Yes -- short thread on it here. scarybeasts originally flagged it as troublesome in beebjit. He didn't fully investigate it but noted it could be defeated just by immediately acknowledging ACIA TX writes. My latest B-Em tape fork used to load PBS, but I replaced its ACIA with beebjit's so I could get Castle Quest to load, and now it fails. Conclusion was that you need to model accurate TDRE delays for baud rates other than 1200. (Or cheat and set them to zero).Is the reference to Pro Boxing Simulator a kind of tape protection system?(Currently, everything in B-Em-TOH looks embarrassingly like 1200 baud, which is why Pro Boxing Simulator pukes.)
For PBS, no: It keeps the serial ULA divider at /16, which is what MOS uses for 1200 baud, just switching the ACIA's divider from 64 to 1; so this is an ACIA problem.On the 1200 baud, is the issue that the serial ULA is not emulated?
Broadly, though, the serial ULA divider is not emulated, and it needs to be. Disregarding TDRE delays, I don't think there is anything stopping you reading and writing cassettes at e.g. 75 baud (that's /256 & /64), and frankly somebody has to try this purely for the novelty value. I will maybe write a test for this eventually. (Unless somebody else would like to do it. There is code in the AUG for custom tape reading and writing routines).
Statistics: Posted by Diminished — Wed Dec 04, 2024 6:52 am