|
ok, 2 options here:
first: you didn't send us the full compile log
second: i found a part of your problem...
read the compile log, and found it ending like this:
Loading c:\users\fts\desktop\jump_torii.bsp
Setting up ray-trace acceleration structure... Done (0.41 seconds)
5770 faces
30135516 square feet [4339514368.00 square inches]
0 Displacements
0 Square Feet [0.00 Square Inches]
5770 patches before subdivision
343388 patches after subdivision
sun extent from map=0.000000
16 direct lights
BuildFacelights: 0...1...2...3...4...5...6...7...8...9...10 (19)
BuildVisLeafs: 0...1...2...3...4...
** Executing...
** Command: Copy File
** Parameters: "C:\Users\fts\Desktop\jump_torii.bsp" "C:\Program Files (x86)\Steam\steamapps\common\Team Fortress 2\tf\maps\jump_torii.bsp"
to put it bluntly:
it starts compiling rad *Setting up ray-trace acceleration structure... Done (0.41 seconds)*
builds face lights
and just abruptly stops on VisLeafs...
find out why it stops the compile there, and you have found the problem. . .
small explanations:
what are vis leafs? Vis leafs are boxes without anything to break line of sight on your map. those boxes together make up the visual base structure of your map, and enable *vis* on the second compile step to compile visibility in your map. So basically from which room you can look into which other room. So when ingame, it can use those informations to draw only the things you can actually see, so you don't have to render the whole map all the time.
rad (lighting calculation) also uses those informations, to see which lights reach what solids, so lights don't shine through walls and stuff (id guess in reality its a bit more complicated, but as far as i know, thats pretty much what that step does)
so, what can lead to our compile suddenly stopping there?
a lot of things can lead to that:
a) your processor is overheating (had that happen before... compiles stopped at totally random times, and i had no clue why)
b) your light information is too large for rad to compile (check windows performance monitor if rad hits over 1.2 gig of memory usage, thats about when it gets critical without a workaround)
c) something about your vis leafs is weird / broken
C is the option i would place my bet on, since when you compile with a leak, vis is pretty much ignored, breaking water rendering on the way... and your lighting compiles with a leak.
Try setting vis to "fast", or to "no", and see what happens with that. Does the problem still persist? If it doesn't we just got a step closer to finding your problem, if it does... well, that at least rules out something xD
another option to go for would be a)... even if it sounds stupid at first, because:
305320 patches after subdivision
16 direct lights
0...1...2...3...4...5...6...7...8...9...100...1...2...3..
this is how the first log ended, in midst of some compile step, and it was a different one than the second step!
So mb your hardware stops the compile from working properly (as in not enough memory accessable, or cpu overheating, or something like that)
id go for one of those 2 things, but you really don't give us much to work with here xD
for the second option to be ruled out:
can you post your appropriate specs? as in cpu model, ammount of physical memory (ram + virtual memory), and cpu temperature when compiling?
|
