On Fri, Jan 13, 2017 at 4:42 PM, Antonio Carlini <a.carlini at ntlworld.com> wrote: Digital Technical Journal, Number 2 March 1986 http://www.dtjcd.vmsresource.org.uk/pdfs/dtj_v01-02_mar1986.pdf The MicroVAX 78032 Chip, A 32-Bit Microprocessor All existing VAX processors implement memory management with a large address translation cache (at least 128 entries), with system and process addresses in separate halves. A translation cache must have a high hit rate to be effective. Since most caches are direct mapped, many entries are required to achieve a high cache rate. Implementing a comparable number of translation cache entries in the 78032 was out of the question, due to die size constraints. However, the VLSI technology in the 78032 is very amenable to using a fully associative translation cache with least-recently-used (LRU) replacement. Such a cache needs many fewer entries to achieve the same hit rate as the direct-mapped version. In addition, the tight coupling to local memory, as explained in the next paragraph, made it possible to reduce drastically the amount of time required to process a translation cache miss. Thus the translation cache in the chip has only eight entries, but the cache is fully associative, uses true LRU replacement, and is supported by highly optimized microcode for fast processing of misses. Moreover, simulation studies showed that the best use of the eight entries was with a homogeneous structure. Therefore, the system and process addresses are cached together. The team also decided to forgo the use of an external memory cache, which required a complex external interface. Use of an internal memory cache had already been ruled out due to die size constraints. Accordingly, the speed of memory access is 400 ns, or two microcycles, which is the speed of local memory. Thus the chip encounters no wait states, and its average time to access memory is approximately the same as the 11/780's. In a typical program, there is little difference between the integer instruction performance of the two CPUs.

On 1/15/17 7:20 PM, Alan Perry wrote: Good luck with that. I tried with my rarest, and they claimed they already had one. Someone mentioned Vectrex, which isn't all that rare. However, my one-off homebrew S-100 Vectrex interface is rare enough. Basically it's just dual ported ordinary memory with a bit on the S-100 side that says whose turn it is to use the memory. Another rare item is my Texas Instruments SR-22 calculator. It does full floating point in decimal, hexadecimal, and octal. Scientific notation is in the same base, with an octal point and base-8 exponent, for example. Burroughts E1400 deskside unit for the accounting machine. It's 1966 vintage, and has a couple core planes. I have the printset, too. It makes a dandy table, and hasn't seen any moving electrons (or holes) for 30 years. The local museum refused it, though. -- Jeff Woolsey {{woolsey,jlw}@jlw,first.last@{gmail,jlw}}.com Nature abhors straight antennas, clean lenses, and empty storage. "Delete! Delete! OK!" -Dr. Bronner on disk space management Card-sorting, Joel. -Crow on solitaire

On 1/15/17 8:58 PM, Jeff Woolsey wrote: I forgot that I had photos of this thing online somewhere: http://www.jlw.com/retro/slafmac/ -- Jeff Woolsey {{woolsey,jlw}@jlw,first.last@{gmail,jlw}}.com Nature abhors straight antennas, clean lenses, and empty storage. "Delete! Delete! OK!" -Dr. Bronner on disk space management Card-sorting, Joel. -Crow on solitaire

On 1/18/17 5:57 AM, Mike Stein wrote: Quite. I did have the other half for a while, and it worked as well as we (philistines--college students having fun) thought it could. We'd had enough fun with it and decided there wasn't anything more constructive to do with it, so we did destructive things to it. We took it to pieces to see all the neat stuff in it. I still have a couple parts, one being a cylindrical structure with lots of shafts and gears, and another being a short chain on a sprocket wheel/axle that is something to keep your hands busy when bored. That's where I found out what model it was, though the printset differs. Mine's just holding stuff off the ground for now. -- Jeff Woolsey {{woolsey,jlw}@jlw,first.last@{gmail,jlw}}.com Nature abhors straight antennas, clean lenses, and empty storage. "Delete! Delete! OK!" -Dr. Bronner on disk space management Card-sorting, Joel. -Crow on solitaire

On 01/13/2017 06:42 PM, Antonio Carlini wrote: Right, I am misremembering. The 780 and the KA650 had caches. The KA630 did real well, then, without a cache. So, I never understood why the Nat Semi 32016 performed so poorly. I did build a multiprocessor using the 32016 WITHOUT the MMU, and it looked to be around 300KIPS (roughly 1/3 of a VAX 780). So, maybe it was the MMU that really hurt the performance. Jon

On 01/16/2017 03:02 PM, Pete Lancashire wrote: Hmm, what's a 360/55? I know the 360/50 and the 360/65 quite well. Seems there is some mention of a /55, but what is different about it from a model 50? Thanks, Jon

On 01/17/2017 06:16 PM, Pete Lancashire wrote: Ahh, the 360/50, I knew it well. Maybe, too well! But, I always thought the /50 and /65 were just about the classiest front panels around! Jon