Vis taking a long time to finish Created 16 years ago2008-09-23 01:01:56 UTC by Kill_the_Bug Kill_the_Bug

Created 16 years ago2008-09-23 01:01:56 UTC by Kill_the_Bug Kill_the_Bug

Posted 16 years ago2008-09-23 01:01:56 UTC Post #255941
Ok not used to source compiling yet but this one's been 100% CPU usuage and 4 days - big TF2 Map.

Have launched task manager and stripped computer done to 26 essential services to free up RAM here are some details:

Totals:
Handles-8284
Threads-406
Processes-26

Physical Memory(k)
Total - 2620912
Available - 1549796
System Cache - 1070064

Commit Charge(k)
Total - 964640
Limit - 4555628
Peak - 1224032

Kernal Memory (K)
Total - 85096
Paged - 653388
Nonpaged - 19708

Processes - 26 CPU usage 100% Commit Charge 941M/4448M

Ok what's all the above mean? LOL

lastly AMD(Athlon)64 processor
300+
2.01GHz, 2.50GB Ram
Physical Address Extension

When I last saw it (hammer shows black screen - can no longer see the compile window) I was at stage 4 of the 2nd Vis line - yesterday about 4 pm - it now 1 am.

Any ideas how I can speed it up? I have lots opf hin brushes in it.

How much longer do you think I'll have to wait?

As always, thanks for your help.
Posted 16 years ago2008-09-23 01:21:05 UTC Post #255943
Not sure about Source, but 4 days is a tad much, even for a big map. If nothing's wrong with it, however, you can always ask someone with more processing power to compile it for you.
Daubster DaubsterVault Dweller
Posted 16 years ago2008-09-23 01:40:41 UTC Post #255945
AMD(Athlon)64 processor 300+
I hope you mean 3000+...
4 days of compile = bad. You're doing something wrong. Try using an external compile program, or restarting your compile.
Penguinboy PenguinboyHaha, I died again!
Posted 16 years ago2008-09-23 04:06:08 UTC Post #255946
If your compiles takes more than 2 hours, you're doing things wrong. Source VIS is quite fast compared to the goldsource one. A detailed prision doesn't take more than 3 minutes to full compile.
Posted 16 years ago2008-09-23 08:55:30 UTC Post #255952
suggestions for that external compile program?

for the ompile - I hit F9 and picked normal for all - should I do something else?

maybe vis got stuck?
Posted 16 years ago2008-09-23 08:56:48 UTC Post #255953
will put the vmf into problem maps area to see if anyone has the same problem
Posted 16 years ago2008-09-23 12:00:06 UTC Post #255956
Question - anyone else having a problem loading maps into the problem area?

Tried 5 times now and it won't work.
Posted 16 years ago2008-09-23 13:08:02 UTC Post #255958
Just read a post by Madcow in different area - he said that engine doesn't like wide open areas and that fog helps this.

My map is canyon style, there's 4 - 5 large combat zones all outdoors - could this add into the compile time?

Oh - and lots of lights(on normal) - torches lighting the map - it's a twighlight dusk / early night time map - shadows and dark areas make it great for spies / snipers - LOL
Posted 16 years ago2008-09-23 13:11:02 UTC Post #255961
Oh my fucking God!

QUADRIPLE POST!?

Use the fucking edit button man... :rly:
monster_urby monster_urbyGoldsourcerer
Posted 16 years ago2008-09-23 14:44:36 UTC Post #255972
I tried - it wasn't working for me :gak: - it think it might be time delayed

two of them close together but the other 2 are at least an hour apart
Posted 16 years ago2008-09-23 19:02:54 UTC Post #255992
ok after 5 days I killed VIS - here's my compile report. Any suggestions:
** Executing...
** Command: "c:\program files\steam\steamapps\kill_the_bug\sourcesdk\bin\orangebox\bin\vbsp.exe"
** Parameters: -game "c:\program files\steam\steamapps\kill_the_bug\team fortress 2\tf" "C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a"

Valve Software - vbsp.exe (Mar 11 2008)
1 threads
materialPath: c:\program files\steam\steamapps\kill_the_bug\team fortress 2\tf\materials
Loading C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a.vmf
Patching WVT material: maps/pl-bugscanyon-beta-a/nature/blendrockdirt008c_wvt_patch
Patching WVT material: maps/pl-bugscanyon-beta-a/nature/blendrockgravel001a_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 (1)
Chop Details...done (0)
Find Visible Detail Sides...done (0)
Merging details...done (0)
FixTjuncs...
PruneNodes...
WriteBSP...
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
NODRAW on terrain surface!
done (0)
writing C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a.prt...Building visibility clusters...
done (0)
Creating default LDR cubemaps for env_cubemap using skybox materials:
skybox/sky_day01_01*.vmt
! Run buildcubemaps in the engine to get the correct cube maps.
Creating default HDR cubemaps for env_cubemap using skybox materials:
skybox/sky_day01_01*.vmt
! Run buildcubemaps in the engine to get the correct cube maps.
Finding displacement neighbors...
Found a displacement edge abutting multiple other edges.
Warning: overflowed 2 displacement corner-neighbor lists.Finding lightmap sample positions...
Displacement Alpha : 0...1...2...3...4...5...6...7...8...9...10
Building Physics collision data...
done (1) (425972 bytes)
Placing detail props : 0...1...2...3...4...5...6...7...8...9...10
Compacting texture/material tables...
Reduced 5291 texinfos to 2681
Reduced 43 texdatas to 34 (1260 bytes to 782)
Writing C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a.bsp
19 seconds elapsed

** Executing...
** Command: "c:\program files\steam\steamapps\kill_the_bug\sourcesdk\bin\orangebox\bin\vvis.exe"
** Parameters: -game "c:\program files\steam\steamapps\kill_the_bug\team fortress 2\tf" "C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a"

Valve Software - vvis.exe (Mar 11 2008)
1 threads
reading c:\documents and settings\john\desktop\maps\pl-bugscanyon-beta-a.bsp
reading c:\documents and settings\john\desktop\maps\pl-bugscanyon-beta-a.prt
3222 portalclusters
12500 numportals
BasePortalVis: 0...1...2...3...4...5...6...7...8...9...10 (126)
PortalFlow: 0...1...2...3...4...5..
** Executing...
** Command: "c:\program files\steam\steamapps\kill_the_bug\sourcesdk\bin\orangebox\bin\vrad.exe"
** Parameters: -game "c:\program files\steam\steamapps\kill_the_bug\team fortress 2\tf" "C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a"

Valve Software - vrad.exe SSE (Mar 11 2008)
Valve Radiosity Simulator
1 threads
[Reading texlights from 'lights.rad']
[34 texlights parsed from 'lights.rad']

Loading c:\documents and settings\john\desktop\maps\pl-bugscanyon-beta-a.bsp
No vis information, direct lighting only.
Setting up ray-trace acceleration structure... Done (5.69 seconds)
7135 faces
1 degenerate faces
1503323 square feet [216478560.00 square inches]
179 Displacements
226606 Square Feet [32631388.00 Square Inches]
174 direct lights
BuildFacelights: 0...1...2...3...4...5...6...7...8...9...10 (188)
Build Patch/Sample Hash Table(s).....Done<0.0896 sec>
FinalLightFace: 0...1...2...3...4...5...6...7...8...9...10 (35)
FinalLightFace Done
Computing detail prop lighting : 0...1...2...3...4...5...6...7...8...9...10
0 of 0 (0% of) surface lights went in leaf ambient cubes.
ThreadComputeLeafAmbient: 0...1...2...3...4...5...6...7...8...9...10 (139)
Writing leaf ambient...done
Ready to Finish

Object names Objects/Maxobjs Memory / Maxmem Fullness
---------- ------------- ------------- ------
models 206/1024 9888/49152 (20.1%)
brushes 1155/8192 13860/98304 (14.1%)
brushsides 8779/65536 70232/524288 (13.4%)
planes 8590/65536 171800/1310720 (13.1%)
vertexes 12017/65536 144204/786432 (18.3%)
nodes 6617/65536 211744/2097152 (10.1%)
texinfos 2681/12288 193032/884736 (21.8%)
texdata 34/2048 1088/65536 ( 1.7%)
dispinfos 179/0 31504/0 ( 0.0%)
disp_verts 14275/0 285500/0 ( 0.0%)
disp_tris 22528/0 45056/0 ( 0.0%)
disp_lmsamples 519056/0 519056/0 ( 0.0%)
faces 7135/65536 399560/3670016 (10.9%)
hdr faces 0/65536 0/3670016 ( 0.0%)
origfaces 4390/65536 245840/3670016 ( 6.7%)
leaves 6824/65536 218368/2097152 (10.4%)
leaffaces 9142/65536 18284/131072 (13.9%)
leafbrushes 3723/65536 7446/131072 ( 5.7%)
areas 2/256 16/2048 ( 0.8%)
surfedges 48973/512000 195892/2048000 ( 9.6%)
edges 27580/256000 110320/1024000 (10.8%)
LDR worldlights 174/8192 15312/720896 ( 2.1%)
HDR worldlights 0/8192 0/720896 ( 0.0%)
leafwaterdata 6/32768 72/393216 ( 0.0%)
waterstrips 588/32768 5880/327680 ( 1.8%)
waterverts 0/65536 0/786432 ( 0.0%)
waterindices 9096/65536 18192/131072 (13.9%)
cubemapsamples 0/1024 0/16384 ( 0.0%)
overlays 0/512 0/180224 ( 0.0%)
LDR lightdata [variable] 7675588/0 ( 0.0%)
HDR lightdata [variable] 0/0 ( 0.0%)
visdata [variable] 0/16777216 ( 0.0%)
entdata [variable] 347936/393216 (88.5%) VERY FULL!
LDR ambient table 6824/65536 27296/262144 (10.4%)
HDR ambient table 6824/65536 27296/262144 (10.4%)
LDR leaf ambient 13626/65536 381528/1835008 (20.8%)
HDR leaf ambient 6824/65536 191072/1835008 (10.4%)
occluders 0/0 0/0 ( 0.0%)
occluder polygons 0/0 0/0 ( 0.0%)
occluder vert ind 0/0 0/0 ( 0.0%)
detail props [variable] 1/244 ( 0.4%)
dtl prp lght [variable] 1/14 ( 7.1%)
HDR dtl prp lght [variable] 1/4 (25.0%)
static props [variable] 1/7892 ( 0.0%)
pakfile [variable] 214986/0 ( 0.0%)
physics [variable] 425972/4194304 (10.2%)
physics terrain [variable] 24679/1048576 ( 2.4%)

Level flags = 0

Total triangle count: 17765
Writing c:\documents and settings\john\desktop\maps\pl-bugscanyon-beta-a.bsp
6 minutes, 9 seconds elapsed

** Executing...
** Command: Copy File
** Parameters: "C:\Documents and Settings\John\Desktop\maps\pl-bugscanyon-beta-a.bsp" "c:\program files\steam\steamapps\kill_the_bug\team fortress 2\tf\maps\pl-bugscanyon-beta-a.bsp"
Posted 16 years ago2008-09-24 00:50:42 UTC Post #256002
Thanks for all the help so far- Daub I'm taking your advice...

here's a link to the vmf file - can someone try to do a compile on a good machine and see if they have any issues?

http://jecphotography.com/gamestuff/Pl_bugscanyon.htm
Posted 16 years ago2008-09-24 09:21:56 UTC Post #256010
"compiled" it with a quad core. I killed vis after the 10% VIS, it's an unoptimized mess of map, wtf you put a hollowed sky box around the map!!!

This map would take 1 hour to compile on my machine, but if things done correctly it wouldn't take more than 4 minutes. Go learn optimization and how Source works, because you can't make maps like this.
Posted 16 years ago2008-09-24 11:05:27 UTC Post #256014
Hey thanks a lot Spike I really appreciate it! I'm new to the whole Source World - this is my 1st real big map - I mostly do day of defeat and cs 1.6 stuff.

Will learn the optim stuff.
Posted 16 years ago2008-09-24 11:33:27 UTC Post #256016
use sv_cheats1 1 and mat_wireframe 1 on official maps to see how they're optimized, it will help you a lot.
Posted 16 years ago2008-09-24 12:01:06 UTC Post #256020
Thanks again - I was wondering where to start

I think the skybox route I did was the old fashioned way of doing it- is this right? Would I be better off doing a 3D skybox thing? I was looking at the tutorial on the site here - would that cut down the compile tim too?

Do you know anything about area_portals? I've heard they're also good but hear they don't work with displacements - you've seen the map - I tried to use them in the tunnels and caves but am getting leak errors when I do so I just went with hint brushes.
Posted 16 years ago2008-09-24 12:22:08 UTC Post #256022
I heven't used areaportals yet. And for the skybox thing... you have to put sky brushes on the limits of your map, dont round it with a sky box.

Use func_detail for complex brushes, don't make large areas, and put nodraw brushes behind displacements, because displacements doesn't block VIS
Posted 16 years ago2008-09-24 14:07:01 UTC Post #256024
OH WOW SPIKE - HUGE TIP!!!! my map contains a ton of diplacements none of which has anything behind it - this tip alone will probably save me a huge amount of compile times - will get to work blocking VIS LOL

For the Rocks ontop of the map & other models - make them func_details? I think I saw somewhere that VIS will carve a map into a lot of leaves around a pillar for example - thus lagging compile time and gameplay time.

Will models/props effect vis too if they touch the floor, walls etc? or should I try to leave them 1 unit off the floor? Haven't found this out yet but am reading up here:

http://developer.valvesoftware.com/wiki/Controlling_Geometry_Visibility_and_Compile_Times
Posted 16 years ago2008-09-24 14:59:12 UTC Post #256033
no entities will carve into the world when rendered in-game.

This includes un-tied entities such as the Clip brush
Archie ArchieGoodbye Moonmen
Posted 16 years ago2008-09-25 17:52:28 UTC Post #256077
I'm putting this into here instead of another question addon to penguin boy on the 3D Sky boxes tutorial.

Displacements make up a lot of the exterior walls of my map (currently fixing this thanks to Spike's mesage above). If I wall them up with the nodraw texture behind them can I put another brush ontop of it then attach the skyobx texture to it so that I can then do the 3d Skybox thing? Will the 3D skybox work?

Sorry to be such a pain but one last thing on the skybox thing - if you had a map that was shaped like a donut (or other holes in your map) and put in the skybox in the interior and outer cirlces will the skybox still work properly or do you want me to copy/paste this question onto the tutorial?

Lastly - similar to the dounut - let's say your map from above looked like the letter "E" but with the top part of the E went to the left - for the top skybox section do you mold it to fit or can you just make one big square to fit and the program will blend it in the game automatically?
Posted 16 years ago2008-09-25 18:15:47 UTC Post #256116
I don't really know what you're asking. Can you clarify a bit?
Penguinboy PenguinboyHaha, I died again!
Posted 16 years ago2008-09-26 01:40:20 UTC Post #256153
in your 3d skybox tut u use a 4x4 floor with walls - textured

can u put a skybox on a nodraw textured wall

can u manipulate the skybos to any shape u like - ie your map is in a u shape therefore your skybox will have to be u shapd or o they all have to be like a 4x4 box?

thanks!
Posted 16 years ago2008-09-26 01:50:34 UTC Post #256155
The shape of the skybox doesn't matter (with exceptions - do some testing of your own to figure out what works and what doesn't), but it must be textured with the tools/toolsskybox texture.

Does that answer your question?
Penguinboy PenguinboyHaha, I died again!
Posted 16 years ago2008-09-30 00:40:20 UTC Post #256361
Yes thanks Penguin but only 1 last question - what happens if I have tunnels that go outside of the skybox area - ie you're tutorial is 4x4 with the sky going off the walls what would happen if you ran a tunnel from 1 wall around the map to have it come around to the other wall? If the tunnel is wholy enclosed with no leaks, then the skybox doesn't matter in that case - am I correct in assuming there will be no errors?

It's been a real bugger trying to put in blocks behind my displacements (without having to redo the entire map lol) but you gotta do what you gotta do.

I guess one last question re vis - is there some way I can force my pc to show me the compile window to see what's going on? On long compiles my screen saver comes on and when I reactivate the monitor my hammer window is black and I can't see what's going on. I'm doing a compile again the block are helping with the speed but it's still going slow - the 1st line of vis happened really fast and I got to 30% fairly quickly on the 2nd - now it seems to be chugging again. Sorry that's all I can tell you due to the black screen.

Does Vis also work like this?

Part a of map doesn't have a lot of stuff so done fast and it's done, now I'll work on part b and since there's a lot it will take time before I can go to part c - in other words it's all added together slowly in chunks

or

Part a is finished, now I'll use all the data and comiple part b then when part b is completed it uses both a and b to finish part c - in other words it's all added together slowly in chunks but the data of the first chunks is used to build the other chunks

or

It looks at your entire map as a whole and does absolutely everything at the same time?

I'm just trying to figure out why vis seems to go from 0-30% and then chugs slower and slower as it goes.

The 1st compile took 4 days and I only got to 40%, the last one before this in 4 hours, wasn't using comp so thought I'd run a test to check my optimzation got it to 50% and now I've been going 24 hours and don't want to stop least I kill it at 90% and go DOH!!!
Posted 16 years ago2008-09-30 01:04:05 UTC Post #256362
disable your screensaver :P

as for your 3D skybox question you should probably just test it yourself with a small map, but it should be fine.

There's still something seriously wrong if it's taking that long to compile.
Penguinboy PenguinboyHaha, I died again!
Posted 16 years ago2008-09-30 12:36:57 UTC Post #256389
Yep - studying up on map optimization now :-)

Found a good link - so posting it here, hopefully this post that I started will give other noobers like myself good info.

http://www.student.ru.nl/rvanhoorn/optimization.php?chapter=intro&PHPSESSID=fb997c30d05321af4f16961ffb59b73a

Although this is good I'm getting some conflicting messages - do they or do they not (func_detail brushes) lower VIS compile times?

I went func_detail happy last night :P
Posted 16 years ago2008-10-10 12:40:29 UTC Post #256872
Thanks for all the help guys! I rebuilt a lot of the map - sealed displacements - map now comiles in about 5 hours now - beter than the 4 days I started with and I'm happy with this time :) I still have to optimize it too :-) huge map being done - will have some screens to show you soon
Posted 16 years ago2008-10-17 16:34:04 UTC Post #257269
Thanks for all your help! The map - pl_bugscanyon-beta2 - is now in the vault for your perusal :-)
You must be logged in to post a response.