Date   : Fri, 17 Aug 2007 10:08:10 +0100
From   : julesrichardsonuk@... (Jules Richardson)
Subject: Acquired Beeb with symptoms...

Robert Schmidt-Cisternas wrote:
> Sprow wrote:
>> It's the 6850 ACIA. You can also yank out the ADC (u7002), any speech
> chips
>> (TMS5220/TMS6100), the disc controller (8271), econet (68B54), and
> still
>> have something that'll boot. Note, not all these parts may be present
> since
>> some were offered as upgrades,
> 
> To make it easier to test the speed changes, I used a simple FOR-loop
> (FOR I%=1 to 10000:NEXT):
> 
> - BeebEm: 2 seconds (I'd appreciate a rough measurement from a real
> working Beeb, anyone?)

Yep, 5 seconds here on a model B with everything in place, with the FDC 
removed, with the ADC removed, with both FDC and ADC removed - i.e. it seems 
to make no difference. (Which is interesting in light of Peter's comments 
about S9)

> - Beeb with no changes and 8271 out: 7 seconds
> - Beeb with ADC out: 4 seconds (Repton is also correspondingly faster,
> but still unplayable)

See, that is strange. I'm not quite sure why removing the ADC would partially 
fix it, as the ADC doesn't appear to generate interrupts anyway - it's purely 
polled by the system.

What was the figure above for no changes and FDC in? Or doesn't that make any 
difference?

If you've got a multimeter, checking that the +5V power rails are behaving 
themselves is probably a good check; I've seen flakey power supplies do some 
really weird things to systems before and it'd be an easy one to rule out (if 
you have a meter!). The fact that pulling random chips seems to make a 
difference just makes me wonder if it's not actually an interrupt problem 
(although I'd agree with Sprow; that would have been my first assumption after 
ruling out the PSU)

cheers

Jules