-
Notifications
You must be signed in to change notification settings - Fork 1
Description
Hi Neil,
Now that VCF is over, I'm taking a fresh look at rtedit and microware abasic.
I'm dedicating one of my D2 kits for the implementation as well as a M68MM01A2 micromodule with AC-30 cassette attached.
Kudos for the start at disassembly!
In looking at my original 1970s margin notes, a cassette version (binary from CMD file) from flexdisk archives, and your disassembly, we might consider the following fixes.
My original manual fades on the abasic listings from $0F10 to $0F30. I can't read the s19 record checksums anymore.
From my vintage notes I had written down the databytes as I originally gave you.
Currently in your git: (with checksums I computed before sending you)
S1130F1002973EDE315FD72CD729D72DD751D75230. - readable in manual with checksum $30. Note 97 $3E
S1130F2039DE318DD48D0B240896522603BD10FA78 - very faint in manual, computed checksum from databytes in my notes
S1130F300D398D378D2E812B26088D2B5FBD10FC2E- very faint in manual, computed checksum from databytes in my notes
Looking at abasic.bin (converted from ABASIC.CMD within the FLEX2-01 disk image) - V1.0c compiled for 12K and slight modifications for flex OS
We see the following:
S1130F1002974EDE315FD72CD729D72DD751D75220 <---note $4E instead of $3E. $3E is utilized by 16 bit index reg instructions elsewhere in code
Make more sense to store accumulator in $4E as with other code, but unclear given weird coding practices microware used.
checksum is computed to be consistent with changed value to $4E
S1130F2039DE318DD48D0B240896522603BD10FA78 <--- identical in both manual version and Flex image
S1130F300D398D378D2E812B26088D2B5FBD10BC6E <---note BD $10BC and not BD $10FC. Disassembly suggests $10BC correct. Could be error in my original vintage notes. I no longer have readable cassette tapes of the object code to confirm.
checksum is computed.
so bottom line - we should probably use $10BC as JSR target. Further investigation will be needed for the $97 (STAA direct) $3E vs $4E.
Stan
PS, when I clicked on your youtube link for my vcf talk, the timestamp started just after my interview. Might adjust.
I got raw video files from another attendee who interviewed me w/B-roll footage. I'm editing and will send you a link when done.
screenshot of binary differences either side of $0F10 ($0E10 in screenshot as file offset 0x0100)
