On Sun, Feb 03, 2013 at 06:54:06PM +0100, Gerrit Heitsch wrote: > >I've decided to try to build some minimal system based on a 65xx CPU. The > >"best" CPU I have for this purpose seems to be a 8501 from an otherwise > >non-working C16, fortunately it was sockected (I am not sure if it's a > >standard thing being socketed or not). > > Problem is, the 8501 is prone to die easily. So the reason why this > C16 is not working might be that the 8501 is dead. Maybe, however I have two of them, so I believe in my luck. I have even found an older mail here about the heating problem (a not needed resistor on the mask of the CPU or so?), but since I would drive the CPU at a constantly 1MHz clock (afaik in C16/Plus4 it is clocked at 1.76MHz during the TED video blanking periods), I guess I won't fry it if they managed to survive their lifespan in their original place (Commodore 16) ;-) I also have a Commodore Plus4, but I wouldn't like to cannibalize it, as it's a working machine otherwise. As I've read many reports about dead TEDs too, I thought the unworking state of my C16s can be caused by TEDs as well, so I can have to hope to find the CPUs working actually. I would try to test them with the "NOP executing" breadboard config :) however my first problem was even the unknown pin-out of 8501 ... The test circuit has UM6502 currently, so some modification is needed, I guess (I even don't know if 8501 needs two phase clock or not, etc). Anyway I don't think that 8501 is _that_ easy to die, as I was in a "computer camp" in my teenager period having about 40 machines of perfectly working C16s ... :) :) Hungary is somewhat "famous" for having more C16s and Plus4's than the world's average density of these machines :) Actually I got these C16s for free as an "extra" when I bought a Plus/4 some years ago. > >Also what seems to be a magic for me: it's often quoted as an imporant > >feature of 8502 that being capable running on 2MHz, not only 1. However even > >8501 runs at 1.76MHz (during blanking times, when TED does not need to > >access memory?) in Plus4 so 8501 is allowed to run over 1MHz cleanly, it > >seems. Does it mean that I can run 8501 at about 1.7MHz constantly without > >problems, or this is only the effect that the average clock of a 8501 in C16 > >is about 1MHz (slightly more though) and 1.76 is only the peak? > > If you tell TED to disable the screen output (border color displayed > on the whole screen), the CPU will run on 1.7 MHz all the time. The > only exception being the 5 DRAM refresh cycles happening every can > line. Well, yes, just I thought that maybe 8501 is not stable enough to run constantly at 1.7MHz clock. And I was somewhat surprised as the "original" 6510 (and 6502) is specified as an 1MHz clocked CPU, while 8502 is being 2MHz, so I thought 8501 should be 1MHz, but this ~1.7 seems to be strange for me. Maybe is it clocked at an "unofficial, out of specification" clock in C16&Plus4? I am just curious here, this is one reason I would love to see the "official" MOS 8501 specification/datasheet. > >Currently I am "playing" with an UM6502 (made by UMC) but I would be happy > >to use 8501 instead (the integrated i/o port is great for a hobby project > >like mine!), but I would need some information/datasheet first to understand > >the differences between 8501 and 6502 (there are much more information > >available on 6510, that's why I wamt to compare 8501 with that). > > If you really must use a 65xx with integrated I/O, you should use > the 6510/8500. Much easier to get and stable. Also most of them I see. However I wouldn't enjoy to cannibalize my C64s for 6510 and also (afaik I can rembeber, I am not sure) the CPU is not even socketed there by default. > should be able to run on 2 MHz. I have a 6510 stamped with an 'A', > indicating 2 MHz capability. > > Otherwise... How about using a 6502, a 6532 and an EPROM for the > code? The 6532 will give you lots of I/O, timer and 128 Bytes of > RAM. Well, my needs is about what I have ... Unfortuntely it's almost impossible here to get 65xx parts, but I managed to get an UM6502 (this was my original "playground") and now I've just realized that I also have two 8501s in two C16s. The best solution would be get a 65C02 or 65C816. The integrated I/O is just because that it can simplify my situation with my very first 65xx hw project, I would use anyway some more modern CMOS part for the next one. - Gábor Message was sent through the cbm-hackers mailing listReceived on 2013-02-03 19:00:30
Archive generated by hypermail 2.2.0.