mismatch histogram *extremely* slow

[antonop]

In the Tools, you can examine mismatch losses by generating graphs and histograms.

When I try to generate a histogram for a mismatch setup (e.g. a pretty normal setup of 12 modules in series and 5 strings in parallel), it is essentially impossible as PVsyst just sits there consuming 50% CPU cycles for as long as I leave it (the longest I tried was 10 minutes, after which I forced close PVsyst).

Does anyone else have difficulty generating mismatch histograms like this?

Granted, I am not running PVsyst 6 on the latest CPU (rather, an Intel dual core 1.86 GHz with 3 GB of RAM), but I also noted this problem years ago in PVsyst 5. I don't think that one would require a super CPU to run such a process.

[André Mermoud]

I never heard about such a problem. This calculation is immediate (fraction of second for 20 random generations), and you can ask to add statistics as much times as desired.