VLIW pre-history
|
From: Leif Harcke on 27 May 2007 00:03 On Mon, 21 May 2007 09:26:11 -0800, Eugene Miya wrote: > What the DSN replaced all those Modcomp IIs IVs and Classics with a > VAX?! Which dish? the 264 ft. one or one of the smaller ones? > Which model? I wonder if that used to be the 120B in 334. The VAX belonged to the Goldstone radar, and sat in the pedestal of the 70m antenna. Most of the solar system radar hardware is not part of the regular DSN tracking network. Obvious things like the dish surface and pointing motors are shared. The VAX belonged to 331, it was not part of the 334 setup (which I think was SEL 32's with attached AP-120B's for SIR-A and maybe SIR-B data processing). -Leif
From: James Dow Allen on 29 May 2007 03:51
On May 11, 5:13 pm, Stephen Fuld <S.F...(a)PleaseRemove.att.net> wrote: > Mark Smotherman wrote: > > Anyone know of other VLIW-like designs pre-1980? > > I'm not sure if you want to count these, but in the 1970s, there were a > series of parts by AMD called bit slice components (the 2900 series). > You could build your own "computer" ... I wrote the firmware for such a disk controller in 1980 (*almost* "pre-1980"). The custom machine instruction was converted into a Long Instruction Word via Decode and Secondary Decode Proms. I got 16 16-bit registers from the four slice chips and an extra 16 from external register chips. There was no other RAM, so I had a total of 64 bytes (not kilobytes, *bytes*) for programming needs: playing with timing chip, supporting seek overlap with multiple pending commands, etc. I actually kept 2 bytes in reserve in case of emergency so only used 62 byes of "Ram". Remember me when your engineers complain they need more *megabytes* of Ram for their controllers. There was another 512 bytes or so in a Fifo used to buffer disk data. I needed to use it as an auxiliary memory to run the "Chinese Remainder" algorithm used with some ECC codes. jamesdowallen at gmail |