Just been playing with nistune version 0.9110b.
I am working on the K11 micra again with the grid board. I can now offset it correctly, however when I have done so consult wont work correctly and shows totally incorrect (all pretty much maxed out) figures. If I remove the offset this works again, but then obviously I cannot tune.
Matt any advice?
Offset Issues
Moderator: Matt
Sounds like the CONSULT table which NIStune uses in 'tuner mode' is being adjusted to the wrong location once you change the OFFSET value
I checked the code and the OFFSET is being used to help with location of the table
For example default K11 values are OFFSET = 0x8000 and CONSULT = 0x0F80. This means that table in physical memory sits at 0x8F80 but adjusting the offset will screw up the consult location
You will need to override the offset by adding an extra parameter in the address file
CONSULT=&Hxxxx
Where xxxx is 0x8F80 - OFFSET value to get the consult table in the correct location again
I checked the code and the OFFSET is being used to help with location of the table
For example default K11 values are OFFSET = 0x8000 and CONSULT = 0x0F80. This means that table in physical memory sits at 0x8F80 but adjusting the offset will screw up the consult location
You will need to override the offset by adding an extra parameter in the address file
CONSULT=&Hxxxx
Where xxxx is 0x8F80 - OFFSET value to get the consult table in the correct location again
thats a difficult one...
the original ROM Editor address entries start from the offset. So a 32K ROM would have addresses 0000h-7FFFh in the address file (with an imaginary offset of 8000h where the maps really lived)
so the CONSULT parameter was done the same.
changing the OFFSET however is having two effects
- Romulator offset where the image gets loaded in the emulator
- ECU offset where the image expects to be run
Have to think about this one and see what the best solution is
the original ROM Editor address entries start from the offset. So a 32K ROM would have addresses 0000h-7FFFh in the address file (with an imaginary offset of 8000h where the maps really lived)
so the CONSULT parameter was done the same.
changing the OFFSET however is having two effects
- Romulator offset where the image gets loaded in the emulator
- ECU offset where the image expects to be run
Have to think about this one and see what the best solution is