The Ur-Quan Masters Discussion Forum

I moved Beta Corvi in Pkunk space (I'm completely sure that this is the only deviation from vanilla UQM in this build)(via plandata.c, ordering after Y value, no other change to the star except position) and, just when the encounter starts (immediately after the Slylandro comm image loads completely, immediately before they say their first line), the game crashes with this error message:
(http://img82.imageshack.us/img82/7786/noslyerrorjp0.th.jpg)
Always.

Help please!

The location of the planet is used to initialise the random number generator for generating the star system.
The code that generates the Slylandro homeworld merely changes the fourth planet. If with the new random seed there is no fourth planet, you'll get a crash.

Yes of course, but, as I said(it looks like you're really tired, you didn't read my post carefully on Bugzilla (Bug 983) too), there was a fourth planet. The crash occurs just when the encounter with the homeworld initializes.

This is the stderr.txt file generated : stderr.txt (http://ceruopran.googlepages.com/stderr.txt)

Just because there are four planets after the Slylandro home system generation code says "planet 4 is a gas giant and has 1 moon", doesn't mean there was a fourth planet before that, as generated randomly. A half-initialised planet is a problem.
But a closer look at the code suggests that this is not enough to explain the problem, as the number of planets is kept separately, so the code should never look at the 4th planet if there wasn't one before. Maybe the actual problem is with the moon.
Tracking down the exact problem will take some more time than I'm willing to put in it now, but I'm confident that (if your code really is otherwise unchanged) your problem can be found with the specific code for the Slylandro system generation (in gensly.c) overriding the randomly generated system.
You could just use a bit of trial and error to find a location for the star system where the planet generation happens to work out, but that doesn't guarantee that there isn't still some other problem which isn't immediately apparent.

And your screen shot isn't readable (although I have a good idea what it will say).

As for bug 983, my remark "Are this the recordings as you got them from Paul?" was a rhetorical question. Because we asked for the original, you say "OK, these are the raw recordings.", and yet you named them "Near-Raw recording (75% tempo)".

Sorry. The fault is mine.

Oops! It seems that there was no problem with the mod, instead the problem was with my Slylandro voicepack - slyhome000.ogg, which should have played at the first meeting, had another bit rate than 11025 Hz. Strangely, the problem manifested only with the mod, or probably it occurs with any UQM compiled with MinGW... <tries it> Yes! This is it.

Thanks you for helping, anyway.