| Title: | * * Computer Music, MIDI, and Related Topics * * |
| Notice: | Conference has been write-locked. Use new version. |
| Moderator: | DYPSS1::SCHAFER |
| Created: | Thu Feb 20 1986 |
| Last Modified: | Mon Aug 29 1994 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2852 |
| Total number of notes: | 33157 |
I have asked before if other COMMUSIC readers are interested in 'old style'
computer music generation, a la MUSIC5 or MUSIC4BF, for example.
Well, now that the uVAX I is becoming available via the Employee Purchase
Program, it seems possible do buy a computer with sufficient (though admittedly
not overwhelming) power to do so.
My questions along these lines are:
1) How fast is a uVAX I? VAX-11/730 class? I assume that a 730 class
machine would be able to handle four channels at 40kHZ. Am I wrong?
2) Is a Qbus D/A converter available? How much does it cost?
How many channels at what speed?
3) Is a Qbus 9-track mag tape drive available? How much does it cost?
What tape density(ies) does it write?
4) Are additional and perhaps denser disk drives available? How much
do they cost? One 31MB wini doesn't hold very much music.
5) Can we get a Vaxstation-1 through the Employee Purchase Program?
How much does it cost? It would be worth the money to be able
to have decent graphics.
6) Can one get a copy of Vercoe's MUSIC-11 that runs on a uVAX?
How much does it cost? It should be available since I have
heard they are running on a 750 (albeit under UNIX). Maybe
some of you ex-MITers know the answer to this one.
7) Are there better music programs of the MUSIC5/4BF strain available
in the public domain? Isn't there a Stanford tape available?
Would that be of interest here?
8) Should those of us interested in this type of computer music wait
for the uVAX II to be available through the Employee Purchase Program?
Given the customer announcement hasn't happened yet, I would expect it
to be at least a year before the EPP makes them available to us.
Probably longer since the uVAX II will sell!
-jim
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 78.1 | PIPA::JANZEN | Thu Apr 04 1985 10:01 | 29 | ||
The AAV11 is the Q-Bus D/A. It has four channels but read the spec. The fourth channel has special options. 12 bits, probably 250MHz. I don't see a tape controller, but I'm sure an outside vendor like Emulex makes one. You're talkin big money. 40KHz*4 is 160K Words/second. The Q-Bus poops between 125 and 250KW/s. We were building a general purpose parallel interface between 780's and 11/23's in Costa Mesa before DEC closed the office last year. I think 200KW/s was about the limit. That's too close, even DMA's. UNIBUS has problems at that rate, too. That's why we sold the A/D controller we had. I even signed off one I made a small change in; anyway it was only for the 11/70 (was it called ADC70-F?) on the 32-bit S-Bus (Mass Bus) and every bit was packed as far as it could go, i.e., 12 bits packed tightly overlapping the 16-bit boundaries, and even then they had problems getting it out to disk without special disk management techniques and shutting down the rest of the machine. It did not have clocks or the A/D or anything, just the logic to get the data into the 70 memory. This is a non-trivial problem. It might be nice if there were a custom memory with a port out the back. You could stuff a couple megawords (14 seconds) into it slowly, and then tell it to burst it out the back end into your convertors and audio system. Having fast I/O is much harder than building fast CPU's. Make sure you get figures on I/O performance. You don't really care about CPU performance, not first anyway. I'd forget the whole thing. Buy a big-screen TV instead. THomas | |||||
| 78.2 | PIPA::JANZEN | Thu Apr 04 1985 19:40 | 13 | ||
================================================================================ ORPHAN::LIONEL MicroVAX's Kinky Hints 4-APR-1985 17:06 Note 148.1 -< MicroVax Perofmrance >- 1 of 1 -------------------------------------------------------------------------------- The MicroVAX I is, overall, about the speed of a 730 for non-FP calculations, and somewhere between a 730 and 750 for FP (assuming no FPA in the 750). That makes it about half a 750 and a third of a 780. I/O throughput is highly dependent on the type of device you are talking to. My understanding is that the QBUS can perform nearly as well as a UNIBUS in many situations, but most QBUS devices are slow. The RC25 is an exception - it is quite fast. Steve | |||||
| 78.3 | SAUTER::SAUTER | Fri Apr 12 1985 18:11 | 6 | ||
I agree with .1--it costs big bucks to get the I/O bandwidth to do decent
music from a general purpose computer. However, you might consider using
MIT's original solution to this problem: building a special box which reads
a mag tape at high speed and plays it into a D/A. How about a TK50 interfaced
to four 44Khz D/As? Likely be a lot less expensive.
John Sauter
| |||||