Martin Gregorie  wrote:
> On Tue, 17 Apr 2018 16:17:13 +0100, Rodney Pont wrote:
>
>> On Tue, 17 Apr 2018 15:00:49 +0000 (UTC), Martin Gregorie wrote:
>>
>>> 1900 mainframes didn't have blinkenlights at all, though there was a 24
>>> lamp engineering display with a set of perspex masks that labelled the
>>> lamps - each mask matched a different socket in the CPU cabinet.
>>> Anything else would have been superfluous because the machine only had
>>> two hardware registers, which pointed to the datum and limit addresses
>>> for the running program. Everything else of significance (the 8
>>> accumulators,
>>> PC and CC, i.e. all 10 registers, were the first 10 words in every
>>> program and no, there was no SP register because it was a stackless
>>> architecture.
>>>
>>> Everything else I've programmed has, like the ICL 1900s, also showed a
>>> total lack of blinkenlights.
>>
>> The 1907 had a whole panel of lamps and I spent many a happy hour
>> loading registers to locate a a problem :-)
>
> I never saw a 1907 - just 1901,1902,1903S and 1904T - and, of course, the
> 2903.

The AN/FSQ7 was probably the most light-bedecked machine I’ve ever seen,
though the Stretch was also impressive.

Among the early machines, the Burroughs 220 has a very impressive front
panel. It was the last commercial vacuum tube computer in the US, and had
240 neon lights on its “main” panel and another 150 on its
“maintenance”
panels straddling the main panel. You may have seen it (with neons replaced
with incandescents) in several movies and as Batman’s computer in the Adam
West TV show. ;-)

--
-michael - NadaNet 3.1 and AppleCrate II:  http://michaeljmahon.com

--- SoupGate-Win32 v1.05