Newsgroups: comp.sys.transputer
From: paul@walker.demon.co.uk (Paul Walker)
Subject: Re: T801 availability
Organization: Paul Walker Consultancy
Date: Wed, 23 Nov 1994 07:37:21 +0000
Message-ID: <785576241snz@walker.demon.co.uk>

Many thanks to David Boreham who replied to the newsgroup, and to others
who replied privately. I have been offered enough devices now to satisfy
the need. 

Athough enough devices have surfaced this time, I should probably
emphasise INMOS' recommendation that the T801 is not used in any 
new designs.

A comment on the "B016 test" below

Many thanks again

Paul Walker

In article <1994Nov19.115352.13580@eso.org>
           rreiss@hq.eso.org "Roland Reiss" writes:

>In article pni@info.eng.octel.com,  david.boreham@octel.com writes:
>>
>>Remember that there were many 801 lots with tremendous bad karma (25MHz 
>>binnings) so whatch out what dross you're using. Udeally test using Neil
> Campbell's B016 tester and at temperature before using.
>
>What is this B016 tester, anyway?
>

It is often rather difficult to design a device test program which is
as rigorous as a few seconds of system test on a PCB with RAM and
peripherals. On a number of occasions, board testing showed up faults
in transputer silicon, and the results were fed back to improve the 
silicon test. In the case of the T801, used on very few boards, the B016
VME master board used the T801 very thoroughly, and was able to detect
faults which had not previously been found. It is possible that 
the T801 was taken out of production before all the results had been
incorporated in the silicon test.

So there may be some devices around which would fail the B016 test.
Fortunately most of them will be 25MHz parts, and hopefully the bugs
would not appear in all systems.

Paul

