DRD Team

Approximately half the time I start GZdoom 0.9.21 with my project as the main IWAD it crashes. I don't even get as far as the game and I have added no command ine parameters. ie, all I do is double click the GZDoom.exe icon.

This IWAD has started fine with previous GZDooms. Nothing has changed with it since the last version that worked with it.

I can see where it crashes but not why. I'm afraid that without the WAD I won't be able to fix this. The problem is the same in all cases.

Fair enough. It seems to have stopped happening now. I don't know why, or if it will start doing it again but I've been able to start and run GZDoom with the project 30-40 times now without issue.

I noticed that it said "access violation" in the report. At various times I've had a few of those - at different address values - with different programs. So perhaps there is something wrong with my machine. I've used a variety of memory checking programs and had them running for hours, I've checked my directX installation, my video drivers and my soundcard drivers - all of which are things that various places have recommended and all of which have come up clean. I've cleaned my registry, tested and defragged my drives and done all the usual housekeeping tasks - all clean. So apparently there is nothing wrong - or nothing I can find - but I have experienced a few "access violations" so perhaps there is something screwy somewhere. Buggered if I can find it though.

The crash happens in the code searching the WAD directory. This is most certainly not something caused by external circumstances. I am suspecting an uninitialized variable somewhere.

One thing you can do is try to get a listing of the IWAD's directory. Unfortunately I don't know many tools that can do such a thing.

Do you mean something like this?

[edit]removed and zipped the post because it was almost 600KB[/edit]


[edit2]There are a few things listed that aren't actually in the WAD - eg all the SCRIPTXY lumps and ~MAPXYZ lumps (which is how DeePsea names script lumps over SCRIPT99) at 0 length are not there, nor is the d_e e_p lump mentioned at the end.[/edit2]

Looks ok. But that's a *lot* of data, isn't it?