Newsgroups: comp.sys.transputer
From: paul@walker.demon.co.uk (Paul Walker)
Subject: Re: Information on T9000 wanted
Organization: Paul Walker Consultancy
Date: Wed, 25 May 1994 07:42:25 +0000
Message-ID: <769851745snz@walker.demon.co.uk>


If you are after specifications, there are a couple of large data books,
"The T9000 Transputer Reference Manual" and 
"The T9000 Transputer Instruction Set Manual"
both available from SGS-THOMSON/INMOS. Also the "Networks, Routers 
and Transputers" book, from IOS or from the HENSA archive.

INMOS has been embarassed by the delay to the T9000 reaching the market,
and so is being very cautious in its public statements about 
availability and actual performance. At the April WoTUG conference
in Bristol, those users who had managed to get their hands on
actual T9000 devices were full of praise for their performance 
and ease of use, in spite of a number of minor problems with the 
versions of the silicon they used.

Speed is currently around 20MHz, but that is a equivalent to
something like a 100MHz T800, and the links and virtual channels
seem to be good. 

What INMOS did say at WoTUG was that they had a continuous improvement
programme in place for removing the problems at a solid 25MHZ, and 
increasing the speed to 40MHz. They gave tentative dates but I'd prefer
you get these from INMOS rather than from me.

But until the devices are available in volume greater than the 
pent-up demand, and with the minor problems fixed, expect INMOS and
the SGS-THOMSON sales force to concentrate on selling the products
that they can already make in volume. (and for the existing transputer
family there appears to be an increasing demand, particularly for
embedded systems.)

In article <2rc2i5$bfp@nnrp.ucs.ubc.ca>
           abeatty@unixg.ubc.ca "Andrew Beatty" writes:

>Hi, 
>
>        I havn't read messages posted in this group for 
>quite a while. I understand that T9000 is on the market
>now and wonder if you could forward me some information
>regarding its technical specifications. A summary will
>be greatly appreciated, though any pointer will help.
>
>        Thank you in advance.
>
>Jonathan
>
>

