19 Jun
2012
19 Jun
'12
8:10 p.m.
were created in the 80s). But as far as learning
computers, I don't
IMHO if you want to learn how computers really work you shouldn't be
using a microprocessor at all :-). The _real_ way to do it is to build
your own (simple) processor from TTL or similar (IMHO doing it in a FPGA
is significantly less educational), or almost as good is to get a machine
with such a processor (early 1970s mini...) and do battle with the
schematics and a 'scope/LA.
necessarily see the benefit of a sbc (much less
an sdk w/a keypad),
as opposed to an established system.
That all depends on what you want to learn. If you want to learn how
computers actually *work*, which (IMNSHO) is a requirement for being
able to program worth a damn, this is a good route...as long as it's not
just treated like an appliance.
An "established system" (by which I think you mean "a commercial
packaged/closed system") just plain denies a person that option, unless
you go dig up the schematics for it.
And in that case, say for an Atari or an Amiga, there are so many
complex custom chips in there, a beginner isn't going to get much of an
education out of it.
AFAIk the Mac hardware was never rully documetned. There may be
reverse-engineered schematics of some older Macs, but they all have
either custom chips or copy-protected PALs, so you can't get very far...
Maybe something like an HP9816 (or another 9000/200 amchine). If you
avoild the ones with the MMU (9817, 9826U, HP9836U, basically), there are
no PALs, it's all standard ICs. The 9816 is quite simple, acutally,
Torch XXX? There are PALs, but they're not copy-protected IIRC.
Any other candidates?
-tony