bixio60 wrote:
- better situation will be to emulate Qx0 if ever possible
I saw Peter Graf partecipate to this , at least to support QL-SD, any opinion?
I guess my contribution for QL on MIST is not mainly QL-SD, but debugging the TG68 FPGA core, so it does not just run Atari and Amiga Code, but also QL code. The TG68K.C (which is probably used on the MIST) is a successor of the TG68 and already has the QL bugfixes included.
The original author of the TG68 and TG68K.C, Tobias Gubener, had debugged his core by letting it run in parallel with a real 68K CPU in hardware. Since I do not have such a setup, it was a much harder task to debug the CPU. A misimplemented instruction usually does not cause a problem immediately, but often tens or hundreds instructions later, or never at all.
I have not yet looked at the MIST implementation, but from what I heard, it is a close implementation of the original QL. The Q68 is closer to the Qx0, especially keyboard and highcolor graphics are identical, so a SMQS/E port might be easier for the Q68 than for the MIST.
I could port QDOS Classic from the Q60 to the Q68 myself in one day, so if SMSQ/E is as nicely structured as QDOS Classic, the amount of work should be similar (for someone who is familiar with SMSQ/E, which I am not).