From: "David Brownlee" <abs at absd.org> The keyboard was all kinds of awful, the mouse had a metal ball that needed the right mat to get any traction and its buttons kept popping off. It made the contemporary sun 3/50 seem like an amazing speed demon, and in a university lab of 20 I'm not sure i ever saw all machines working. People would pick an adm5 in preference to a Whitechapel - for an adm3a the decision was less clear cut. As you can tell, I miss them and wish I had been able to keep hold of one :) <<<< At the University to which abs is referring above, I wrote the operational requirement that led to the the purchase of those machines. I thought I still had a copy but can't find it at the moment. Three proposals made the shortlist: Apollo, Sun, and Whitechapel but, essentially, the MG-1s were the only proposal that offered enough machines to be useful for a teaching room. (for the same money ISTR that Apollo offered 3 machines!, can't remember how many Sun offered but, looking at the benchmark results, they couldn't have had FPUs) At the next major procurement Sun had clearly learned a lesson. We got huge quantities of 386is at a brilliant price (they only guaranteed in the contract to supply the minimal model but a large proportion of them were better than that including some with large colour screens) I acquired one of the (by then) redundant MG-1s but passed it on to someone else when my PC was seriously outperforming it. Talking of benchmark results MG-1 tested at 145 Kwips (slightly worse - 136.5 - non-optimised) Sun (model unknown) at 64.5 (63.5 no) - but for integer results it was /much/ faster than the MG-1 The Gould 9000 supermini we bought at the same time, by comparison rated at 2000* * my notes are confusing - it might have been 2597 (single precision) from 2000 iterations. The ICL 1905E mainframe that we had 6 years earlier rated at 64.9 with the optimising compiler and about 10% better (!!!!) with the non-optimising - CCTA perhaps did too well in making the benchmark resistant to optimisation.

From: tony duell Sent: Sunday, November 30, 2014 8:03 AM It was renamed after the 32032 hit the market, same architecture with 32-bit external data paths instead of the 16 bits of the earlier chip. Somewhere in the storage locker I still have original data sheets for both. I think. Rich Rich Alderson Vintage Computing Sr. Systems Engineer Living Computer Museum 2245 1st Avenue S Seattle, WA 98134 mailto:RichA at LivingComputerMuseum.org http://www.LivingComputerMuseum.org/

On 11/30/2014 02:36 PM, Jon Elson wrote: The 16032 with graphics instructions addition (Bitblt) as the 32CG16 found some use in laser printers, also--Canon and Panasonic used it on some of their models. I called it the "whenever" chip. Sometime around the time that the 68K burst onto the scene, a couple of guys from NSC to give us a "preview" of the 16-bit version. Promising silicon the upcoming December (ISTR that this was about March) and pre-production samples "Real Soon Now", we had the next generation of systems designed and into production before that came about a couple of years later. Make no mistake, this was a remarkable design for the time, architecturally speaking. According to WikiP, the original intention was to clone the VAX, but the legal challenge to DEC was abandoned. I think that some avoided NSC when it came to MPUs--support wasn't what it should have been, nor the marketing attention span. NSC was fine as a second-source for many things, but there was a certain reluctance, I think, in the industry to use them as the prime source. I think I've still got an 32CG16 rattling around in my "MPU morgue"--PLCC package, if memory serves. --Chuck

On 11/30/2014 04:54 PM, Jules Richardson wrote: I never got to that point--the NSC chip was so late that it wasn't even a candidate. I seem to recall some discussion that the early 16032 chips were a little on the flaky side, but that may have been rumor. --Chuck

On 11/30/2014 07:21 PM, Chuck Guzis wrote: I have no idea what the chip set cost new from NSC, but I don't think it was terribly expensive. Maybe a couple hundred $ for the complete set. I think the Logical Micro system was about $2500 for a complete system in a large IBM PC sized box, with hard drive, floppy, memory, Multibus backplane and power supply, dumb serial board and Genix OS. Jon

Matrox QG-640... a third party video card for the Q-bus?! Neat. Will it drive a VGA display? I looked at the PDF of the manual and the specs sound similar. What does the bulkhead look like? Did it use a standard DEC keyboard and puck mouse with breakout cable like the VCB02? Do you have a complete system? It would be cool to see a picture if you could get them to drive a common VGA display. Curious to know more. Were there other third party vendors in the Q-bus video market? Best, Sean On Mon, Dec 1, 2014 at 12:12 AM, Glen Slick <glen.slick at gmail.com> wrote:

On 11/30/2014 05:21 PM, Chuck Guzis wrote: It was a fact, but even if/when they (mostly) fixed it the "rep" continued to dog them. There are certainly indications they had those kinds of issues with products besides the 32k family. And I think they always struggled to achieve the higher-clocked yields and performance they needed across designs. This is all based on odd bits of reading over the years as I wasn't interested in (and had barely heard of) the ns32k until Bruce Culbertson shared his homebrew 32016 design around 1990 - long after the competition had left Nat Semi behind. On 11/30/2014 04:54 PM, Jules Richardson wrote: I'm not clear on the pricing, but the ns32k was designed and delivered as a complete chipset to support multi-user, virtual memory systems with a 32bit-wide CPU design. Even though the MC68000 was slower at 32bit operations it was available several years earlier, and (I gather) was at least equally targeted at "standalone" use like a traditional MPU, or using 6800-family peripherals, so that as the price curve fell it could move into mid- and low-end applications and develop an enormous volume advantage... And then you've got a solid track record, vast adoption, and the forward-compatibility of the instruction set with the fully-32bit 68020 when it shows up circa 1984. Even though the 32032 was released that year, and the 32016 had been out for two years, NSC still had some nasty errata and only a 24-bit physical address space, and were both much slower than the '020. You can certainly find examples of the 32016 being used stand-alone, but not with anything like the ubiquity of the 68000... In terms of performance, you can still find a lot of lively "discussion" in a NetNews archive that includes net.micro.16k, net.micro.32k, and comp.sys.nsc.32k. The 32032 only outpaced the 32016 when the 32bit wide data bus came into play despite coming two years later, and still topped out at 10MHz. The 68020 seems to have beaten them both handily clock for clock, and of course that eventually went up to what, 30MHz? The 32332 arrived in 1985 but only reached 15MHz and was never more than 50% faster best case than the '032. The 32532 of 1987 was a complete redesign and finally seemed to trump the 68030 and 80386 - only to be pummeled by the emerging RISC designs... Despite the issues with early silicon, there were some notable adopters. Encore and Sequent delivered multiprocessor systems based on the ns32k family. Siemens-Nixdorf built a few models running SINIX. Definicon and OPUS shipped add-in boards for PCs, using the x86 for I/O. Tektronix delivered a couple models of UTek workstations. More notable than "successful" since this wasn't commercial, but Niklaus Wirth et al at ETH Z?rich built their Ceres workstations using the ns32k family. Bill Jolitz, later of 386BSD fame, built and sold the very portable Symmetric S/375 using a 32016 chipset (and running 4.2BSD). And you've got others like Whitechapel, Acorn, BBC Micro - CompuPro even shipped an S-100 32016 board. But for my money the most interesting two designs were Bruce Culbertson's homebrew 32016 design, and the pc532 by Dave Rand and George Scolaro (both of whom had been part of the Definicon team). Culbertson ported MINIX to his design, and later to the pc532. Both designs were freely available, but Dave and George organized a run of around 200 of the complex multi-layer pc532 system board in the early 90s and a group-buy of the more expensive parts like the processor complex. The pc532 would eventually run MINIX, Mach, and NetBSD/pc532. BTW - If you've got a spare pc532 sitting in a closet, I'd like to get one... ;) On 11/30/2014 04:54 PM, Jules Richardson wrote: The NS32081 enjoyed some popularity as an FPA with non-ns32k systems. In fact the only application note in National's 1984 Series 32000 Databook interfaces the 32081 with a 68000... On 11/30/2014 05:46 PM, Al Kossow wrote: Nat Semi also offered some flavor of System V Release 2 by 1986-87. And I think SysV was offered by OPUS and some other OEMs. Never have found out what software - if any - CompuPro offered for their 32016 board. --Steve.

Steve Jones wrote: Yes, the first machine, whose designation I cannot recall at the moment, was 32032 based, and was a desk-side unit, quite large. The UTek OS was based on 4.2bsd, though there were definitely some Tek-specific modifications. A quite complex graphics processor was available for the machine. It was quite capable for its time. However, Tektronix didn't really know how to market and sell computers (after all, they were a Test & Measurement company), and sales were less than stellar. Despite this rather lousy sales of the big machine, a smaller line of 32016-based machines were made. The first was the 6130. It was about the size of an original IBM PC, and had 3 dual-wide or 6 single-wide expansion slots, which could be mixed and matched with peripheral and memory cards. The machine came with a 30MB (IIRC) Micropolis 5 1/4" hard disk (ST-506), and a Wangtek 1/4" cartridge tape drive (DC-300) and a 5 1/4" double density half-height floppy disc drive. I think memory maxed out at 4MB. The machines made a great little desktop Unix machine hooked up to a CRT terminal. I bought all the parts to build one from Tektronix engineering stock back in the day (Tek employees got cost+10% for any parts Tek stocked), and put a complete system together. Great fun, and it worked straight away. The CPU board (single board) was purchased completely assembled, so that made things a lot easier. The next was the 4132. It was very similar to the 6130, but it changed connector on the back panel of the CPU for adding external drives. It also had a built-in IEEE-488 interface, and was really designed to be an instrument controller, which Tek was pretty good at, and could sell the machine as such. I ended up buying one of these at the Tektronix Country Store for something like $30 when the line was canceled and a bunch showed up at a "Fire Sale". It was a bit faster than the 6130, partly because of faster disk I/O, and also I think that the 32016 CPU was a faster-clocked version. At this point, denying the fact that Tek didn't really know how to sell computers very well (despite the successes with the 4051/4052/4054), they dumped the NS processors, and went with the Motorola processors for a line of machines, that were desktop form-factor (a little larger, though), and had built-in graphics. Pretty nice, and were much faster than the NS-based machines. However, again, not much market success. Lastly, they dumped the NS processors, and went with the Motoroal 88K CPU for a line of XD88 "superworkstations". They were fast, and had pretty amazing graphics capabilities, but alas, the company never really figured out how to sell computers as computers/workstations, and eventually the development group was disbanded, and that ended Tek's foray into Unix workstations. While on the subject, a plea -- if anyone out there has floppy images for the standalone boot and an image of the DC300 UTek installation tapes for the 4132, I'd love to get my hands on them. I still have the 4132, but no media -- the original floppies were low-quality, and are unreadable today, and the installation tapes that I have don't seem to want to read either, likely due to print-through from sitting for so very long (since like, oh, 1987 or so). I'd love to get this beast running again. Stupidly, I never imaged them back in the day. -Rick Bensene

On 2014-12-01 16:32, Eric Smith wrote: 1984?

On 11/30/14 4:54 PM, Jules Richardson wrote: Probably the easiest ways to find one is find an old HP9000/200 floating point board. There is one on there or take your chances on a part from Hong Kong http://www.ebay.com/itm/NEC-NS32081D-10-DIP-Arithmetic-Processor-/271552360… Note the pic is an 8MHz part, not 10. National released 4.1BSD for it. Opus Systems had it running on their PC coprocessor card.

The 32081? I think the 16081 was the same chip, if it ever existed under that number. I can't remember if my ACW has it fitted or not. It has a socket for it, certainly. I think all Whitechapel MG1s have a 32081 fitted. I have a spare MG1 motherboard somewhere, perhaps I should pull the FPU chip and try it in an Acorn. -tony

The processor part numbers changed from NSxxnyy to NSyynxx, and IIRC some of the peripherals changed from NS16nnn to NS32nnn. This caused great confusion if you had a NS32032, because there was no easy way to tell whether that specific part was in the old-style or new-style numbering.

On 11/30/2014 12:51 AM, tony duell wrote: I think I've seen four of the machines in total; every single one of them had suffered significant corrosion to the main PCB beneath because of the batteries. I know the CG-200 at NMoC has such an adapter. I've actually got a photo of it here - it has 3x 8-bit ISA slots toward the rear of the system, as you say, but then there are PCB traces at the other end of the adapter board for three 64-pin sockets, although only one is physically fitted on NMoC's machine. Purpose unknown, and it'd be interesting to know if anything ever made use of it. That particular system has an ISA serial board and some flavour of ISA video board (whatever the latter is, it has 8x 41264 dual-port RAM chips, and there's a 20-way ribbon cable attached which runs off somewhere beneath the ISA adapter board). Unfortunately the machine - like the MG-1's - had also suffered significant corrosion. cheers Jules

On 11/30/2014 11:53 PM, tony duell wrote: Yes, I seem to recall that those were bad - but the corrosion had spread quite far around the connectors. :-( Unfortunately I don't seem to have any MG-1 photos here to compare it against, and it's been 7 years since I've seen one. In the CG-200 the memory boards sit along the left-hand edge of the system, and I think that is true of the MG-1, but four of the primary nsxxxxx ICs live on a little plug-in daughterboard - I have a feeling that wasn't the case with the MG-1 (its primary ICs lived in sockets directly on the motherboard), and hence that it's a different animal. The amount of RAM on the video board is 512KB, I think - which is an odd number; not enough for an 8bpp version of the MG-1's 1024x800 display, and yet rather high for a mono display (I think CRTs in the multi-megapixel range likely existed then for specialist applications where money was no object, but if someone had the cash to burn they'd probably be buying something other than a Whitechapel). Of course maybe it outputs 1024x800 at 4bpp, but I thought that by the mid-80s serious colour-capable hardware was generally outputting 8bpp minimum (and non-palettised outputs were appearing on the scene). While I can see a 6845 IC on the board in my photo, I can't make out what the other large (28 pin) IC is on the board near to the video connector - I suspect it's a DAC of some kind, but knowing exactly what might be useful. cheers Jules