In addition to my previous speed bump with stim_times, I'm beginning to wonder if my experiment design is such that I can't use the 'stim_times' option at all and will have to stick to the older method of generating 1D files form log files.
This is completely unrelated to the previous question I had about the response type, 'rtype'.

The experiment consisted of 4 runs. Each run had one of two stimulus types (conditions) presented to the subject in addition to the other stimuli (delay, response etc.). So 4 runs -> two runs of each stimulus type, randomly ordered.
To use stim_times, I'd have to generate timing files per stimulus type with one row for each run.
Since the program calculates the global timings from the timings in previous runs, is there a way to specify a 'time-offset' for a run in which the particular stimulus type was not presented ?

I hope I've been able to explain my question clearly enough. Thanks for your help.

Mithun