using an FPGA to emulate a vintage computer
in [FPGA]
|
From: rickman on 6 Feb 2010 13:19 On Feb 5, 1:51 pm, "(see below)" <yaldni...(a)blueyonder.co.uk> wrote: > On 05/02/2010 18:19, in article > badc12c3-cb2b-4ce9-9543-237d60fc2...(a)o8g2000vbm.googlegroups.com, "Eric > > Chomko" <pne.cho...(a)comcast.net> wrote: > > Has anyone created a copy machine of an old system using an FPGA? I > > was wondering if it would be possible to take an entire SWTPC 6800 and > > compile the schematics and have it run on an FPGA board.? Wouldn't > > even have to be the latest Xylinx product, I suspect. > > I think such a project would valuable, and perhaps even more valuable if it > aimed to recreate a machine of the "heroic" era -- a 7094, an Atlas, or a > KDF9, say. Perhaps even a Stretch. > > KDF9 had about 20K transistors, a few K logic transformers, and a comparable > number of diodes; less than 50K devices in total. I imagine this would be > easily accommodated on a modern FPGA. The big question would be whether to > go for functional equivalence, or whether to try to replicate the original > internal structures. > > Documentation would be the main challenge for the latter. > > -- > Bill Findlay > <surname><forename> chez blueyonder.co.uk Heck, on an iCore 2 you might be able to run that under Spice! You could probably even provide a graphical display of any front panel lights! Rick
From: Jecel on 6 Feb 2010 13:34 I try to keep a reasonably updated list of such projects at http://www.merlintec.com:8080/hardware/31 -- Jecel
From: glen herrmannsfeldt on 6 Feb 2010 14:10 In comp.arch.fpga Anne & Lynn Wheeler <lynn(a)garlic.com> wrote: (snip) > in the early 80s los gatos did custom hardware for chip logic simulation > (LSM ... "losgatos state machine" ... then "logic simulation machine" > for publication) ... dozen plus rack boxes ... ran 50,000 times faster > faster than logic simulation in software on 3033 I remember when I first started working with computers I had a book from our library about ECAP, IBM's Electronic Circuit Analysis Program. I never saw or used the actual program, and haven't heard about it since. I wonder where it went... -- glen
From: Anne & Lynn Wheeler on 7 Feb 2010 11:51 glen herrmannsfeldt <gah(a)ugcs.caltech.edu> writes: > I remember when I first started working with computers I had a > book from our library about ECAP, IBM's Electronic Circuit > Analysis Program. I never saw or used the actual program, > and haven't heard about it since. I wonder where it went... re: http://www.garlic.com/~lynn/2010c.html#71 using an FPGA to emulate a vintage computer no direct knowledge and web search is rather sparse ... a couple IEEE citations: http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F23%2F4335780%2F04335910.pdf%3Farnumber%3D4335910&authDecision=-203 http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F6%2F5218140%2F05218152.pdf%3Farnumber%3D5218152&authDecision=-203 in the aftermath of the troubles of the early 90s ... there was push to move to industry standard tools ... part of which involved transfer of internal tools to chip tool vendors (and some number of the internal chip tools people spending a lot of time with these vendors ... and then some number leaving and joining external vendor). I've mentioned recently porting nearly 60k statement pascal program (that did circuit layout) to other platforms, as part of such a tool transfer. http://www.garlic.com/~lynn/2010b.html#74 Happy DEC-10 Day http://www.garlic.com/~lynn/2010c.html#29 search engine history, was Happy DEC-10 Day in the mid-80s ... internally, there was big push to expand a lot of mainframe manufacturing capacity anticipating the market would double in size by the early 90s. Not particularly "career enhancing" ... I made some observation that computer hardware was becoming increasingly commoditized ... resulting in thinner margins & profits ... which would at least require significantly cutting the number of related employees to stay out of the red. misc. past posts mentioning various (non) "career enhancing": http://www.garlic.com/~lynn/2007e.html#48 time spent/day on a computer http://www.garlic.com/~lynn/2007f.html#30 The Perfect Computer - 36 bits? http://www.garlic.com/~lynn/2007r.html#6 The history of Structure capabilities http://www.garlic.com/~lynn/2008l.html#23 Memories of ACC, IBM Channels and Mainframe Internet Devices http://www.garlic.com/~lynn/2009c.html#54 THE runs in DOS box? http://www.garlic.com/~lynn/2009p.html#34 big iron mainframe vs. x86 servers http://www.garlic.com/~lynn/2009r.html#49 "Portable" data centers http://www.garlic.com/~lynn/2009r.html#50 "Portable" data centers http://www.garlic.com/~lynn/2009s.html#4 While watching Biography about Bill Gates on CNBC last Night -- 42yrs virtualization experience (since Jan68), online at home since Mar1970
From: James Dow Allen on 7 Feb 2010 16:44
On Feb 7, 12:10 am, Al Kossow <a...(a)bitsavers.org> wrote: > On 2/6/10 3:29 AM, James Dow Allen wrote: > > a National Semiconducter subsidiary once tried to > > emulate an IBM 3033 at full speed using Fairchild 100k parts. > > ... the reason for failure is interesting ... > . > I would be interested in what the reason for failure was. > I assume it wasn't the obvious chip-chip delays using commodity > ICs. I think you mean that chip-chip delays would be too obvious to be interesting. :-) The commodity high-speed ECL chips were almost in the same ballpark as IBM's chips for speed and density, but IBM's packaging was better in various ways. The one difference I found "interesting" and which seemed to be a significant factor in the slowdown was that IBM used smaller circuit boards. Each signal was therefore closer to the backplane, so closer to more chips total; in other words the smaller circuit boards allowed IBM's wiring to take better advantage, in some sense, of the 3rd dimension! I think there were other important factors in that project's failure, but there's no need to start any anti-NatSemi flamefest. :-) James Dow Allen |