Re: How much of a ][ is needed to twitch?

[Tristan Mumford <xtristan.xmumford@xgmail.xcom>]

...what a mess.

Michael J. Mahon wrote:

> Tristan Mumford wrote:
>> Michael J. Mahon wrote:
>> 
>> 
>>>Tristan Mumford wrote:
>>>
>>>>Michael J. Mahon wrote:
>>>>
>>>>
>>>>
>>>>>Tristan Mumford wrote:
> 
> <snip>
> 
>>>Whatever the power-on pattern in the video RAM is, viewed as text.
>>>
>>>Often this pattern is something like:  00 00 FF FF...
>>>
>> 
>> Heyyy..... That is most intriguing. viewed as text, as in standard 40col
>> text mode. What I'm seeing seems to be something approximating HGR, I
>> think anyway.
>> 
>> I may have to try to take a screenshot for everyone to look at.
> 
> That sounds unusual, but if the processor isn't initializing the
> system, it could come up in HGR mode.

It'll probably be two days until I can attempt the screenshot. The apple is
still in the shed and I need to attach its scary XT power supply.

-snip-


>> They -(RAM)- are all there in body, but I can't guarantee their sanity.
> 
> If there is more than one bank, you might try swapping with bank 0.
> The odds of more than one failure are pretty low.

It's just in possession of the bog standard complement of everything, bar
the weird 16k RAM expansion.




> 
>>>>I'll have to make a logic probe. I think I used to have one a long time
>>>>ago. I'm really tempted to cheat and make a uC based one. That way I
>>>>could have a few custom functions like a tone or light when a pulse is
>>>>detected. Just a thought anyway.
>>>
>>>True, and you could even make a simple logic analyzer!
>> 
>> 
>> What exactly do you have in mind?
> 
> You could actually save the times of logic transitions to capture
> the "waveform" at whatever sample clock your uC can handle.
> 
> If you sample multiple lines, you could save the states into a circular
> buffer and "trigger" on preset conditions to see what happened before
> and after the trigger.
> 
> Since most of what happens around the processor is 1MHz, sampling at
> 5MHz or so would provide reasonable detail.

Hmm... The dallas semiconductor 8051 clone may be up to that job. IIRC it
can do a max of 1 instruction per clock tick at 33MHz. I guess that would
be 33 MIPS. It would allow enough time to sample, store, and process
multiple sources.

still, a logic probe would be adequate for now :)

> 
>> Something came to me last night while trying to get to sleep. One of my
>> 40pin uCs, and some pullups on all the used IO pins. Depending on the
>> states of the pins, it would be enough to power, earth, provide input and
>> monitor output for the standard 74 series and maybe some RAMs.
>> Essentially a programmable tester.
> 
> Interesting approach--I like it!

I see no reason why it wouldn't work. I know the PIC uCs at least could do
it. I kind of like the ability to configure all logic states and
powersupply in software. I wonder if there would be a market for something
like that. It could even be used to generate truth tables from a chip,
which would be the easiest way to test a chip anyway. Come to think of it
that means it could also do things like help expose the darkest secrets of
closed chips, and also dump most types of ROMS. I really have to build one
now.



> 
>> But I already have a "minor" project in the works already. it's a2
>> related and it's a secret :) It will be freely available. I just don't
>> want to get people's hopes up if it doesn't work right.
>> 
>> For now I think a logic probe will be adequate.
> 
> Very simple and very handy.

Yes.

I tested my "secret" project today. It didn't work. When I start seeing
results / make it look less pathetic / pose less danger to people's
precious hardware I'll reveal more. Or I guess if I'm asked nicely.

> 
>>>>Theoretically, if one were to pull every RAM chip out and make an
>>>>expansion card with the correct amount of RAM (or at least disabled in
>>>>the ROM region) would that work? Ie would an otherwise healthy Apple
>>>>start?
>>>
>>>No, since reset will switch to main board RAM.
>> 
>> 
>> As a certain jaundiced animated man would say, "D'Oh!"
> 
> I really should have said "main board ROM", since that's the
> address space that a RAM card maps to.  It is possible for a
> slot card to disable all main board address decoding, and so
> take over for main board ROM *and* RAM, but I'm not aware of
> any manufactured cards that do so.  This would be the only way
> to get page zero, etc., mapped to a slot card.

Shame. Such a card would be very useful.


-- 
-----> http://members.dodo.com.au/~izabellion1/tristan/index.html <-----
=====       It's not pretty, it's not great, but it is mine.       =====