brain wrote:
Testing is much appreciated.
I had a go tonight.
First test was aborted when having programmed a 27c256 and gone to insert into the socket, I realised the socket is for 32 pins. I then reread the lemon64 thread and saw the note on JP4. I didn't want to fiddle with cutting traces or soldering pads, but I found a 29c040 and programmed it up with a file consisting of 32 copies of my 16KB "nb65" cart image.
I then inserted the eeprom into the cart, put the cart into my c64, and powered up.
Unfortunately it didn't work

With the switches as outlined above (and an eeprom in the socket) my C64 basically hangs on boot up. If I disable the eeprom (with either of S2 or S3) it boots into BASIC. But the code on the eeprom doesn't seem to work.
The code in the eeprom starts like this:
;first let the kernal do a normal startup
sei
jsr $fda3 ;initialize CIA I/O
jsr $fd50 ;RAM test, set pointers
jsr $fd15 ;set vectors for KERNAL
jsr $ff5B ;init. VIC
cli ;KERNAL init. finished
jsr $e453 ;set BASIC vectors
jsr $e3bf ;initialize zero page
;set some funky colours
LDA #$04 ;purple
STA $D020 ;border
LDA #$00 ;black
STA $D021 ;background
This code seems to work OK in VICE. I assume if it's starting to run at all, it's not getting as far as the "STA $D020" line, since the border/background colours when the C64 hangs are the normal startup blues (with no text anywhere on the screen - just the 2 blues).
Not sure what could be going on here. I assume that if I'd completely botched the burning of the eeprom, then the C64 wouldn't see the "CBM80" autostart signature and would just do a normal boot? A