Keeping old hardware alive?
Peter Ibbotson
news:3e68e481$0$12819$afc3...@news.optusnet.com.au...
> I'm not sure if my old Apple II+ is old enough to be relevant to this
> group? :) Anyway, I was just wondering if anyone had tips on keeping
> old hardware alive. My Apple still fires up as if it was new, but I
> only pull it out once a year or so. I've heard that keeping old
> electronic devices in good condition requires powering them up every
> now and then? My question is, how often should I turn this old beast
> on? Any tips on keeping it alive over the next 10 or 20 years would be
> appreciated...
>
Old enough, but not really the right group try comp.sys.apple2 instead.
--
Work pet...@lakeview.co.uk.plugh.org | remove magic word .org to reply
Home pe...@ibbotson.co.uk.plugh.org | I own the domain but theres no MX
Rupert Pigott
news:3e68e481$0$12819$afc3...@news.optusnet.com.au...
> I'm not sure if my old Apple II+ is old enough to be relevant to this
> group? :) Anyway, I was just wondering if anyone had tips on keeping
> old hardware alive. My Apple still fires up as if it was new, but I
> only pull it out once a year or so. I've heard that keeping old
> electronic devices in good condition requires powering them up every
> now and then? My question is, how often should I turn this old beast
> on? Any tips on keeping it alive over the next 10 or 20 years would be
> appreciated...
In general I've found stuff like power supplies and capacitors
are the biggest risks. Oh yeah, and EPROM/EEPROM fade too.....
Cheers,
Rupert
Rupert Pigott
> up over time. Haven't heard of EPROM/EEPROM fade before. I don't
I've heard that electrolytics can go closed-circuit
with lack of use and need "reforming" (basically
trickling charge into them). Can't say I've actually
seen that first hand though. Other types of caps
die with time too, although I can't for the life of
me remember the timescales etc. It's been a very
long time since I cared... Telecomms kit often has
components guaranteed for 15 years minimum - that
puts a fairly hefty premium on the prices.
> think (but could be very wrong) that the Apple used eproms. Hmm..
> just popped the lid, I can see some of the ROM's have labels that
> could be covering an erase window...
Most of the early "Home computer" type micros I've come
across use PROMs for the vendor supplied important stuff
(like basic interpreters, sub-routines etc) - which are
great because they don't fade.
Cheers,
Rupert
Foobar T. Clown
>
> EPROM/EEPROM fade? Ouch. I've heard that caps are likely to dry
> up over time. Haven't heard of EPROM/EEPROM fade before. I don't
> just popped the lid, I can see some of the ROM's have labels that
> could be covering an erase window...
I don't remember what's under the hood of an Apple ][, but werent
the classic 27XX E-PROMS just common as dirt in those days?
Anyway, just because you don't see a window doesn't mean it's not
an E-PROM. A chip in a windowed, ceramic package might have cost
fifteen dollars each whereas the same chip in a windowless, plastic
package would have sold for two or three dollars in quantity. They
were called OTP (one time programmable) parts. (I guess it would've
been too weird to call them Non-Eraseable EPROMS.)
-- Foo!
Lars Duening
> EPROM/EEPROM fade?
Yep. E(E)PROM work by keeping a charge on an isolated gate (a capacitor
in effect), but the isolation can't be perfect, otherwise it would not
be possible to program the chip in the first place. So over time the
stored charge trickles off the gate (insert appropriate quantum
semiconductor mechanical terms here).
dgr...@cs.csbuak.edu
>> Yep. E(E)PROM work by keeping a charge on an isolated gate (a capacitor
>> in effect), but the isolation can't be perfect, otherwise it would not
>> be possible to program the chip in the first place. So over time the
>> stored charge trickles off the gate (insert appropriate quantum
>> semiconductor mechanical terms here).
> Great. I guess my grandchildren may not see what computing was for the
> average person in the 70's/80's. I suppose emulators will have to do,
> although it's really not the same. I'm off to play Wavy Navy before the
> PROM's expire....
Well, as long as you keep the ROM images around and keep copying them onto
newer media, you should be able to show your grandkids the old hardware.
Of course, some other rebuilding may be needed (caps, new [E|EE]PROMS,
etc).
Another thing you might want to explore is reimplementing old hardware on
FPGA. Examples: http://c64upgra.de/c-one/ (C64) and
http://www.mindspring.com/~2600onachip/ (Atari 2600).
--
David Griffith
Charles Richmond
>
> > Yep. E(E)PROM work by keeping a charge on an isolated gate (a capacitor
> > in effect), but the isolation can't be perfect, otherwise it would not
> > be possible to program the chip in the first place. So over time the
> > stored charge trickles off the gate (insert appropriate quantum
> > semiconductor mechanical terms here).
>
> average person in the 70's/80's. I suppose emulators will have to do,
> although it's really not the same. I'm off to play Wavy Navy before the
> PROM's expire....
>
or electric bill...or maybe the department store bill. Most
people in the 1970's had *no* computers...and many had *no*
hand-held calculators.
--
+-------------------------------------------------------------+
| Charles and Francis Richmond <rich...@plano.net> |
+-------------------------------------------------------------+
CBFalconer
Of course you can take those 'faded' eproms, windowless or not,
and reprogram them at any time without erasure. Just make sure
you are refreshing with the identical code. If they are 2708s you
may have trouble finding a programmer.
--
Chuck F (cbfal...@yahoo.com) (cbfal...@worldnet.att.net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home.att.net> USE worldnet address!
Glen Herrmannsfeldt
"Lars Duening" <la...@bearnip.com> wrote in message
news:1frgk4w.10tcpot17d0itcN%la...@bearnip.com...
> Biggles <nos...@optusnet.com.au> wrote:
(snip)
> > > In general I've found stuff like power supplies and capacitors
> > > are the biggest risks. Oh yeah, and EPROM/EEPROM fade too.....
> >
> > EPROM/EEPROM fade?
>
> Yep. E(E)PROM work by keeping a charge on an isolated gate (a capacitor
> in effect), but the isolation can't be perfect, otherwise it would not
> be possible to program the chip in the first place. So over time the
> stored charge trickles off the gate (insert appropriate quantum
> semiconductor mechanical terms here).
They are supposed to last many years if kept dark. Fluorescent lights have
enough
UV to erase them in some number of weeks, though, and sunlight in hours.
Those black stickers should get them to 50 years or so, if properly
programmed in the first place.
-- glen
jmfb...@aol.com
Then this begs the question similar to the light going out
in the refrigerator. How do you know that you're keeping the
light out?
/BAH
Subtract a hundred and four for e-mail.
bfra...@jetnet.ab.ca
>
> Then this begs the question similar to the light going out
> in the refrigerator. How do you know that you're keeping the
> light out?
Well I unplug the fridge? :)
Ben.
PS. Mind you that is easy to test, and my light does go out.
Charles Richmond
>
> In article <Dasaa.13163$sf5....@rwcrnsc52.ops.asp.att.net>,
> "Glen Herrmannsfeldt" <g...@ugcs.caltech.edu> wrote:
> >
> >
> > Those black stickers should get them to 50 years or so, if properly
> > programmed in the first place.
>
> Then this begs the question similar to the light going out
> in the refrigerator. How do you know that you're keeping the
> light out?
>
"I don't snore. One night I stayed up all night to see if I
snored, and I didn't."
And of course everyone remembers the famous quote from the
British Foreign Secretary prior to World War I:
"The refrigerator lights are going out all over Europe..."
Lars Duening
> Then this begs the question similar to the light going out
> in the refrigerator.
Now you reminded me of one episode of the TV show 'Sledge Hammer', where
Hammer tries to open a refrigerator and shoot the light before it goes
on.
Dosius
Now if only the Apple //e could be put on a single chip. ;)
-uso.
Neil Franklin
> dgr...@cs.csbuak.edu wrote in message news:<b4c013$23mrj$1...@hades.csu.net>...
> >
> > Well, as long as you keep the ROM images around and keep copying them onto
> > newer media, you should be able to show your grandkids the old hardware.
> > Of course, some other rebuilding may be needed (caps, new [E|EE]PROMS,
> > etc).
> >
> > Another thing you might want to explore is reimplementing old hardware on
> > FPGA. Examples: http://c64upgra.de/c-one/ (C64) and
> > http://www.mindspring.com/~2600onachip/ (Atari 2600).
>
> Now if only the Apple //e could be put on a single chip. ;)
Any sufficiently simple computer (that includes any 8bit micro, by a
wide margin) can today be put into an single FPGA chip.
The way the 2600 guy is doing it, is correct. Why the c64 project is
using 2 FPGAs and an separate 6502 chip (originally soldered in, now
an slot and processor modules) is beyond me. The reason given on the
website (processor needs too much space in the FPGA) is seriously
outdated.
Todays $50 FPGAs such as XC2S200 will fit any 8bit microprocessors in
1/4..1/2 of an single chip! I am aiming an PDP-10 (KA10) clone at that
chip. 1/2 of the instruction set and all central mechanisms are done
in 1/4 space used. Aim is 1/3..1/2 processor, 1/4..1/3 memory managment,
rest IO devices.
For an project that wants to be an universal (and reloadable!) 8bit
clone having to have hard swappable modules for every processor type
is nonsensical. In particular as completed 6502 and Z80 clones are
available for free from the net.
--
Neil Franklin, ne...@franklin.ch.remove http://neil.franklin.ch/
Hacker, Unix Guru, El Eng HTL/BSc, Programmer, Archer, Blacksmith
- hardware runs the world, software controls the hardware
code generates the software, have you coded today?
Charles Richmond
>
> st...@dosius.zzn.com (Dosius) writes:
>
> > dgr...@cs.csbuak.edu wrote in message news:<b4c013$23mrj$1...@hades.csu.net>...
> > >
> > > Well, as long as you keep the ROM images around and keep copying them onto
> > > newer media, you should be able to show your grandkids the old hardware.
> > > Of course, some other rebuilding may be needed (caps, new [E|EE]PROMS,
> > > etc).
> > >
> > > Another thing you might want to explore is reimplementing old hardware on
> > > FPGA. Examples: http://c64upgra.de/c-one/ (C64) and
> > > http://www.mindspring.com/~2600onachip/ (Atari 2600).
> >
> > Now if only the Apple //e could be put on a single chip. ;)
>
> Any sufficiently simple computer (that includes any 8bit micro, by a
> wide margin) can today be put into an single FPGA chip.
>
> The way the 2600 guy is doing it, is correct. Why the c64 project is
> using 2 FPGAs and an separate 6502 chip (originally soldered in, now
> an slot and processor modules) is beyond me. The reason given on the
> website (processor needs too much space in the FPGA) is seriously
> outdated.
>
a 65816 chip, which is a sixteen bit version of the 6502. That
would certainly take up more space in the FPGA.
Dosius
> For an project that wants to be an universal (and reloadable!) 8bit
> clone having to have hard swappable modules for every processor type
> is nonsensical. In particular as completed 6502 and Z80 clones are
> available for free from the net.
Not to mention the C-1's own 65C816.
-uso.
Neil Franklin
> Neil Franklin wrote:
> >
> > st...@dosius.zzn.com (Dosius) writes:
> >
> > > Now if only the Apple //e could be put on a single chip. ;)
> >
> > Any sufficiently simple computer (that includes any 8bit micro, by a
> > wide margin) can today be put into an single FPGA chip.
> >
> > The way the 2600 guy is doing it, is correct. Why the c64 project is
> > using 2 FPGAs and an separate 6502 chip (originally soldered in, now
> > an slot and processor modules) is beyond me. The reason given on the
> > website (processor needs too much space in the FPGA) is seriously
> > outdated.
> >
> IIRC, the new C64 project is *not* using a 6502 chip...it is using
> a 65816 chip, which is a sixteen bit version of the 6502.
Which is roughly the same thing.
Index registers and stack pointer widended 8->16bit. A few more
instructions. And an very primitive 16->24bit address extension
section. Still smaller than an Z80.
> That
> would certainly take up more space in the FPGA.
Hardly any. Not relevantly more.
Foobar T. Clown
>
> Great. I guess my grandchildren may not see what computing was for
> the average person in the 70's/80's. I suppose emulators will have
> to do, although it's really not the same.
It might not be too bad, assuming a full immersion virtual reality
emulator. I'm picturing a full, photo-realistic simulation of the
computer, the 12" color TV set sitting on top of it, the cassette
recorder, cassettes, etc. Throw in appropriate audio and tactile
cues, and it could be a pretty decent recreation of the experience.
-- Foo!
Foobar T. Clown
>
> Charles Richmond <rich...@ev1.net> wrote:
> >>
> > Computing for the average person in the 1970's was their phone
> > or electric bill...or maybe the department store bill. Most
> > people in the 1970's had *no* computers...and many had *no*
> > hand-held calculators.
>
Maybe so for computers, but calculators are dirt cheap, and they
don't need to be plugged in, and a lot of them don't even need
batteries. I'm pretty sure that there aren't many places on
earth where the average person could not have one if he really
wanted it.
OTOH, there probably are a lot of places on earth where you'd
have to search far and wide to find somebody who wouldn't gladly
give up a calculator in exchange for a good plastic water jug or
a garden hoe.
As for the 70's, that was a decade of considerable change.
In 1970, There was only ONE hand held, four function calculator:
http://www.hpmuseum.org/ffcurta.htm
You could buy it from an ad that ran in the back of Scientific
American for something like $350 dollars. (Remember, a gallon
of gas cost about a quarter back then).
By 1979, when I was a high school senior, we were REQUIRED to
have a scientific calculator with trig functions for some of our
math and science classes.
-- Foo!
Bernd Felsche
Was that because they thought you were hopeless at using tables and
slide-rules? :-)
--
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ / ASCII ribbon campaign | I'm a .signature virus!
X against HTML mail | Copy me into your ~/.signature
/ \ and postings | to help me spread!
Steve O'Hara-Smith
"Foobar T. Clown" <fu...@gazonk.del> wrote:
FTC> By 1979, when I was a high school senior, we were REQUIRED to
FTC> have a scientific calculator with trig functions for some of our
FTC> math and science classes.
Hmmm, I took my A levels in 1977. We were permitted to use
calculators, but we were expected to be able to use tables and/or slide
rules and most of us were faster with a slide rule (8 years of daily
practice) than with a calculator - which oddly enough didn't stop most
of us choosing to use calculators. IIRC calculators were not permitted
in my O levels in 1975, but a decent slide rule was essential.
--
C:>WIN | Directable Mirrors
The computer obeys and wins. |A Better Way To Focus The Sun
You lose and Bill collects. | licenses available - see:
| http://www.sohara.org/
Foobar T. Clown
>
> "Foobar T. Clown" <fu...@gazonk.del> writes:
>
> >By 1979, when I was a high school senior, we were REQUIRED to
> >have a scientific calculator with trig functions for some of our
> >math and science classes.
>
> Was that because they thought you were hopeless at using tables and
> slide-rules? :-)
1978, or maybe 1979: One of my math teachers brought out a big ol'
demonstration slide rule (a foot and a half wide and maybe five or
six feet long.) He showed us how it was used back in The Old Days.
It was basically a history lesson. We never were expected to use
one for real.
Tables... I vaguely remember learning about how to read logarithm
tables and, about linear interpolation. We probably even had home
work on that subject, but I don't remember ever having to USE tables
in the solution of some OTHER homework problem.
-- Foo!
Brian Inglis
"Foobar T. Clown" <fu...@gazonk.del> wrote:
Pffhh! Newbie! ;^>
Thanks. Take care, Brian Inglis Calgary, Alberta, Canada
--
Brian....@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca)
fake address use address above to reply
ab...@aol.com tos...@aol.com ab...@att.com ab...@earthlink.com
ab...@hotmail.com ab...@mci.com ab...@msn.com ab...@sprint.com
ab...@yahoo.com ab...@cadvision.com ab...@shaw.ca ab...@telus.com
ab...@ibsystems.com u...@ftc.gov spam traps
Steve O'Hara-Smith
FTC> 1978, or maybe 1979: One of my math teachers brought out a big ol'
How old were you then ? That was about the time I left school.
FTC> demonstration slide rule (a foot and a half wide and maybe five or
FTC> six feet long.) He showed us how it was used back in The Old Days.
Ye gods. For learning the principles of slide rules - we made
our own Napiers Bones. That would have been first year maths in 1970
I think.
FTC> It was basically a history lesson. We never were expected to use
FTC> one for real.
Damn but that was recent history - between 1970 and 1975 slide
rule practice was a daily event. It's amazing how quickly they became
subjects of a history lesson.
FTC> Tables... I vaguely remember learning about how to read logarithm
FTC> tables and, about linear interpolation. We probably even had home
FTC> work on that subject, but I don't remember ever having to USE tables
FTC> in the solution of some OTHER homework problem.
Gronk, my four figure tables were the most dog eared publication
I had in my school days.
Nick Spalding
> On Mon, 17 Mar 2003 01:31:58 GMT
> "Foobar T. Clown" <fu...@gazonk.del> wrote:
>
> FTC> 1978, or maybe 1979: One of my math teachers brought out a big ol'
>
> How old were you then ? That was about the time I left school.
>
> FTC> demonstration slide rule (a foot and a half wide and maybe five or
> FTC> six feet long.) He showed us how it was used back in The Old Days.
>
> Ye gods. For learning the principles of slide rules - we made
> our own Napiers Bones. That would have been first year maths in 1970
> I think.
>
> FTC> It was basically a history lesson. We never were expected to use
> FTC> one for real.
>
> Damn but that was recent history - between 1970 and 1975 slide
> rule practice was a daily event. It's amazing how quickly they became
> subjects of a history lesson.
>
> FTC> Tables... I vaguely remember learning about how to read logarithm
> FTC> tables and, about linear interpolation. We probably even had home
> FTC> work on that subject, but I don't remember ever having to USE tables
> FTC> in the solution of some OTHER homework problem.
>
> Gronk, my four figure tables were the most dog eared publication
> I had in my school days.
I still have mine and I left school in 1949. The most recent set I have is
a 1958 edition of Chambers's Seven-Figure Mathematical Tables. I also have
a 1940 edition of Inman's Nautical Tables and a 1953 edition of the
three-volume set of AP3270 Sight Reduction Tables for Air Navigation. I
like tables.
--
Nick Spalding
Pete Fenelon
> I still have mine and I left school in 1949. The most recent set I have is
> a 1958 edition of Chambers's Seven-Figure Mathematical Tables. I also have
I have a 1942 edition on my desk, purchased for ten bob a couple of
weeks back ;)
> a 1940 edition of Inman's Nautical Tables and a 1953 edition of the
> three-volume set of AP3270 Sight Reduction Tables for Air Navigation. I
> like tables.
I used to have a three-volume "Manual of Gear Design" that was
hand-written and typeset from that rather than typeset. Gorgeous. I
don't know where it's gone :(
pete
--
pe...@fenelon.com "there's no room for enigmas in built-up areas" HMHB
David Powell
Steve O'Hara-Smith <ste...@eircom.net> in alt.folklore.computers
wrote:
>On Mon, 17 Mar 2003 01:31:58 GMT
>"Foobar T. Clown" <fu...@gazonk.del> wrote:
>
>FTC> 1978, or maybe 1979: One of my math teachers brought out a big ol'
>
> How old were you then ? That was about the time I left school.
>
>FTC> demonstration slide rule (a foot and a half wide and maybe five or
>FTC> six feet long.) He showed us how it was used back in The Old Days
>
> Ye gods. For learning the principles of slide rules - we made
>our own Napiers Bones. That would have been first year maths in 1970
>I think.
>FTC> It was basically a history lesson. We never were expected to use
>FTC> one for real.
>
> Damn but that was recent history - between 1970 and 1975 slide
>rule practice was a daily event. It's amazing how quickly they became
>subjects of a history lesson.
>
"O" levels for me in about 1960. All calculations performed with log
tables. Slide rules forbidden. (Hint - the old farts who taught
couldn't use slip sticks) Allowed for "A" levels. Bloody
frustrating as a kid, Dad taught me to use a slide rule at an early
age, at least I could use it for homework to check my logs work for
gross errors. In retrospect, like so many other things at school
(Latin, etc) boring at the time, but I found the payback later in
life.
The only tabular data that find an occasional use for now is my 1963
copy of Callendars' Steam Tables: if I needed to use them more often,
I'm sure that I could knock up some ForTran (or FOCAL)...
>FTC> Tables... I vaguely remember learning about how to read logarithm
>FTC> tables and, about linear interpolation. We probably even had home
>FTC> work on that subject, but I don't remember ever having to USE tables
>FTC> in the solution of some OTHER homework problem.
>
> Gronk, my four figure tables were the most dog eared publication
>I had in my school days.
In my day, that would have been a copy of "Lady Chatterley's Lover".
D.H.L. was an old boy of the school, the subject was "sensitive" and
like the products of another old boy, John Dane Player, something
secreted in hidden pockets, produced only in quiet places, when
authority was looking the other way.
Regards,
David P.
Andrew Gabriel
"Foobar T. Clown" <fu...@gazonk.del> writes:
>
> 1978, or maybe 1979: One of my math teachers brought out a big ol'
> demonstration slide rule (a foot and a half wide and maybe five or
> six feet long.) He showed us how it was used back in The Old Days.
> It was basically a history lesson. We never were expected to use
> one for real.
My dad has a high accuracy one, but instead of being linear, it's
wrapped round a brass cylinder like a screwthread with a second
cylinder moving/rotating like a telecope. I would probably
recognise the make if someone mentioned it, but I can't recall it
off the top of my head although it might be German.
I used slide rules during my secondary schooling in the 1970's.
Also had a calculator then -- I think we could use either/both
in exams (and logbooks), but programmable calculators were not
allowed.
I wish I still had one of my slide rules. I seem to recall going
through a few whilst at school (typically cursor would break).
I still seem to have all my 1970's calculators;-) First was a
Prinztronic(sp?) with just 4 functions and sqrt, LED display and
probably about 1 hour battery life, and second was a Sharp
scientific one (which was the first time I heard of Sharp).
Both still work.
--
Andrew Gabriel
Consultant Software Engineer
Anne & Lynn Wheeler
> I used slide rules during my secondary schooling in the 1970's.
> Also had a calculator then -- I think we could use either/both
> in exams (and logbooks), but programmable calculators were not
> allowed.
i saved up so i could buy the cheapest one from the sears catalogue
when i was 12 ... i think it was something like $3.89. no idea where
it is now.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Prof Karl Kleine
> In article <3E752568...@gazonk.del>,
> "Foobar T. Clown" <fu...@gazonk.del> writes:
>>
>> 1978, or maybe 1979: One of my math teachers brought out a big ol'
>> demonstration slide rule (a foot and a half wide and maybe five or
>> six feet long.) He showed us how it was used back in The Old Days.
>> It was basically a history lesson. We never were expected to use
>> one for real.
> My dad has a high accuracy one, but instead of being linear, it's
> wrapped round a brass cylinder like a screwthread with a second
> cylinder moving/rotating like a telecope. I would probably
> recognise the make if someone mentioned it, but I can't recall it
> off the top of my head although it might be German.
google for "Otis King"
[...]
> --
> Andrew Gabriel
> Consultant Software Engineer
________________________________________________________
Prof. Karl Kleine http://www.fh-jena.de/~kleine
Fachhochschule Jena kle...@fh-jena.de
Carl-Zeiss-Promenade 2 +49-3641-205-502 [fax -503]
D-07745 Jena, Germany
none Carl Lowenstein
Andrew Gabriel <and...@cucumber.demon.co.uk> wrote:
>In article <3E752568...@gazonk.del>,
> "Foobar T. Clown" <fu...@gazonk.del> writes:
>>
>> 1978, or maybe 1979: One of my math teachers brought out a big ol'
>> demonstration slide rule (a foot and a half wide and maybe five or
>> six feet long.) He showed us how it was used back in The Old Days.
>> It was basically a history lesson. We never were expected to use
>> one for real.
>
>My dad has a high accuracy one, but instead of being linear, it's
>wrapped round a brass cylinder like a screwthread with a second
>cylinder moving/rotating like a telecope. I would probably
>recognise the make if someone mentioned it, but I can't recall it
>off the top of my head although it might be German.
Otis King. British. Sometimes sells for exorbitant prices on eBay.
carl
--
carl lowenstein marine physical lab u.c. san diego
clow...@ucsd.edu
Charles Shannon Hendrix
> Todays $50 FPGAs such as XC2S200 will fit any 8bit microprocessors in
> 1/4..1/2 of an single chip! I am aiming an PDP-10 (KA10) clone at that
> chip. 1/2 of the instruction set and all central mechanisms are done
> in 1/4 space used. Aim is 1/3..1/2 processor, 1/4..1/3 memory managment,
> rest IO devices.
Yes, but what about the support chips?
For example, the Atari had several custom chips that were non-trivial.
Seems doable, but I would imagine it would need more than one
FPGA.
Of course, it would be cool to see it.
> For an project that wants to be an universal (and reloadable!) 8bit
> clone having to have hard swappable modules for every processor type
> is nonsensical. In particular as completed 6502 and Z80 clones are
> available for free from the net.
Can you get one to the point where you can hook up a terminal and
boot up CP/M and run WordStar? That would be interesting.
I'd also like to use that for really small embedded projects.
Marada C. Shradrakaii
>> clone having to have hard swappable modules for every processor type
>> is nonsensical. In particular as completed 6502 and Z80 clones are
>> available for free from the net.
Well, why not make a 6502-on-a-FPGA and plug that in the socket?
Actually, I can see using the hard-wired processor as a money-saver, as it may
well be cheaper than the analogous FPGA.
Moreover, what you can do with a box like that depends on configuring the
entire machine-- a fairly intensive task. What good is it to be able to
emulate the Florko 7000's CPU, if the rest of the Florko 7000 hardware is not
implemented? Similarly, if all that's available out of the box is a 6502-based
model, why not just stick with a 6502 you can replace if and when replacements
appear. In addition, designing your own CPU may prove to be an unneeded
feature-- the most common 8-bit CPUs are still available, and it's not
something that many people are going to need to modify if their goal is either
new development or emulating their fave 8-bit.
--
Marada Coeurfuege Shra'drakaii
On the Internet, all roads lead to either pornography or a GNU/Linux HOWTO.
Which way are you going?
Mail hint: Not in Russia
Neil Franklin
> In article <6u65qpi...@chonsp.franklin.ch>, Neil Franklin wrote:
>
> > Todays $50 FPGAs such as XC2S200 will fit any 8bit microprocessors in
> > 1/4..1/2 of an single chip! I am aiming an PDP-10 (KA10) clone at that
> > chip. 1/2 of the instruction set and all central mechanisms are done
> > in 1/4 space used. Aim is 1/3..1/2 processor, 1/4..1/3 memory managment,
> > rest IO devices.
>
> Yes, but what about the support chips?
>
> For example, the Atari had several custom chips that were non-trivial.
Unlikely to be much larger than 10-20'000 transistors, given the date of
introduction of the VCS. So that is 2'500-5'000 2-input NANDs. Same size
as the 6502 itsself.
Harmless in comparison to the XC2S200 FPGA, that fits the equivalent of
30-200'000 2-input NANDs (depending on use). Strictly the XC2S200 is
56x84=4704 4-input random function elements, each paired with an
hardwired 2:1 Mux and an Flipflop, plus 14 512byte SRAMs, plus 140-264
user IO pins. That is a cheap $50 FPGA. If you can fit in 16x24=384
element/Mux/FF and 4 SRAMs, you are in at $5 (XC2S15).
So space is no problem. Knowing what those chips did, may be, given
how Atari was closed on documentation.
> Seems doable, but I would imagine it would need more than one
> FPGA.
Not with todays large ones. 10 years ago it would have still required
multiple.
> > For an project that wants to be an universal (and reloadable!) 8bit
> > clone having to have hard swappable modules for every processor type
> > is nonsensical. In particular as completed 6502 and Z80 clones are
> > available for free from the net.
>
> Can you get one to the point where you can hook up a terminal and
> boot up CP/M and run WordStar? That would be interesting.
Sure one can. Anything that can be done on a chip, can be done on an
FPGA, subject to size and speed limits, which correspond to hardwired
chips about 6-9 (2-3 generations) years earlier.
> I'd also like to use that for really small embedded projects.
No problem. Just programming work.
Neil Franklin
> >> For an project that wants to be an universal (and reloadable!) 8bit
> >> clone having to have hard swappable modules for every processor type
> >> is nonsensical. In particular as completed 6502 and Z80 clones are
> >> available for free from the net.
>
> Well, why not make a 6502-on-a-FPGA and plug that in the socket?
Now that would have all disadvantages. Modules that need plugging, and
having to clone the CPU.
> Actually, I can see using the hard-wired processor as a money-saver, as it may
> well be cheaper than the analogous FPGA.
Not any more today.
CPU + module print + plug + socket on main print + smaller FPGA for
the custom stuff. All adds up. Just an larger FPGA is cheaper today.
Add in that the CPU/module/plug are pay multiple times if you want
multiple CPUs.
And the extra flexibility of no hardwired connection module<->main, and
"rebooting" into an different CPU at any time without open case and
replugging. And instant from-Internet CPU upgrades.
> Moreover, what you can do with a box like that depends on configuring the
> entire machine-- a fairly intensive task.
That was implicit in the entire (sub-)thread. We were talking about an
re-implemented Atari VCS and C64. The VCS full-FPGA and the C64 with
an hardware CPU and the rest in FPGA. And contrasting the 2 approaches.
> What good is it to be able to
> emulate the Florko 7000's CPU, if the rest of the Florko 7000 hardware is not
> implemented?
That has to be implemented anyway.
> Similarly, if all that's available out of the box is a 6502-based
> model, why not just stick with a 6502 you can replace if and when replacements
> appear.
You can not download an newer processor from the Internet, install and
reboot. You can with the FPGA, just like downloading and installing an
new OS release or even an different OS.
Charles Shannon Hendrix
> Unlikely to be much larger than 10-20'000 transistors, given the date of
> introduction of the VCS. So that is 2'500-5'000 2-input NANDs. Same size
> as the 6502 itsself.
I mean the 8-bit computers, not the VCS.
Antic, POKEY, and the gang are probably small, but there are quite a few
of them to cram into one FPGA it seems.
My old univeristy now has a graduate degree in parallel computation,
and they use FPGAs to build parallel computers.
> So space is no problem. Knowing what those chips did, may be, given
> how Atari was closed on documentation.
Sounds good then.
I think there is enough information about Atari chips. While they were
closed, there was a lot of information on them.
You probably have enough information to create a compatible clone,
and there are plenty of them still working for testing purposes.
> Not with todays large ones. 10 years ago it would have still required
> multiple.
Good.