On 03/04/2015 07:47 PM, firstname.lastname@example.org wrote: > On 2015-03-04 18:40, Gerrit Heitsch wrote: > >>> I already run two swinsids in one machine - they do a relatively good >>> job so - in case - there is a direct replacement. Things look worse with >>> e.g. TED. or even VIC. >> >> VIC is not that much of a problem. Despite the 6569 running quite hot it >> doesn't die very often and there are still lots of them available. TED >> is a different story since not as many were made and the earlier ones >> (R1 and R2 with '84 datecode) seemed to die easily. >> >> I'd be more worried about the availability of the 6526 / 8521. > > 8520? No, the 8520 is the CIA for the Amiga. If you own a C64 made at the end of 1986, check the CIAs, with some luck you will find a 8521R0 instead of a 6526 (I own a 250466 board with a 8521R0 on U2 and a 1571 with one ). Later Commodore went back to the '6526' label, but the die inside is still a 8521. Any CIA with a '206A' or '216A' next to the datecode is a 8521 with the digit after the '2' showing the revision. The number after the datecode seems to be another code indicating the process (1=NMOS, 2=HMOS, 3=?, 4=CMOS) and revision of the chip. > TED is critical as we know. I agree that VIC is fairly hard to break (I > recall that was probably the least breaking MOS component in my > practice) but the reason I placed it over CIAs is that the latter should > be easier to achieve full compatibility when reimplementing. And also > require far lower gate count, possibly enabling it to fit on a limited > footprint. The CIAs are not as easy to implement as one would expect. Digitally yes, but then there's the analog side (the way the output drivers behave) that will be hard to duplicate with todays technology. Some programs use that with the CIA for the keyboard. Interested in die shots? http://oms.wmhost.com/misc/ Includes one of the 6526/8521. Gerrit Message was sent through the cbm-hackers mailing listReceived on 2015-03-04 20:00:05
Archive generated by hypermail 2.2.0.