The Unix Heritage Society mailing list
 help / color / mirror / Atom feed
From: Lawrence Stewart <>
To: ron minnich <>
Cc: The Eunuchs Hysterical Society <>
Subject: [TUHS] Re: Burroughs funded $30,000 for a port to E-mode stack machine in 1983!
Date: Sun, 15 Jan 2023 20:59:56 -0500	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

[-- Attachment #1: Type: text/plain, Size: 4979 bytes --]

I rise to defend the honor of ECL.  I had designed with TTL through college and the ‘80s: network cards, graphics controllers, the Etherphone, the CPU modules for the Firefly MicroVAX multiprocessor, and so forth.  My early designs were SSI and later ones PAL based mostly (Like in Soul of the New Machine).  In 1990 or 1991 Digital Systems Research Center started work on the Alpha Demonstration Unit, intended for software porting.  The Alpha EV-3 and EV-4 were I/O compatible with either TTL or ECL levels, so we chose ECL 100K with some of the new ECLinPS stuff and even some custom Gallium Arsenide address drivers for the DRAM arrays.  This all let the system run at 100 MHz, to keep up with the 100 or 200 MHz Alphas.

The short version is that if you can afford the power, ECL is <awesome>.  Every wire is a transmission line. Every signal is terminated. Almost every part (except the 5 NS cache rams) has complementary outputs.  What’s not to like about 350 ps NOR gates? The signal integrity alone makes it worthwhile. The CACM paper is at <> but the Digital Technical Journal version is on Bitsavers at <>

Naturally all the production systems were TTL and CMOS, but that ECL stuff was great fun.  We had a local welding shop make 18” cables for the 400 Amps of -4.5v and 200 Amps of -2v terminator power.  They said “what kind of a welder is this for anyway?

The Unix angle, such as it is, is that the ADU ran OSF-1 and Alpha VMS <at the same time> on different subsets of the CPU and memory boards.  Not a hypervisor, but a hard partitioning of resources.  There was enough I/O stuff to run three “computers” I think.

Earlier I had been a user of the Xerox Dorado, which was 16 MHz ECL (probably MECL) in the early 80’s.  It made a wonderful personal computer :).  Ken Pier’s retrospective at <> talks a bit about debugging on page 32.  The Dorado had JTAG like scan logic threaded through it, which made debugging and repair of such a complex beast possible.  IIRC quite a lot of work went into the control software for the scan logic, devoting an Alto to having a decent UI for it.


> On 2023, Jan 15, at 7:51 PM, ron minnich <> wrote:
> <>
> has the burroughs info.
> On Sun, Jan 15, 2023 at 11:09 AM ron minnich < <>> wrote:
> I just stumbled across an old letter, from a VP of Burroughs to me and Steve Bartels, authorizing $30,000 for a port of Unix to the E-mode stack machine. I had forgotten getting it.
> Burroughs was famed for its stack machines. E-mode was a kind of last gasp attempt to save the stack architecture, which failed as far as I know, see this table:
> <>
> I worked as a hardware engineer on the A15. I also had been a Unix user for 7 years at that point and kept pointing out how awful the Burroughs CANDE time-sharing system was, and how much better Unix was. At some point I guess they asked me to put up or shut up. I got that money, and left Burroughs a week later for grad school. 
> Funny note: A15 was Motorola ECL (MECL), and ran at 16 Mhz., considered fast at that time.  We used a technique called "stored logic" which was, believe it or not, using MECL RAM to map logic inputs to outputs, i.e. implement combinational logic with SRAM. Kind of nuts, but it worked at the time. We also used a precursor of JTAG to scan it in. Those of you who know JTAG have some idea of how fun this had to be.
> One side effect of working with MECL is you realized just how well designed the TI 7400 SSI/MSI parts were ... MECL always just felt like an awkward family to design with. 
> Another funny story, pointing to what was about to happen to Burroughs. We had an app that ran for hours on the stack machine. We quick ported it to a VAX, started it up, and headed out to lunch -- "this will take a while, let's go eat." We got to the front door and: "Oh, wait, let me hop back into the office,I  forgot my jacket". And, noticed, the program was done in ... about 3 minutes. Not 8 hours. 
> That's when we knew it was game over for Burroughs. 
> If a picture of this letter would be useful in some archive somewhere, let me know, I can send it.

[-- Attachment #2: Type: text/html, Size: 6659 bytes --]

      reply	other threads:[~2023-01-16  2:00 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-15 19:09 [TUHS] " ron minnich
2023-01-15 19:19 ` [TUHS] " Luther Johnson
2023-01-16  0:51 ` ron minnich
2023-01-16  1:59   ` Lawrence Stewart [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \ \ \ \ \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).