eric@topaz.rutgers.edu (Eric Lavitsky) (03/12/90)
Title: Corrections to DSP Article <In article <41152.25f147bf@drunivac.drew.edu> tlimonce@drunivac.drew.edu writes: <WARNING: This is really long because I wanted to include a lot of <detail. I'm really excited about this product. <I have a BIG SCOOP for everyone. I appreciate Tom's enthusiasm about the Active Circuits, Inc. "Bonsai 2000" (TM) DSP board, but there are a few things that have to be put in perspective... <At the February JAUG (Jersey Amiga Users Group) meeting we were <witness to a new product. I think this will be a real ground-breaking <product for the Amiga; as I explain at the end of this article. <The product was "The Bonsai 2000 DSP board" for the Amiga 2000 by John <Cadona of Cadona Research & Engineering (CRE... pronounced like CRAY). <After 2 years of research & development, this was the first public <showing. The aforementioned DSP board was a product previewed at the February JAUG meeting. No official product or delivery announcements were made. The manufacturer is Active Circuits, Inc. and the principal designer of the board and the person who did the the technical presentation was Dr. Johanan L. Codona. I also did some talking (don't you remember Tom? :-), as I have been involved in the hardware design, have been designing the software architecture for the board and am chiefly responsible for its production. <DSPs In General: <---------------- <The talk began with an explanation of DSPs in general. A DSP (Digital <Signal Processor) is a board that can do the high-speed calculations ^^^^^ DSP is "Digital Signal Processing". DSP Chips are chips which perform DSP operations at high speed and in convenient packaging. <required to do sound manipulation. They were originally designed for <Amiga sound stores 8-bits for every "number" when it digitizes or plays <sound. A CD player uses 16-bits (or 18 ...depending on how you think <about it) per number. A CD player plays at 44.1MHz; this means that it ^^^^ Er - that's 44.1Khz, 44Mhz would be pretty damn impressive and useless for most audio purposes. <The AT&T WE 32c DSP Chip: < ....<stuff deleted< <Since this chip is the choice of chips for scientists; and since it is <*the* DSP chip used by AT&T/Bell-foo/Bell-frob researchers it has a C Tom was a little overzealous here. While in our opinion, the DSP32C is an excellent choice for scientific applications, it is not the only chip available. We feel strongly that the combination of the DSP32C's power and the Active Circuits flexible design, the Bonsai 2000 can fulfill the needs of many of the most demanding scientific applications requiring DSP. <Graphics Applications: <---------------------- <....<stuff deleted< < -- Scientific Visualization often requires a CRAY. How many people <would settle for 1/2 a CRAY all to themselves at this price instead of <millions of dollars for a CRAY that they have to share with other <researchers? A nice thought, but it's not fair to call even two DSP32Cs half a Cray: We're single precision We have less I/O bandwidth through the system Crays can compute in the GigaFLOPS range < -- ALMOST REAL-TIME RAY-TRACING. Ray-tracing is slow because it <requires so much floating point math to be performed. This is what the <DSP does well! At 25 MFLOPS each, two DSP32Cs are nearly 100 times the performance of a 25Mhz 68882. If the ray-tracer can be parallelized to run on both DSPs, you could get some pretty nice performance out of the Bonsai 2000. You won't get real-time ray-tracing for complex scenes, but it should be quite fast for a personal computer. <The Card Itself: <---------------- <John offered to pass the card around. After hearing it's price no body <wanted to take responsibility of passing it person-to-person. It was <pretty funny. Actually, I was the one who offered to pass it around :-) The price quoted was based on what it cost to build the first 5 prototype boards (no economies of scale here). <I was impressed when I got hold of the card. The hardware is finished; <no patches were showing on the board. I remember when the first RAM <cards for the Amiga were released and AmigaWorld did a review of a <bunch of them. They all had hand-soldered last-minute patches all over <them except for the ASDG board, which looked completely professional. <This card looked the same way. It looked like it had been in <production for a while and was done by a very serious developer (which <describes CRE). Thanks - we really pored over our design time after time to make sure we did it right the first time. It's hard to debug a six layer fine-line board and we didn't want to have to do it. There's definitely something to be said for design reviews that involve both the software and hardware people. We did all the schematic capture on a Commodore PC40-III and the layout guys did a terrific job. <Configurations are flexible: You can vary the kind of performance you <want based on how much you can afford to pay: < -- 1 or 2 DSPs running at 50MHz. < -- 0, 32, 64, 128, 256K RAM per DSP. < -- Using PALs, you can reconfigure the memory map of the DSP. < -- The less-expensive "DSP32" can also be used (slower performance). < -- 8Kx8 or 32Kx8 SRAM (8- or 32-bit bus) These specifications are preliminary and are in no way any offical commitment by Active Circuits, Inc. This is what the current design supports, but it may change. <16-bit parallel interface to Amiga w/DMA access to all RAM memory and <control registers. In other words, you can go "straight to the chip". The DMA for the Bonsai 2000 means that you can go straight to the on board memory, expansion port, or second DSP/memory by DMA'ing through the first DSP. This allows access to on board subsystems while all the DSPs continue to run. <General I/O support via daughter boards: Personality modules on each <DSP. Attach samplers, CDs, etc. Intentions are for specifications for daughter board design to be made available for third party developers. <Price: <------ <Native Development Kit is <500 for the software. This does everything <on-chip. Results are echoed back to the Amiga. < <Amiga Development Kit's price is unknown. This lets you compile, etc. <on the Amiga and download code to the DSP. It will include a .library <for accessing the DSP so that all languages can access it. 1/2 a CRAY <accessible from AmigaBASIC! <The boards themselves will run from less than <1500 (1 DSP with no <external RAM) to <4000-<5000 (2 DSPs with 512K RAM). No prices on <"personality modules" or "daughter boards" yet. Ummm - it's really inappropriate to discuss prices for something that hasn't been officially announced as a product. In any event, (and without making any commitments) since it's been brought up, it's expected that the complete developer environment (C Compiler, Assembler, Linker, Simulator, and Amiga device driver support) will be ~$500. It will all run native on the Amiga. No other specifications on support libraries have been announced at this point. It's expected that a 0K board will be closer to $1,000 and a 2 DSP/512K(35ns) (it may be possible to add more memory than this) board will be closer to <4,000. The two DSP model is appropriate for certain scientific and high-end multimedia applications. Sorry - no dates announced for product availability. If you have specific needs or development requirements, please contact Active Circuits, Inc. directly at 201-223-5999. <Implications: <------------- <The NeXT has a Motorola DSP 56001; which is inferior to the AT&T model. <The AT&T chip is the chip-of-choice for serious research. It's faster <and it's floating-point. This means the NeXT has one sophomoric DSP <while your Amiga could have 2 superior DSPs. In article <5943@sbcs.sunysb.edu< root@sbcs.sunysb.edu writes: << << I understand your enthusiasm, but remember that the DSP56K used << in the NeXT is not apples to apples comparison. The DSP 56K is << roughly same generation as the ATT dsp16 or so. Motorola has << a FP DSP called the DSP96000. Rick Spanbauer is correct in his comment here; this is not quite a fair statement. As far as the DSP32C is floating-point and the 56001 is fixed point this is true, but there are still plenty of useful things one can do with a 56001. However, try and find a DSP96000 chip for less than $500, no make that $800... <AT&T/Bell-frobs can buy Bonsai boards because AT&T/Bell-frobs must all <use the AT&T DSP chips. This means it has a large potential market. Umm - I think he means that one of the target users for this board may be AT&T, but this is just speculation. No one particular customer has been targeted for this product, but we would hope that large companies would be interested in the Bonsai and the Amiga as a DSP platform. <Summary: <-------- <Wow! This board can make an Amiga 2500/30 more capable than a NeXT for <many applications. It is the least expensive price/performance ratio <that can be found. The software support is excellent. The design is <great too. It was shown at EuroDevCon '90 and it made many people <interested. I predict that if this board is marketed correctly it can <really make the scientific community take notice of the Amiga. All <this comes at a time when C-A is stepping up its college/ <university-orientented marketing. We certainly hope so. <-Tom <(John Cadona is on the network and might be reading this, hopefully <he'll correct any factual errors.) John is not available on Usenet. Any questions or comments regarding the ACI Bonsai 2000 should be directed to me or directly to Active Circuits, Inc. at 201-223-5999. -Eric Bonsai 2000 is a trademark of Active Circuits, Inc. Amiga is a registered trademark of Commodore-Amiga, Inc. Eric Lavitsky Active Circuits, Inc. TechStream/ACI/201-223-5999 106 Hwy 71, Suite 101 Running Amiga UUCP on an Amiga 2500 Manasquan, NJ 08736 -- Eric Lavitsky UUCP: ...rutgers!tstream!eric Active Circuits, Inc. eric@topaz.rutgers.edu 201-223-5999 BIX: eric
panon@cheddar.cc.ubc.ca (Paul-Andre Panon) (03/12/90)
<<In article <41152.25f147bf@drunivac.drew.edu> tlimonce@drunivac.drew.edu writes: <<I was impressed when I got hold of the card. The hardware is finished; <<no patches were showing on the board. I remember when the first RAM <<cards for the Amiga were released and AmigaWorld did a review of a <<bunch of them. They all had hand-soldered last-minute patches all over <<them except for the ASDG board, which looked completely professional. <<This card looked the same way. It looked like it had been in <<production for a while and was done by a very serious developer (which <<describes CRE). << <<-Tom < In article <Mar.11.11.01.40.1990.12117@topaz.rutgers.edu> eric@topaz.rutgers.edu (Eric Lavitsky) writes: < I also did some talking (don't you remember Tom? :-), <as I have been involved in the hardware design, have been designing the <software architecture for the board and am chiefly responsible for its <production. < <-Eric < <-- <Eric Lavitsky UUCP: ...rutgers!tstream!eric <Active Circuits, Inc. eric@topaz.rutgers.edu <201-223-5999 BIX: eric < I seem to remember that Mr. Lavitsky was initially strongly associated with ASDG (maybe even a founding member?) back in the early days (daze? :-)) of the Amiga. Perhaps this is part of the reason why the design and layout quality is so high and why Tom might find it reminiscent of early ASDG products. -- Paul-Andre_Panon@cheddar.cc.ubc.ca or USERPAP1@UBCMTSG or Paul-Andre_Panon@undergrad.cs.ubc.ca or USERPAP1@mtsg.ubc.ca "What should the role of the University be? It should be to enlighten Society." -Luis Sobrino
kosma%stc-sun@stc.lockheed.com (Monty Kosma) (03/16/90)
<Graphics Applications: <---------------------- <....<stuff deleted< < -- Scientific Visualization often requires a CRAY. How many people <would settle for 1/2 a CRAY all to themselves at this price instead of <millions of dollars for a CRAY that they have to share with other <researchers? A nice thought, but it's not fair to call even two DSP32Cs half a Cray: We're single precision We have less I/O bandwidth through the system -> Crays can compute in the GigaFLOPS range not typically. Depends on what Cray. A multiple-headed Y/MP can get close (still can't beat my Connection Machine, though), but the cray X-MP gets one or two hundred MFlops, and the cray 1 got only about 50 or 60. sorry, I know this isn't comp.sys.supercomputers, but I can't resist the opportunity to cray-bash :-).
limonce@pilot.njin.net (Tom Limoncelli) (03/16/90)
In article <14002@snow-white.udel.EDU> kosma%stc-sun@stc.lockheed.com (Monty Kosma) writes: > Tom Limoncelli wrote: > < -- Scientific Visualization often requires a CRAY. How many people > <would settle for 1/2 a CRAY all to themselves at this price instead of > <millions of dollars for a CRAY that they have to share with other > <researchers? Looking at my notes, I see that the "1/2 a cray" phrase was coined by John Cadona. I understand that it might have been a hyperbole. [ cray-bashing and cray-defending deleted ] > sorry, I know this isn't comp.sys.supercomputers, but I can't resist the > opportunity to cray-bash :-). Wait a second; some really nice guys from Cray almost saved my life at Usenix last January. Don't knock Crays! :-) -Tom -- tlimonce@drew.edu Tom Limoncelli As seen in USA Today & tlimonce@drew.uucp +1 201 408 5389 Rec.Humor.Funny! tlimonce@drew.Bitnet Stock quote: Commodore stock closed limonce@pilot.njin.net at $8.50 (-.25) on 3-14-1990.