14 Jun
2006
14 Jun
'06
4:03 p.m.
Tony Duell wrote:
Many (small) PALs/GALs can be reverse engineered "empirically".
Depends on how much buried state they have. If there is none
*buried* (i.e. if all of the state is visible on pins) I have
a routine I wrote many years ago that will exhaustively apply
test vectors and track state transitions to rebuild a state table
suitable for applying to a logic compiler (which will then
generate the actual logic equations). IIRC, it is a NS32000
COFF binary, though... I'd have to hunt to see which tape has
the sources (something I probably should do, anyway, since
tape isnt very long-lived)
Exactly. Rarely does the published design exactly meet my needs, I want
to modify it. And that may mean modifying the firmware too.
I don't expect them to support all my old machines, but I wish they'd let
_me_ do so by documentign the interface properly. If a device plugs into
a serial port and uses the Tx and Rx line conventionally, then I'd like
to know what commands I can send the device, and how it responds. Or
alternatively, give me the source to the firmware and/or the PC program
and I'll work it out.
Exactly. After all, these publications are intended for people
who *can* do that sort of thing!
Of course. Nor does collecting old computers! :>
Tony Duell
wrote:
Microcontrollers, PALs/GAls, CPLDs. All come with the security fuse
blown. Obviously EPROMs are readable :-) Elektor is the best of a dwindling number of
electronics magazines
available in the UK. It's improved in the last 5 years or so as they now
do make firmware sources available for about 50% of their projects (but
not the interesting ones, alas). Before that (and for the other 50% now)
you have to buy the programmed device from them, and it comes copy-protected.
"Copy protected" -- implying an MCU implementation?
I.e. *small* pieces of code? (not hundreds of KB) In a few cases, they make the object (binary) file
available to program
your own device, but not the source. Which is not particularly helpful.
Are these "old school" MCU's? ARMs? etc. I have quite a number
of dissasemblers that would make quick work of this. If the code
was written in a HLL, then it's usually very easy to deduce the
corresponding language constructs used (most compilers for tiny
processors are very literal). If it's written in ASM, then you'd
have to walk through it to deduce the appropriate commentary to
add.
Yes, I am well aware there are ways to
reverse-engineer some of these
devices, but I have to weigh up the effort to do that against the effort
to design the whole thing from scratch (and have it just the way I want
it).
Exactly. Something vendors fail to realize when the produce a
product that they think is so "special"/unique. But, for other
things, you don't have much choice (I disassembled the code in
my PBX -- about a megabyte -- just to figure out how it *really*
worked since the "help" was written by a nonenglish speaker :-(
Since I generally build in order to learn, this is a
major reason for me
not to build said project.
Yup. Or, at least to be able to *build* on their
ideas
(things they didn't think of or didn't implement well) Anotehr
disadvantage is that their computer-related projects invariably
have Windows software. The sources of that may not be available, the link
between the PC and the project (RS232, USB, parallel ports) may not be
properly documented.
This is true here (US), as well. Its as if the idea of
something
*other* than a PC never crosses their mind... I wonder if
the *apparent* (? unsubstantiated by hard numbers)
prevalence of hobbyists on your side of the pond has anything
to do with economics? I recall (when working with a firm in
Manchester) that many components that were dirt cheap here
were quite "dear", there. And, the "pounds = dollars"
relationship means things are effectively (i.e. from *my*
point of view) 60 - 80% more expensive than they would be,
here.
I've heard economic arguments (mostly as a reason for not homebrewing)
many times before, and I don't believe it's the whole story... Let me go
off at a tangent for a moment.
All my life I've been interested in clocks, and one day I intend to make
a real mechanical clock. I'll have to get a lot better at machining
first, but I still want to do it.
Economically, this makes no sense at all. To make a clock from scratch
probably needs \pounds 5000 to \pounds 10000 worth of tools (many of
which i have, it's true, but I'd still need to get gear cutters, etc and
they're not cheap). The metal to make said clock is around \pounds 100.
And the result is a much worse timekeeper than a 5 quid quartz clock from
the local household shop. But I still want to do it, to learn, and to have the
enjoyment of making
it. And it appears others do likewise. There's at least one magazine
which has an article every month on desinging/making a clock (and other
articles on making parts as part of antique clock restoration). There are
many books and plans for homebuilt clocks. So I can't be alone in wanting
to do this.
I share the avocation -- though most of the clocks I build are
electronic and just play with unique ways of displaying the time
(e.g., in Braille, etc. -- think about it! :> )
Now lets get back to computers. The difference in
performance between a
homemade mechanical clock and a quartz clock is probably comparable to
the difference in performance between a modern PC and a homebuilt
processor. The tools needed to make a homebuilt processor are cheaper
than those needed to make a clock, thouygh. And yet, I've not seen one
book or magazine article on making a processor from scratch in the last
10 years.
I think once the 2901 "fell from grace" (?), this became a thing
of the past. I've designed two processors "from scratch" (TTL
with bipolar ROMs for the microcode store) and found it quite
an interesting exercise. Not just the "logic design" but
actually thinking about what the instruction set should be
for that particular application domain, etc.
But, nowadays, I think it would be a lot less tedious if you
could do it in a big FPGA using synthesis tools. You could
get better test vector coverage (and *generation* of those
by the toolchain!) instead of having to design vectors from
scratch and single step the microcode (hint: fully synchronous
designs really off big, here!)
If some people think it's worth making a clock
from scratch, why not a
processor.
I will end with a wuote from a book entitled 'Every Boy his own
Mechanic', published about 90 years ago. It's the start of a chapter on a
home-made telephone system
' For some years previous to 1914 the constructive instinct and ability
inherent in many of us was almost snuffed out because most of the
mechanical and electrical things we wished to possess were so cheap that
they could be bought ready made. Often the prices quoted were actually
below the cost to us of the raw materials, and so we came to think that
<grin> I guess things never change! :>
constructive hobbies were a bit futile. Lots of those
cheap things were
amazingly good, but some where exasperatingly bad, and the bulk
indifferent. Still, we bought the stuff, learned nothing from it, and
clean missed all the solid satisfaction that lives for ever in building
and contriving and creating
[snip]
And that is exactly how I feel about homebrewing
electronic devices now.
So, is it your opinion that the "build it" mentality is
so individualistic that the differences in attitudes towards
it between our sides of the pond are more *cultural* (of
yet, undetermined origins) than anything else? E.g., the
US "disposable" society, consumerism, materialism, etc.?