Map crashing my game upon loading.

  • If you're asking a question make sure to set the thread type to be a question!

R0GAN

L1: Registered
Oct 16, 2015
18
0
Hi all, I've been working on trying to make a more open version of 2fort by opening up the fences and such.
It's all been going fairly smoothly until after one of my compiles, my game crashed when trying to load the map.

I've tried fixing up just about all the errors I can with the interlopers.net diagnostic tool and the hammer itself finds no errors but it still continues to crash my game every launch.

I'm very new to mapping and would really be thankful for any help!

In case it's of interest, there's a copy of my vmf file here. I'll give 2 mann co. keys to anyone kind enough to help optimize it so it compiles and runs smoothly: www.mediafire.com/download/sgt3q4qh780ahed/sdk_ctf_2fort1%283%29.vmf
My compile log is as follows:

** Executing...
** Command: "D:\Program files\SteamApps\common\Team Fortress 2\bin\vbsp.exe"
** Parameters: -game "D:\Program files\SteamApps\common\Team Fortress 2\tf" "C:\Users\Matt\Desktop\sdk_ctf_2fort1.vmf"

Valve Software - vbsp.exe (Oct 14 2015)
8 threads
materialPath: D:\Program files\SteamApps\common\Team Fortress 2\tf\materials
Loading C:\Users\Matt\Desktop\sdk_ctf_2fort1.vmf
ConVarRef mat_reduceparticles doesn't point to an existing ConVar
Patching WVT material: maps/sdk_ctf_2fort1/nature/blendgroundtograss001_wvt_patch
Patching WVT material: maps/sdk_ctf_2fort1/nature/blendgroundtogravel001_wvt_patch
Patching WVT material: maps/sdk_ctf_2fort1/nature/blendgroundtograss003_nodetail_wvt_patch
fixing up env_cubemap materials on brush sides...
ProcessBlock_Thread: 0...1...2...3...4...5...6...7...8...9...10 (1)
ProcessBlock_Thread: 0...1...2...3...4...5...6...7...8...9...10 (1)
Processing areas...done (0)
Building Faces...done (0)
Chop Details...done (0)
Find Visible Detail Sides...
Merged 1160 detail faces...done (0)
Merging details...done (1)
FixTjuncs...
PruneNodes...
WriteBSP...
done (1)
writing C:\Users\Matt\Desktop\sdk_ctf_2fort1.prt...Building visibility clusters...
done (0)
*** Error: Skybox vtf files for skybox/sky_night_01 weren't compiled with the same size texture and/or same flags!
Can't load skybox file skybox/sky_night_01 to build the default cubemap!
*** Error: Skybox vtf files for skybox/sky_night_01 weren't compiled with the same size texture and/or same flags!
Can't load skybox file skybox/sky_night_01 to build the default cubemap!
Finding displacement neighbors...

** Executing...
** Command: "D:\Program files\SteamApps\common\Team Fortress 2\bin\vvis.exe"
** Parameters: -game "D:\Program files\SteamApps\common\Team Fortress 2\tf" "C:\Users\Matt\Desktop\sdk_ctf_2fort1"

Valve Software - vvis.exe (Oct 14 2015)
8 threads
reading c:\users\matt\desktop\sdk_ctf_2fort1.bsp
reading c:\users\matt\desktop\sdk_ctf_2fort1.prt
2489 portalclusters
6482 numportals
BasePortalVis: 0...1...2...3...4...5...6...7...8...9...10 (1)
PortalFlow: 0...1...2...3...4...5...6...7...8...9...10 (319)
Optimized: 5023 visible clusters (0.90%)
Total clusters visible: 560559
Average clusters visible: 225
Building PAS...
Average clusters audible: 815
visdatasize:674276 compressed from 1553136
writing c:\users\matt\desktop\sdk_ctf_2fort1.bsp
5 minutes, 21 seconds elapsed

** Executing...
** Command: "D:\Program files\SteamApps\common\Team Fortress 2\bin\vrad.exe"
** Parameters: -game "D:\Program files\SteamApps\common\Team Fortress 2\tf" "C:\Users\Matt\Desktop\sdk_ctf_2fort1"

Valve Software - vrad.exe SSE (Oct 14 2015)

Valve Radiosity Simulator
8 threads
[Reading texlights from 'lights.rad']
unknown light specifier type - lights

[56 texlights parsed from 'lights.rad']

Loading c:\users\matt\desktop\sdk_ctf_2fort1.bsp
Patch Sample Radius Clamped!
Patch Sample Radius Clamped!
Setting up ray-trace acceleration structure... Done (9.10 seconds)
16283 faces
11 degenerate faces
1997893 square feet [287696640.00 square inches]
232 Displacements
169872 Square Feet [24461698.00 Square Inches]
16272 patches before subdivision
254172 patches after subdivision

** Executing...
** Command: Copy File
** Parameters: "C:\Users\Matt\Desktop\sdk_ctf_2fort1.bsp" "D:\Program files\SteamApps\common\Team Fortress 2\tf\maps\sdk_ctf_2fort1.bsp"


** Executing...
** Command: "D:\Program files\SteamApps\common\Team Fortress 2\hl2.exe"
** Parameters: -game "D:\Program files\SteamApps\common\Team Fortress 2\tf" +map "sdk_ctf_2fort1" -steam
 
Last edited:

MoonFox

L10: Glamorous Member
Mar 17, 2015
739
74
you may want to check the game's start up files, it could be the game, not the editor
 

R0GAN

L1: Registered
Oct 16, 2015
18
0
you may want to check the game's start up files, it could be the game, not the editor
I really have no idea what you mean by that. My game currently runs other maps, including a previous version of my edited map, fine if that's relevant.
 

MoonFox

L10: Glamorous Member
Mar 17, 2015
739
74
it still wont hurt to check integrity of game cache, trust me: do it once a month always
 

tyler

aa
Sep 11, 2013
5,102
4,621
coltnweston posts are kind of generally confusing and in many cases are better off outright ignored. He means to verify the game cache, which yes, might actually do something, but probably it won't. Good thing to check first though.

Have you changed any of the logic? What edits have you made since your last working compile? I don't have time to look at your VMF at the moment unfortunately, but usually a map crashing your game is due to bad logic.
 
Last edited:

R0GAN

L1: Registered
Oct 16, 2015
18
0
I just verified my game cache and it still crashes sadly.
Thanks for letting me know it's likely due to logic problems. I'm not sure if I've changed the logic since my last working version. In the last working version, I think I had the briefcases deleted and the outputs regarding the intel has been deleted.
 

MoonFox

L10: Glamorous Member
Mar 17, 2015
739
74
derp_exe... well look at my avatar.. it explains exactly what I know. though in general, I try to help in anyway possible, and last time hammer failed to run a game after compile, was due to a bug wherein the loader tries to process, and it just cant run correctly. a typical fix is to reference your logics and check your sdk cache, but the alternate way... well to copy paste everything into a new test and see if that fixes it.
 
S

saph

I didn't see any errors in the compile log, but check your entity setups. One thing might be to not have a logic_auto kill the color_correction. This isn't causing the crash though. A few displacements are a power of four, and this leads to issues often. Your compile goes a lot further than mine does, but try to reduce the number of numportals through optimization. 4,000 seems to be the cap.
 

Zed

Certified Most Crunk™
aa
Aug 7, 2014
1,241
1,025
If you're bad at helping, why are you helping?
 

R0GAN

L1: Registered
Oct 16, 2015
18
0
I think I found the issue, the maximum amount of clusters is 65,536 (two to the sixteenth power I believe). You have:
How do I reduce the number of clusters? It seems I am over the max umber by a very long way but I really haven't added much to the map. I don't really understand the concept of clusters.
 
S

saph

A cluster is a group of visleafs, don't get this confused with the Hammer visgroups. General optimization should be able to reduce the number of clusters drastically. Things such as func_details, hint-skip brushes, occluders, and areaportals are standard ways to manually split and manipulate visleafs. If basic optimization is not present, especially on a map like 2fort with lot of different angled geometry, the number of visleafs (and yes, clusters) will shoot through the roof.
 

MoonFox

L10: Glamorous Member
Mar 17, 2015
739
74
If you're bad at helping, why are you helping?
I don't know what the fuck I do... I eat, I sleep, and I make maps... its what I do.
 

ics

http://ics-base.net
aa
Jun 17, 2010
841
540
I think I found the issue, the maximum amount of clusters is 65,536 (two to the sixteenth power I believe). You have:

This is not the cause. That number is for visible clusters shown at the same time, not total. On my 2 maps, I have total clusters visible 864506 and 728447 and they never crash.

Fault lies somewhere else.
 

R0GAN

L1: Registered
Oct 16, 2015
18
0
Ok so I've messed around with making loads of stuff func_detail and now the map no longer crashes on loading, but I get a massive problem when it's started up. I get 'error: vis decompression overrun' spamming up the console and parts of the map spazz out, going invisible all the time. On the bright side at least it's not crashing any more, so there's that xD

From some searching online, I don't really see any fixes that have helped. I've been trying trying to figure these issues out all day and I really feel like I'm getting nowhere on my own sadly. Anyone know what I can do about it?

My small reward of 2 mann co keys for anyone kind enough to help optimize my vmf file to make it compile into a stable and working map still stands. ->www.mediafire.com/download/sgt3q4qh780ahed/sdk_ctf_2fort1%283%29.vmf
 
S

saph

6482 numportals
You have roughly 2,000 more than what you should have. I'd usually say that you should have 4,000 numportals or less, otherwise, the map begins to lag due to too many numportals. That will cause:
'error: vis decompression overrun'

If you want, I can try to optimize it tonight. (Is that the updated VMF that compiles in your most recent post?)
 

R0GAN

L1: Registered
Oct 16, 2015
18
0
Yes, I've updated the links in all my posts in this thread to my latest vmf file.
That would be amazing if you could get it optimized. I must say, you guys are really helpful!
 

ics

http://ics-base.net
aa
Jun 17, 2010
841
540
You have roughly 2,000 more than what you should have. I'd usually say that you should have 4,000 numportals or less, otherwise, the map begins to lag due to too many numportals.

Numportals are not the cause for his problems, unless the map is very open at all times. You can easily have over 6000 numportals within a map. If the map is big, it's even bound to have lots of those. 6442 and 5353 numportals are in my maps and they don't lag. Infact the framerate is just fine in them.
 
S

saph

Numportals are not the cause for his problems, unless the map is very open at all times. You can easily have over 6000 numportals within a map. If the map is big, it's even bound to have lots of those. 6442 and 5353 numportals are in my maps and they don't lag. Infact the framerate is just fine in them.
Sorry, yes. A high number of numportals doesn't mean a map will lag. You want to reduce them whenever you can though. 5Curve runs fine for me, but lower-end PCs might have a bit of trouble. Really, natural barriers do help a lot when optimizing a map. I've actually seen up to 28,000 or so numportals, but 4,000 is good for Counter-Strike:Global Offensive. 6,000 is a good number for Team Fortress 2.

Really, you want to limit the number of portals because if you have one portal seeing to many other numportals, it'l have to render more of them instead of just one other numportal. The problem with 2fort is that it's middle area is very open with some structures and objects around it.