6526a or 6526?
Not Necessarily Me
Oldsoftware.com has 6526A chips. Are these the same as 6526 chips?
Thanks,
Jeff
MagerValp
J> I need to replace at least one of the 6526 chips on the I/O board
J> in SX-64. Oldsoftware.com has 6526A chips. Are these the same as
J> 6526 chips?
They are 6526 chips that can run at up to 2 MHz. They will work just
fine in an SX-64.
--
___ . . . . . + . . o
_|___|_ + . + . + . Per Olofsson, arkadspelare
o-o . . . o + Mage...@cling.gu.se
- + + . http://www.cling.gu.se/~cl3polof/
White Flame (aka David Holz)
news:p14oeew...@panini.cling.gu.se...
> >>>>> "J" == Not Necessarily Me <n...@nnm.com> writes:
> J> I need to replace at least one of the 6526 chips on the I/O board
> J> in SX-64. Oldsoftware.com has 6526A chips. Are these the same as
> J> 6526 chips?
>
> They are 6526 chips that can run at up to 2 MHz. They will work just
> fine in an SX-64.
The 6526A's timer interrupts also trigger 1 cycle later than the 6526's, but
I don't think anything besides an esoteric demo or two rely on that
behavior. But yeah, Jeff, the 6526A is a drop-in replacement.
I'd like if somebody with both a 6526 and 6526A in their machines lying
around could test this program out to see if it detects it properly:
http://www.white-flame.com/timingtest.prg
--
White Flame (aka David Holz)
http://www.white-flame.com/
(spamblock in effect)
Mika Leinonen
>The 6526A's timer interrupts also trigger 1 cycle later than the 6526's, but
>I don't think anything besides an esoteric demo or two rely on that
>behavior. But yeah, Jeff, the 6526A is a drop-in replacement.
Are these two made different? Different IC-layout masks, so the logic is
slightly changed? I'm asking, because often integrated circuit have speed
grades "stamped" on the package, but they can be from the same production.
White Flame (aka David Holz)
news:cu8d9j$1gn$1...@news.cc.tut.fi...
> Are these two made different? Different IC-layout masks, so the logic is
> slightly changed? I'm asking, because often integrated circuit have speed
> grades "stamped" on the package, but they can be from the same production.
Seeing as there's a slight functional difference of that cycle offset at the
same 1MHz speed between the two versions, I'd say that they reworked the
logic internals. I'm guessing here, but I *think* the 6526's serial port
didn't work quite right, which is why they used a software based serial bus
instead of using the SP/CNT lines of the CIA. Doing a logic rework to fix
the bug might have changed the timer logic to introduce that 1-cycle offset.
Again, that's conjecture from vague recollection.
Mika Leinonen
>didn't work quite right, which is why they used a software based serial bus
My recollection is that the 6522 VIA had a bug, so that is why the serial
shift register wasn't used in the VIC-20. My recollection comes from
something I've read on the Internet.
Nicolas Welte
> The 6526A's timer interrupts also trigger 1 cycle later than the 6526's, but
> I don't think anything besides an esoteric demo or two rely on that
> behavior. But yeah, Jeff, the 6526A is a drop-in replacement.
This difference is not between 6526 and 6526A, but between 6526/6526A
manufactured before 1987 and ones that are manufactured later. They must've
changed the mask, I wonder if they went CMOS. The Aldi-C64, the C64G, the
(later) C64C and the C128DCR were made with this chip.
The trace command of SMON got broken with the change and needed fixing, and I
had a 15s copy program that didn't like the new chip (also fixable).
The interesting thing is that the 8521 (a 6526 replacement) and the 8520
Amiga CIA (even models from the 90s!) have the old timing.
> I'd like if somebody with both a 6526 and 6526A in their machines lying
> around could test this program out to see if it detects it properly:
>
> http://www.white-flame.com/timingtest.prg
Gonna check it out!
Nicolas
--
--> Email address is valid for replies (requires Re: in the subject) <--
- My Commodore hardware projects and the X1541 Shop at http://x1541.de -
Nicolas Welte
>> I'd like if somebody with both a 6526 and 6526A in their machines lying
>> around could test this program out to see if it detects it properly:
>>
>> http://www.white-flame.com/timingtest.prg
>
>
> Gonna check it out!
Okay, checked it with the 128DCR (new CIA) and it outputs 6526A. The old C64
(old CIAs) outputs 6526. Any chance you expand the test to include CIA #1 as
well, and maybe test the 1570/71/81 drive CIA? Here an additional test for
presence of a 8520 (different TOD registers) would be quite nice.
Cameron Kaiser
>http://www.white-flame.com/timingtest.prg
Intriguing. Source code ga arimasu ka?
--
Cameron Kaiser * cka...@floodgap.com * posting with a Commodore 128
personal page: http://www.armory.com/%7Espectre/
** Computer Workshops: games, productivity software and more for C64/128! **
** http://www.armory.com/%7Espectre/cwi/ **
Frank
the 6526A chips from them, and they packaged them poorly. Some of the
prongs were bent, and they just 'stashed' them in a used box instead of
packaging them to protect the chips. They arrived in a timely matter,
and the service was ok, but they just didn't seem to take the care to
package them very well.
MagerValp
NW> This difference is not between 6526 and 6526A, but between
NW> 6526/6526A manufactured before 1987 and ones that are manufactured
NW> later. They must've changed the mask, I wonder if they went CMOS.
I'm guessing they went from NMOS to HMOS, just like they did with the
other chips.
Nicolas Welte
>>>>>>"NW" == Nicolas Welte <welte...@freenet.de> writes:
>
>
> NW> This difference is not between 6526 and 6526A, but between
> NW> 6526/6526A manufactured before 1987 and ones that are manufactured
> NW> later. They must've changed the mask, I wonder if they went CMOS.
>
> I'm guessing they went from NMOS to HMOS, just like they did with the
> other chips.
But the strange thing is that the HMOS-II chips (8520 and 8521) have the
same timing like the old NMOS 6526, and only the new 6526 differs. Is
there any way to check if it's CMOS or not from the electrical
characteristics on the outputs? Or maybe a microscope attack is needed
to answer this question ;)
White Flame (aka David Holz)
news:36q2tuF...@individual.net...
> Okay, checked it with the 128DCR (new CIA) and it outputs 6526A. The old
C64
> (old CIAs) outputs 6526.
Cool, so it actually works? :) I only have a 6526A-based flat 128 at the
moment, and I don't know if I trust VICE's timing for this idiosyncracy.
> Any chance you expand the test to include CIA #1 as well,
Definitely.
> and maybe test the 1570/71/81 drive CIA?
mmm, probably not. Once I finish it and release the source, somebody else
can do that if they wish. I'm not much of a drive coder.
> Here an additional test for
> presence of a 8520 (different TOD registers) would be quite nice.
Are any of these ever found in stock C64s?
--
White Flame (aka David Holz)
Nicolas Welte
>>Here an additional test for
>>presence of a 8520 (different TOD registers) would be quite nice.
>
>
> Are any of these ever found in stock C64s?
Some people used them as a replacement (with limited compatibility), so
it would be nice to be able to detect them.
Mika Leinonen
>same timing like the old NMOS 6526, and only the new 6526 differs. Is
>there any way to check if it's CMOS or not from the electrical
>characteristics on the outputs? Or maybe a microscope attack is needed
>to answer this question ;)
It should stay very cool if it were CMOS.
Nicolas Welte
IIRC, it produces no noticeable heat at all. But the older version also
never got very warm, so it's hard to tell in this case.