I find all that strange for me using "side effect" of a chip such as illegal opcode is a bad practice... and makes bad software... in Commodore software is there any use of these illegal opcode ? Le 15/11/2017 à 15:06, Gerrit Heitsch a écrit : > On 11/15/2017 02:36 PM, smf wrote: >>> Does this still apply to chips made after CSG/MOS went out of business? >>> >> Was anyone still second sourcing 6502 by then? I was under the >> impression that everyone had switched to CMOS designs. >> >>> Also, I have doubts about the later 6502 still being NMOS. >>> >> It's possible. The 6xxx CSG/MOS chips are supposed to be NMOS, 7xxx >> HMOS-I & 8xxx HMOS-II, but they have been known to lie (I think the >> later 6526 were HMOS-II). > > They were... And starting at the end of 1986 on MOS/CSG chips you > could see it from the number after the Datecode. If it starts with a > '2' it's HMOS-II, if it starts with a '1', it's NMOS. Internally, the > CIAs had the number '8521' on the die. The other numbers indicate the > revision. > > > >> Conventiently that covers the c64/plus4/c128, so as long as the >> undocumented opcodes are stable across all three then they are safe >> to use no matter what process the chip was made with. > > Did someone ever test that? > > >> I doubt MOS/CSG made a huge number of 6502's though, they normally >> had better uses for their fab. Although they had time to produce the >> HMOS-II 8501 in 1988 (maybe someone at MOS/CSG loved his plus4, but >> the CPU broke). > > I have 8501 with a datecodes from 1989 and 1990. Also, they did a new > revision of the 8501 in 1986 since I have a '8501R4' with datecode > '4986'. > > Also, I have a 1541-II with a CSG 6502AD with datecode 0791. Process > and revision indicator is '1D'. So Commodore did make the 6502 for a > long time in NMOS. > > > Gerrit > > > Message was sent through the cbm-hackers mailing list Message was sent through the cbm-hackers mailing listReceived on 2017-11-15 15:01:43
Archive generated by hypermail 2.2.0.