old microcode listings
|
Prev: Software vs Hardware
Next: Searching for the PDP-3
From: toby on 9 Jan 2007 18:35 Eric Smith wrote: > I wrote: > > of the MICROM chips, but there are also two PLAs in the control > > chip that can't be easily extracted and affect microinstruction > > sequencing, so the MICROM contents alone would be nearly worthless. > > "Peter \"Firefly\" Lund" wrote: > > PLAs should not be too hard to extract, as long as there's no storage > > involved. Oh, you mean they are embedded inside some other circuitry? > > Yes, there are two levels of PLAs (i.e., two AND arrays and to OR arrays) > which take inputs from several internal registers, and whose outputs > affect the operation of the microsequencer. > > > I think a cheap microscope and a camera is good enough. > > Tried that. Need a _good_ microscope and and a camera. > > > Perhaps just a camera with a "macro" setting > > Tried that too, with a 4 megapixel camera. Seems to be at least two > orders of magnitude off. If you can fill a medium format frame (not sure how close a macro lens will get) and use a suitable emulsion, you could achieve at least 400 megapixels... e.g. RB-67 bellows and 140mm macro? > > > or a camera + a magnifying glass. > > Haven't tried that.
From: kenney on 10 Jan 2007 05:54 In article <TQPoh.80021$493.60430(a)newsfe4-gui.ntli.net>, nospam(a)devnul.co.uk (ChrisQuayle) wrote: > Didn't the z80 also have fully vectored interrupts for > zilog peripheral devices, implemented as a vector > register in the peripheral controller chip itself ?... Yes, the peripheral controller had to put the vector on the data bus, the chips also included the hardware needed for priority determination. Ken Young
From: ChrisQuayle on 12 Jan 2007 17:34 Eric Smith wrote: > ChrisQuayle <nospam(a)devnul.co.uk> writes: > >>Yes it was, but probably the best that could be done at the time in >>commodity devices - ok, bit slice in intent ?. I think the more >>complex AMD 29xx series came quite a bit later than 1972. > > > Sure, but there were bit slice components before the Am2900 series. > The earliest I'm aware of were from Fairchild in 1968-1969. > > Year Vendor P/N description > ---- --------- ---- --------------------------- > 1968 Fairchild 3800 8-bit data path > 1968 Fairchild 3804 4-bit data path > 1972 National MM5750 4-bit data path > 1972 National MM5751 sequencer and microcode ROM > 1974 MMI 6701 4-bit data path, very similar to later Am2901 > 1974 MMI 6700 4-bit sequencer > 1975 Intel 3002 2-bit data path > 1975 Intel 3001 9-bit sequencer > 1975 AMD 2901 4-bit data path > 1975 AMD 2909 4-bit sequencer > 1975 AMD 2911 4-bit sequencer > 1976 TI 74S481 4-bit data path > 1976 TI 74S482 sequencer > 1977 AMD 2903 4-bit data path > 1977 AMD 2910 10-bit sequencer > 1977 Motorola 10800 4-bit data path, ECL > 1977 Motorola 10801 sequencer, ECL > > ? MMI 67110 sequencer > ? AMD 29203 4-bit data path > ? TI 74AS888 8-bit data path > ? TI 74AS890 sequencer > ? TI SBP0400 4-bit data path, I2L > ? AMD 29116 16-bit data path > > I'm missing information on the Fairchild Macrologic bitslice parts, > which were available in both TTL and CMOS. There are probably > some others I'm not aware of. > > I've seen conflicting reports over whether the Four-Phase AL1 (1970) > should be considered a bit slice design. I need to track down a copy of > the paper "Four-phase LSI logic offers new approach to computer > designer" by L. Boysel and J. Murphy from the April 1970 issue > of Computer Design. That's interesting. Nearly 3 decades ago, I was given a Honeywell VIP440 (?) terminal. Inside, were a load of boards, Intel 1103 dynamic ram etc and a cpu board containing around 6 Fairchild 40 pin packages with 1969 date codes. The terminal wasn't working, but after blagging the schematics from Honeywell uk London office, fixed the psu and away it flew. The interesting thing was that the cpu chips had names in the schematic like BLU, or basic logic unit. At the time, thought it was some sort of precursor to the Fairchild F8 micro, as that was also a multichip 40 pin package implementation. Looked, but ever did manage to find any info on the chips, but the above has probably solved the mystery. Do you have any other info or pointers to the 3800/3804 series ?... Chris
From: Del Cecchi on 12 Jan 2007 21:15 "ChrisQuayle" <nospam(a)devnul.co.uk> wrote in message news:F1Uph.83164$493.14454(a)newsfe4-gui.ntli.net... > Eric Smith wrote: >> ChrisQuayle <nospam(a)devnul.co.uk> writes: >> >>>Yes it was, but probably the best that could be done at the time in >>>commodity devices - ok, bit slice in intent ?. I think the more >>>complex AMD 29xx series came quite a bit later than 1972. >> >> >> Sure, but there were bit slice components before the Am2900 series. >> The earliest I'm aware of were from Fairchild in 1968-1969. >> >> Year Vendor P/N description >> ---- --------- ---- --------------------------- >> 1968 Fairchild 3800 8-bit data path >> 1968 Fairchild 3804 4-bit data path >> 1972 National MM5750 4-bit data path >> 1972 National MM5751 sequencer and microcode ROM >> 1974 MMI 6701 4-bit data path, very similar to later >> Am2901 >> 1974 MMI 6700 4-bit sequencer >> 1975 Intel 3002 2-bit data path >> 1975 Intel 3001 9-bit sequencer >> 1975 AMD 2901 4-bit data path >> 1975 AMD 2909 4-bit sequencer >> 1975 AMD 2911 4-bit sequencer >> 1976 TI 74S481 4-bit data path >> 1976 TI 74S482 sequencer >> 1977 AMD 2903 4-bit data path >> 1977 AMD 2910 10-bit sequencer >> 1977 Motorola 10800 4-bit data path, ECL >> 1977 Motorola 10801 sequencer, ECL >> >> ? MMI 67110 sequencer >> ? AMD 29203 4-bit data path >> ? TI 74AS888 8-bit data path >> ? TI 74AS890 sequencer >> ? TI SBP0400 4-bit data path, I2L >> ? AMD 29116 16-bit data path >> >> I'm missing information on the Fairchild Macrologic bitslice parts, >> which were available in both TTL and CMOS. There are probably >> some others I'm not aware of. >> >> I've seen conflicting reports over whether the Four-Phase AL1 (1970) >> should be considered a bit slice design. I need to track down a copy >> of >> the paper "Four-phase LSI logic offers new approach to computer >> designer" by L. Boysel and J. Murphy from the April 1970 issue >> of Computer Design. > > That's interesting. Nearly 3 decades ago, I was given a Honeywell > VIP440 (?) terminal. Inside, were a load of boards, Intel 1103 dynamic > ram etc and a cpu board containing around 6 Fairchild 40 pin packages > with 1969 date codes. The terminal wasn't working, but after blagging > the schematics from Honeywell uk London office, fixed the psu and away > it flew. The interesting thing was that the cpu chips had names in the > schematic like BLU, or basic logic unit. At the time, thought it was > some sort of precursor to the Fairchild F8 micro, as that was also a > multichip 40 pin package implementation. > > Looked, but ever did manage to find any info on the chips, but the > above has probably solved the mystery. Do you have any other info or > pointers to the 3800/3804 series ?... > > Chris Fairchild and Motorola both made a line of ECL chips, MECL1, 2, 3, 10k and 100k
From: Jonathan Thornburg -- remove -animal to reply on 14 Jan 2007 09:20
Peter "Firefly" Lund wrote: > Timing and race conditions are easy. Come on, race conditions, I mean, > really. Races are not hard, okay? ChrisQuayle <nospam(a)devnul.co.uk> wrote: > It may be easy for half a dozen ttl devices, but by the time you have > hundreds of devices, you will need to be a very competent designer to > make it work reliably over component spreads and temperature. Building a > vax in ssi ttl would be herculean task, but even the 11/05 was a two > board hex width unibus set with hundreds of ssi ttl devices. Chris is ++right. Back in the "old days" even professional designers with lots of experience sometimes missed race conditions. There's a classic example in Tracy Kidder's "The Soul of a New Machine", http://en.wikipedia.org/wiki/The_Soul_of_a_New_Machine in the chapter "The Case of the Missing Nand Gate". A design team at Data General, working on what became the Data General MV/8000 (= DG's answer to the VAX), missed a race condition... and spent two months tracking down the resulting intermittent failure. I suspect that even with today's simulation capability, similar bugs still lurk in modern microprocessors. In fact, I'm sure of it -- just look at the errata sheet for your favorite microprocessor. ciao, -- -- "Jonathan Thornburg -- remove -animal to reply" <jthorn(a)aei.mpg-zebra.de> Max-Planck-Institut fuer Gravitationsphysik (Albert-Einstein-Institut), Golm, Germany, "Old Europe" http://www.aei.mpg.de/~jthorn/home.html "Washing one's hands of the conflict between the powerful and the powerless means to side with the powerful, not to be neutral." -- quote by Freire / poster by Oxfam |