Can't discount sensors and mapping on the fade issue.
These mysterious 2S fade issues, and I've experienced them more with poo than doo or cat, have become more common since we have timing and fuel being constantly adjusted and dictated by a program with sensors reading pipe temp, air temp, water temp, calculated air density, throttle position, crank position, etc and poos fuel type settings. Always seems like a takes some build experience to get these maps firmed up, if ever (CFI4 was never adequate due to design, CFI2 didn't really get done till 13, with different injectors than 11, doo etec 8 took a year, etec 6 took two years, SDI was never really resolved, cat ctec took a year, etc.) An out of calibration or sporadically failing sensor causes havoc.
Carbed sleds without sliding timing curves just didn't experience it, WHEN you had it jetted for the day and location. DPM was probably the best of both worlds for most joes and sallys, even with an issue from large elevation changes (under enleanment), but it was very predictable. The doo 700 and 800 DPM motors were probably the most consistent with the least amount of tinkering in the carb days. TRA clutching is another issue...