3 Dec
2014
3 Dec
'14
10:31 a.m.
Rich Alderson <RichA at LivingComputerMuseum.org> writes:
From: Eric Smith
Ah, that explains it. Thanks!
> It was? During my stint at XKL, I was told
the Toad-1 I was
> working with was an FPGA implementation. Maybe it was an
> experiment, but it seemed to work pretty well. But then, I was a
> telecommuter, so maybe some information was jumbled in
> long-distance transmission.
You guys probably have more information about
this than I do, but
my understanding was that the Toad-1 CPU required multiple FPGAs,
CPLDs, and other logic, while the Toad-2 CPU is a single FPGA.
Lars, Eric is correct. The Toad-1 is a multi-board computer system
(at minimum, the XKL-1 CPU, one XMG-1 memory, one XNI-1 Ethernet
interface, and one XRH-1 SCSI interface, in a 7-slot backplane).
Each board used several programmable parts, from Altera and Xilinx.
In particular, the CPU was a pair of Xilinx parts for the microcode
engine, with an AMD 2900 family sequencer (I forget the particular
part number).
The Toad-2 (and the 1U rack mount unit is *not* a Toad-2, but it
does contain two of them, thus the label) is a single FPGA
implementation of the Toad-1 (mutatis mutandis: no SCSI, but it
talks to SD cards, for example).