Newsgroups: comp.sys.transputer
From: ifunltd@xs1.xs4all.nl (ifunltd)
Subject: Re: Just An Observation
Organization: XS4ALL, networking for the masses
Date: 26 Dec 1994 09:39:16 GMT
Message-ID: <3dm304$bi9@news.xs4all.nl>

anuzen@tidal.nosc.mil wrote:

: A few years ago (1986 to be exact), I worked on a supercomputer project 
: using the T400s, and today just out of curiosity, I browse through the 
: articles on this forum. I can't help but notice that most of the 
: participants are from the UK or Germany. At the time, I thought 
: Transputers were the way of the future, and still do now. I wonder why 
: the Transputers never caught on in the US?

Well, it's the same reason the Transputer did not really take off in Europe
either: people just did not trust a European company - let alone a small
startup producer of miroprocessors - to make something good. After a few
years of looking at T's from a safe distance, the Americans decided the
i860 was a lot faster - and showed by this observation that they had not
understood the concept of distributed-memory parallel machines. 

Today, allmost everyone has understood that concept (look at all the big
producers and the architectures of their machines) and Inmos has practically
died. Tragic, but true: Americans are too much "patriots" to buy European,
and Europeans are too uncertain about their own potential to buy European.
If I would have a great idea for processors or machines, I would set up
an American company to sell it, for psychological reasons.

: Anyway, would any of you out there, be kind enough to update me to what 
: the latest and greatest developments in the world of Transputers.

SGS-Thomson has tranferred production of T's to the continent (France),
and is wiping out the name Inmos, as far as I know. The T9000 still seems
to be worked on, but high speed it will never become (at the planned 50 MHz
it would not outperform any of the normally available single processors,
although the communications engine would be quite significant, since most
processors still do not have one). There is talk of a successor, which is
no real successor since the T9000 did not make it. This "successor" would
go into the direction of macro cells, is all I know about it. Again, this
was a European Microprocessor Initiative, but they changed the name from
EMI (isn't that a record company ?) to OMI, with the O indicating "Open".

Hope this gives you some answers

Jang (IF UnLtd)

