JinCuk... you're saying page order is completely random for you... unless you only use one thread. Lol, well, at least you've discovered a workaround.
Hmm... Are they totally random (eg, page 1 goes to page 100) or only somewhat random (page 54 goes to page 52)? Thing is that at the point where multithreading is working, page numbers aren't really able to go out of order. My feeling is that maybe when that step finishes, and all the pages are in one folder, and my script asks Windows to list the contents of that folder.... windows doesn't list it in alphabetical (numerical) order and the pages get created in whatever way they're listed.
On my machine and seemingly most other people's it's always alphabetical, but maybe that can change. Could you do me a favor and right before you do "Press RETURN to close...", you go into temp/result and take a look at the files there? What order are they in? Also, if you take a file, eg, "0025" and rename it to "0025.png" and open it, is it a picture of page 25?
If it's just a problem of ordering, I should be able to fix it easily. I thought i could get away with a more elegant method, but guess not.
jakeluck... thanks for the bugreport. Sounds like it's not so much a bug as just windows running out of ram. What puzzles me is why sleep crashes and nothing else. However, sleep.exe doesn't do much and it doesn't really matter. Let me know if this happens often/repeatably or with plenty of free ram.