o Memory available?

It is easy to conjure up a whole slew of applications, eg fax/data modems, 
voice response systems, music IF a standard, low cost board was produced by 
CBM.  Until this happens, DSP will remain a niche market.

					Rick Spanbauer
					Ameristar

laughlin@fornax.UUCP (Bob Laughlin) (11/24/90)

In article <1990Nov20.161034.25281@sisd.kodak.com> jeh@sisd.kodak.com (Ed Hanway) writes:
>DSPs won't catch on until someone develops an application for them.
>A DSP+ADC+DAC combination, with the proper software, should have many,
>many applications in music, and would probably sell Amigas just like the
>MIDI port sold Atari STs.  The problem is that at this point, I don't know
>of any software being delivered with DSPs, either for the Amiga or any
>other platform, other than "developer's toolkits."

   The "developer's toolkit" available for the NeXT is quite sophisticated.
For example an Orchestra object coordinates all DSP operations at a high
level of abstraction, handling such things as SynthPatch objects that
implement synthesis algorithms, which in turn utilize UnitGenerator classes
that do simple things like behaving as oscillators. I'm kind of surprised
the NeXT MusicKit has'nt caught on---although it has at places like 
CCRMA at Stanford, Princeton Music dept, MIT Media Lab etc.  The price of
a NeXT is a bit much for the average working musician; a void filled nicely
by the Atari.

   I think the Amiga 2000/3000 would be an excellent computer for a DSP
board.  Believe me these things will take off. There are several DSP
boards in the works for the Atari and several available now for Macs
and PC's.  I've seen one ad for an Amiga board called AudioLink from
Beta Unlimited in Brooklyn. At least I thought it was a DSP board.
The brochure they sent me only mentions 16 bit sampling and playback,
MIDI, sample editing, and up to 16megs om the board.  Has anybody
seen or used one of these?  There was also somebody on the net about
6-12 months ago soliciting feedback on a DSP board in the works.
Anybody know what happened to the project?

-- 
 Bob Laughlin  laughlin@cs.sfu.ca 

p554mve@mpirbn.mpifr-bonn.mpg.de (Michael van Elst) (11/26/90)

In article <1990Nov21.013237.4623@ameristar> rick@ameristar (Rick Spanbauer) writes:
>	* Which DSP?  The major players (AT&T, Motorola, TI, Analog Devices)
>		      all have differing word lengths, instruction sets,
>		      architectures, chip resources, etc.
>	o Standard I/O resources?  Should the board have 14, 16, or 18 bit
>		      a/d, d/a?  What sampling rate (variable)?  What
>		      pre/post filtering?  A telephone interface for modems?
>	o IPC model?  Shared memory?  Mailboxes?  Serial interface?
>	o Memory available?

Well, if these parameters were fixed you won't benefit from future
enhancements of todays DSPs. I think the major point would be defining
data structures (f.e. file formats) that you can agree on storing
and processing information.
Another point would be an application interface for standard functions
but you'd need board specific routines for low-level operations.

Regards,
-- 
Michael van Elst
UUCP:     universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
Internet: p554mve@mpirbn.mpifr-bonn.mpg.de
                                "A potential Snark may lurk in every tree."

rick@ameristar (Rick Spanbauer) (11/27/90)

In article <1390@mpirbn.mpifr-bonn.mpg.de> p554mve@mpirbn.UUCP (Michael van Elst) writes:
>In article <1990Nov21.013237.4623@ameristar> rick@ameristar (Rick Spanbauer) writes:
>>	* Which DSP?  The major players (AT&T, Motorola, TI, Analog Devices)
>>		      all have differing word lengths, instruction sets,
>>		      architectures, chip resources, etc.
>>	o Standard I/O resources?  Should the board have 14, 16, or 18 bit
>>		      a/d, d/a?  What sampling rate (variable)?  What
>>		      pre/post filtering?  A telephone interface for modems?
>>	o IPC model?  Shared memory?  Mailboxes?  Serial interface?
>>	o Memory available?
>
>Well, if these parameters were fixed you won't benefit from future
>enhancements of todays DSPs. I think the major point would be defining
>data structures (f.e. file formats) that you can agree on storing
>and processing information.
>Another point would be an application interface for standard functions
>but you'd need board specific routines for low-level operations.

	DSP's are yet fast enough that they can run some standard threaded 
	language or interpreted instruction set that would abstract the 
	basic machine in a useful way.  Much the same way as the Amiga is 
	wedded to the 680x0 instruction set, a standard DSP must be settled 
	on if we are ever to see a rich application set emerge.

>Michael van Elst

					Rick Spanbauer
					Ameristar

danb20@pro-graphics.cts.com (Dan Bachmann, SubOp) (11/27/90)

In-Reply-To: message from laughlin@fornax.UUCP

        I know that Active Circuits makes two DSP boards for the A2000 &
A3000.  That's all really know about it.
-------------------------------------------------------------------------
 ProLine: danb20@pro-graphics           ***************************
    UUCP: ...crash!pro-graphics!danb20  *       Dan Bachmann      *
ARPA/DDN: pro-graphics!danb20@nosc.mil  *  Raritan Valley College *
Internet: danb20@pro-graphics.cts.com   ***************************
  P-Link: DanB20        <-- I only read PLink once a mo. use Internet
U.S.Mail: 509 StonyBrook Drive, Bridgewater, NJ 08807

p554mve@mpirbn.mpifr-bonn.mpg.de (Michael van Elst) (11/28/90)

In article <1990Nov27.020113.13352@ameristar> rick@ameristar (Rick Spanbauer) writes:
>	DSP's are yet fast enough that they can run some standard threaded 
>	language or interpreted instruction set that would abstract the 
>	basic machine in a useful way.  Much the same way as the Amiga is 
>	wedded to the 680x0 instruction set, a standard DSP must be settled 
>	on if we are ever to see a rich application set emerge.

Well, that depends. Most current DSPs are as fast as they are because
of their use of parallelism. I don't think that an abstract language
can benefit much from this unless you have a _very_ smart compiler which
has yet to be built. I'm using a DSP but I'd sacrify more than 50% of its
performance if I didn't make use of this parallelism.

Regards,
-- 
Michael van Elst
UUCP:     universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
Internet: p554mve@mpirbn.mpifr-bonn.mpg.de
                                "A potential Snark may lurk in every tree."

bbs00010@uafcseg.uucp (Don Kennedy) (11/28/90)

Yes, you haven't heard much from us lately, but we're still out here. Vision
Quest is now in a critical negotiation phase with SunRize Industries to 
manufacture and distribute our DSP 56001 based 16 bit audio card. The prototype exists and records to hard drive in stereo @48khz sampling rate. SunRize is 
nearing completion of a video-oriented 16 bit sample editor. We are toying
seriously with the idea of building a full fledged audio workstation. 

The board has stereo A/D and D/A onboard, and may have AES/EBU onboard. It 
will have connectors for daugherboards, which may include a DRAM controller
and/or a SCSI controller. We are also investigating hardware data compression.

Contact us or SunRize for more info or if you would like to be involved in
software development. We are working on a deal for a shareware 56001 assembler.
We are currently using Motorola's Assembler/Simulator for the PC. If you'd like
to take a shot at NeXT's PD assembler source, let us know.

Don Kennedy
Vision Quest
(501) 521-0420
(501) 253-5264

email will be forwarded from kcampbel@uafhp.uark.edu

phil@b11.ingr.com (Phil Johnson) (11/30/90)

>seen or used one of these?  There was also somebody on the net about
>6-12 months ago soliciting feedback on a DSP board in the works.
>Anybody know what happened to the project?
>

I had solicated input for a Zorro-based DSP board last year and had some
interesting discussions with several people (three if my memory serves).
I was planning to design the board (Motorola 56001-based) as a personnel
edification project (versus commercial venture).  I designed and built
the DSP side but did not have the time or patience to decypher the Zorro
bus interface.  I wrote a VERY crude DSP56001 assembler to test the board.

I believe that adding DSP (both single and floating point) capability to
the Amiga would open areas such as audio sampling and reproduction, and
voice recognition and generation that would put the present Amiga cap-
abilities to shame.  Try to imagine you Amiga with a voice that is clear
an completely understandable.  Also, imagine using voice rather than your
mouse to navigate through applications.

I am NOT suggesting that the Amiga will wither away without DSP.  I am only
saying WHY NOT expand the horizons of the Amiga.  I will not get into a
religious war over the pros and cons of the Amiga, but I will say that I have
seen a number of Amiga user sell their Amiga and buy IBM clones or Macs for
the simple reason that they can get their work done on clones and Macs and
can not wait for Commodore to spend another two to three years promising
great things and delivering more promises.  I owned an A1000 and now own a
2000, so I am not a non-Amigan dumping on Commodore, but rather a slightly
frustrated Amigan.