Topas software is very good at solving such short axis problems. The advantage is that it will look at all of the peaks you feed it, instead of using just the first twenty or so to generate candidate solutions (the way that ITO and TREOR work).
There is a way with McMaille to fix the cell parameters deduced from a series of h0l, to enter up to 100 peak positions, and to explore exclusively the lacking parameters. This can be done either by Monte Carlo or by grid-search. But if only 3 or 4 peaks for a total of 31 are not h0l and involve k, then the ambiguity is too large IMHO.
Armel
