Argh! You know, I was in the middle of writing my counterargument to the thread on "too-long help wanted posts," but halfway through changed my mind and dropped it.
Ummm...well all I can do is try a series of follow-ups and possibilities. And ask for more details. So, I'll do all of that.
What calibration frames are you using? Are you using RAW/DSLR or astrocam FITS? Could you provide a sampling of each type of file (say, 10 or less) so we can toss them into DSS to try to find the place things are going wrong? We could then compare DSS settings if the results come out good.
What does the DSS histogram look like when it loads the autosave back in (which is does immediately after stacking completes) for preview? A think spike of nothingness?
What does the stretch you speak of look like (using some program other than DSS I hope)? Blackness except for some overexposing stars?
At this point, my only real guess is that calibration has gone wrong, and the bulk of your data is being subtracted (clipped) off. For example, darks are too bright, or some other issue that's being had with scaling.