Hey guys, I made a large trade map, and I have some problem when making a compile (VBSP, VVIS, and VRAD all on normal.) Normally I would compile just fine with no errors. I checked manually and with the Interlopers checker, but if you REALLY want the compile log I an provide it.
Before I even start showing the "limits list" I just want to say that I myself am not sure as to why the compile issue happens and I am just making assumptions.
Anyways, here's the "limits list" for my map compiling normally:
As you can see, my entdata is way over capacity (I know that the limits that are shown in hammer are dynamic, and are just there for budgeting really.) My brushes and brushsides are very high.
So I decided I would make a minigame (high jump) and add it to the map. I have already ran into issues where the map would be too full, but I managed to make sacrifices. It compiled fine without any problems (I made it in a separate file because the 2d views are crowded in my main map.) Here's the list for it:
The entdata is only at 11.7%, so when I add this to the main map, the total entdata should be around 169%. I do have a list for this compile too, but it only compiled because I had a leak (the previous lists are of compiled after the leak was fixed.)
I knew that 169% was too much as 157% was already borderline on the "No free edicts" error. The map did run though, but obviously half the lighting wasn't done due to the leak and that's unacceptable.
After fixing the leak it wouldn't compile VRAD. VRAD would stop at something like:
And by stop I mean VRAD would finish there, and then it would be written to the bsp, so it's not like VRAD just slowed down a lot. This is evident because
is missing from the compile log along with that long list. When loading up the map it would just be fullbright.
I started removing lights, making more models not solid, etc. I managed to reduce the entdata to 132% (despite not being happy with some of the sacrifices I've made.)
I was sure this would compile with the minigame now, but it didn't. What doesn't make sense is after reducing the main map's entdata to 132%, then compiling with that extra 11%, it wouldn't compile lighting. This doesn't make sense because 132+11=143. 143<157, which was the previous entada, WITHOUT highjump. Nothing else went over the limit, but waterindices went from 74% to 86%, despite not having any water in highjump. I don't really know what waterindices is though.
Could anyone explain what the hell s going on here?
Before I even start showing the "limits list" I just want to say that I myself am not sure as to why the compile issue happens and I am just making assumptions.
Anyways, here's the "limits list" for my map compiling normally:
Code:
Object names Objects/Maxobjs Memory / Maxmem Fullness
------------ --------------- --------------- --------
models 473/1024 22704/49152 (46.2%)
brushes 6748/8192 80976/98304 (82.4%) VERY FULL!
brushsides 60170/65536 481360/524288 (91.8%) VERY FULL!
planes 34032/65536 680640/1310720 (51.9%)
vertexes 38850/65536 466200/786432 (59.3%)
nodes 6207/65536 198624/2097152 ( 9.5%)
texinfos 5379/12288 387288/884736 (43.8%)
texdata 306/2048 9792/65536 (14.9%)
dispinfos 68/0 11968/0 ( 0.0%)
disp_verts 3380/0 67600/0 ( 0.0%)
disp_tris 5056/0 10112/0 ( 0.0%)
disp_lmsamples 236160/0 236160/0 ( 0.0%)
faces 23609/65536 1322104/3670016 (36.0%)
hdr faces 0/65536 0/3670016 ( 0.0%)
origfaces 19289/65536 1080184/3670016 (29.4%)
leaves 6681/65536 213792/2097152 (10.2%)
leaffaces 29145/65536 58290/131072 (44.5%)
leafbrushes 11292/65536 22584/131072 (17.2%)
areas 19/256 152/2048 ( 7.4%)
surfedges 189261/512000 757044/2048000 (37.0%)
edges 118052/256000 472208/1024000 (46.1%)
LDR worldlights 919/8192 80872/720896 (11.2%)
HDR worldlights 0/8192 0/720896 ( 0.0%)
leafwaterdata 1/32768 12/393216 ( 0.0%)
waterstrips 2780/32768 27800/327680 ( 8.5%)
waterverts 0/65536 0/786432 ( 0.0%)
waterindices 48750/65536 97500/131072 (74.4%)
cubemapsamples 96/1024 1536/16384 ( 9.4%)
overlays 63/512 22176/180224 (12.3%)
LDR lightdata [variable] 9690556/0 ( 0.0%)
HDR lightdata [variable] 0/0 ( 0.0%)
visdata [variable] 35149/16777216 ( 0.2%)
entdata [variable] 617168/393216 (157.0%) VERY FULL!
LDR ambient table 6681/65536 26724/262144 (10.2%)
HDR ambient table 6681/65536 26724/262144 (10.2%)
LDR leaf ambient 17325/65536 485100/1835008 (26.4%)
HDR leaf ambient 6681/65536 187068/1835008 (10.2%)
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] 1/12 ( 8.3%)
static props [variable] 1/70558 ( 0.0%)
pakfile [variable] 10310842/0 ( 0.0%)
physics [variable] 2501576/4194304 (59.6%)
physics terrain [variable] 15618/1048576 ( 1.5%)
As you can see, my entdata is way over capacity (I know that the limits that are shown in hammer are dynamic, and are just there for budgeting really.) My brushes and brushsides are very high.
So I decided I would make a minigame (high jump) and add it to the map. I have already ran into issues where the map would be too full, but I managed to make sacrifices. It compiled fine without any problems (I made it in a separate file because the 2d views are crowded in my main map.) Here's the list for it:
Code:
Object names Objects/Maxobjs Memory / Maxmem Fullness
------------ --------------- --------------- --------
models 59/1024 2832/49152 ( 5.8%)
brushes 494/8192 5928/98304 ( 6.0%)
brushsides 3640/65536 29120/524288 ( 5.6%)
planes 1546/65536 30920/1310720 ( 2.4%)
vertexes 3027/65536 36324/786432 ( 4.6%)
nodes 1171/65536 37472/2097152 ( 1.8%)
texinfos 422/12288 30384/884736 ( 3.4%)
texdata 67/2048 2144/65536 ( 3.3%)
dispinfos 0/0 0/0 ( 0.0%)
disp_verts 0/0 0/0 ( 0.0%)
disp_tris 0/0 0/0 ( 0.0%)
disp_lmsamples 0/0 0/0 ( 0.0%)
faces 1642/65536 91952/3670016 ( 2.5%)
hdr faces 0/65536 0/3670016 ( 0.0%)
origfaces 1130/65536 63280/3670016 ( 1.7%)
leaves 1231/65536 39392/2097152 ( 1.9%)
leaffaces 2373/65536 4746/131072 ( 3.6%)
leafbrushes 1032/65536 2064/131072 ( 1.6%)
areas 2/256 16/2048 ( 0.8%)
surfedges 12433/512000 49732/2048000 ( 2.4%)
edges 7771/256000 31084/1024000 ( 3.0%)
LDR worldlights 2/8192 176/720896 ( 0.0%)
HDR worldlights 0/8192 0/720896 ( 0.0%)
leafwaterdata 0/32768 0/393216 ( 0.0%)
waterstrips 233/32768 2330/327680 ( 0.7%)
waterverts 0/65536 0/786432 ( 0.0%)
waterindices 4038/65536 8076/131072 ( 6.2%)
cubemapsamples 28/1024 448/16384 ( 2.7%)
overlays 1/512 352/180224 ( 0.2%)
LDR lightdata [variable] 1580492/0 ( 0.0%)
HDR lightdata [variable] 0/0 ( 0.0%)
visdata [variable] 0/16777216 ( 0.0%)
entdata [variable] 46075/393216 (11.7%)
LDR ambient table 1231/65536 4924/262144 ( 1.9%)
HDR ambient table 1231/65536 4924/262144 ( 1.9%)
LDR leaf ambient 5586/65536 156408/1835008 ( 8.5%)
HDR leaf ambient 1231/65536 34468/1835008 ( 1.9%)
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] 1/12 ( 8.3%)
static props [variable] 1/12 ( 8.3%)
pakfile [variable] 3071764/0 ( 0.0%)
physics [variable] 187659/4194304 ( 4.5%)
physics terrain [variable] 2/1048576 ( 0.0%)
The entdata is only at 11.7%, so when I add this to the main map, the total entdata should be around 169%. I do have a list for this compile too, but it only compiled because I had a leak (the previous lists are of compiled after the leak was fixed.)
Code:
Object names Objects/Maxobjs Memory / Maxmem Fullness
------------ --------------- --------------- --------
models 532/1024 25536/49152 (52.0%)
brushes 7249/8192 86988/98304 (88.5%) VERY FULL!
brushsides 63846/65536 510768/524288 (97.4%) VERY FULL!
planes 35326/65536 706520/1310720 (53.9%)
vertexes 44479/65536 533748/786432 (67.9%)
nodes 10070/65536 322240/2097152 (15.4%)
texinfos 5814/12288 418608/884736 (47.3%)
texdata 404/2048 12928/65536 (19.7%)
dispinfos 68/0 11968/0 ( 0.0%)
disp_verts 3380/0 67600/0 ( 0.0%)
disp_tris 5056/0 10112/0 ( 0.0%)
disp_lmsamples 236160/0 236160/0 ( 0.0%)
faces 25664/65536 1437184/3670016 (39.2%)
hdr faces 0/65536 0/3670016 ( 0.0%)
origfaces 20488/65536 1147328/3670016 (31.3%)
leaves 10603/65536 339296/2097152 (16.2%)
leaffaces 32731/65536 65462/131072 (49.9%)
leafbrushes 12093/65536 24186/131072 (18.5%)
areas 20/256 160/2048 ( 7.8%)
surfedges 203744/512000 814976/2048000 (39.8%)
edges 127275/256000 509100/1024000 (49.7%)
LDR worldlights 353/8192 31064/720896 ( 4.3%)
HDR worldlights 0/8192 0/720896 ( 0.0%)
leafwaterdata 1/32768 12/393216 ( 0.0%)
waterstrips 3198/32768 31980/327680 ( 9.8%)
waterverts 0/65536 0/786432 ( 0.0%)
waterindices 56511/65536 113022/131072 (86.2%) VERY FULL!
cubemapsamples 229/1024 3664/16384 (22.4%)
overlays 64/512 22528/180224 (12.5%)
LDR lightdata [variable] 11415772/0 ( 0.0%)
HDR lightdata [variable] 0/0 ( 0.0%)
visdata [variable] 0/16777216 ( 0.0%)
entdata [variable] 664385/393216 (169.0%) VERY FULL!
LDR ambient table 10603/65536 42412/262144 (16.2%)
HDR ambient table 10603/65536 42412/262144 (16.2%)
LDR leaf ambient 44015/65536 1232420/1835008 (67.2%)
HDR leaf ambient 10603/65536 296884/1835008 (16.2%)
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] 1/12 ( 8.3%)
static props [variable] 1/70890 ( 0.0%)
pakfile [variable] 24401753/0 ( 0.0%)
physics [variable] 2692432/4194304 (64.2%)
physics terrain [variable] 15618/1048576 ( 1.5%)
I knew that 169% was too much as 157% was already borderline on the "No free edicts" error. The map did run though, but obviously half the lighting wasn't done due to the leak and that's unacceptable.
After fixing the leak it wouldn't compile VRAD. VRAD would stop at something like:
Code:
Build Patch/Sample Hash Table(s).....Done<0.2424 sec>
And by stop I mean VRAD would finish there, and then it would be written to the bsp, so it's not like VRAD just slowed down a lot. This is evident because
Code:
0...1...2...3...4...5...6...7...8...9...10FinalLightFace Done
393 of 649 (60% of) surface lights went in leaf ambient cubes.
0...1...2...3...4...5...6...7...8...9...10Writing leaf ambient...done
Ready to Finish
is missing from the compile log along with that long list. When loading up the map it would just be fullbright.
I started removing lights, making more models not solid, etc. I managed to reduce the entdata to 132% (despite not being happy with some of the sacrifices I've made.)
Code:
Object names Objects/Maxobjs Memory / Maxmem Fullness
------------ --------------- --------------- --------
models 474/1024 22752/49152 (46.3%)
brushes 6747/8192 80964/98304 (82.4%) VERY FULL!
brushsides 60143/65536 481144/524288 (91.8%) VERY FULL!
planes 33998/65536 679960/1310720 (51.9%)
vertexes 38833/65536 465996/786432 (59.3%)
nodes 6213/65536 198816/2097152 ( 9.5%)
texinfos 5411/12288 389592/884736 (44.0%)
texdata 372/2048 11904/65536 (18.2%)
dispinfos 68/0 11968/0 ( 0.0%)
disp_verts 3380/0 67600/0 ( 0.0%)
disp_tris 5056/0 10112/0 ( 0.0%)
disp_lmsamples 236160/0 236160/0 ( 0.0%)
faces 23603/65536 1321768/3670016 (36.0%)
hdr faces 0/65536 0/3670016 ( 0.0%)
origfaces 19278/65536 1079568/3670016 (29.4%)
leaves 6688/65536 214016/2097152 (10.2%)
leaffaces 29139/65536 58278/131072 (44.5%)
leafbrushes 11290/65536 22580/131072 (17.2%)
areas 19/256 152/2048 ( 7.4%)
surfedges 189197/512000 756788/2048000 (37.0%)
edges 118025/256000 472100/1024000 (46.1%)
LDR worldlights 812/8192 71456/720896 ( 9.9%)
HDR worldlights 0/8192 0/720896 ( 0.0%)
leafwaterdata 1/32768 12/393216 ( 0.0%)
waterstrips 2778/32768 27780/327680 ( 8.5%)
waterverts 0/65536 0/786432 ( 0.0%)
waterindices 48732/65536 97464/131072 (74.4%)
cubemapsamples 201/1024 3216/16384 (19.6%)
overlays 63/512 22176/180224 (12.3%)
LDR lightdata [variable] 9690420/0 ( 0.0%)
HDR lightdata [variable] 0/0 ( 0.0%)
visdata [variable] 43455/16777216 ( 0.3%)
entdata [variable] 519546/393216 (132.1%) VERY FULL!
LDR ambient table 6688/65536 26752/262144 (10.2%)
HDR ambient table 6688/65536 26752/262144 (10.2%)
LDR leaf ambient 17339/65536 485492/1835008 (26.5%)
HDR leaf ambient 6688/65536 187264/1835008 (10.2%)
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] 1/12 ( 8.3%)
static props [variable] 1/65100 ( 0.0%)
pakfile [variable] 21432292/0 ( 0.0%)
physics [variable] 2500914/4194304 (59.6%)
physics terrain [variable] 15618/1048576 ( 1.5%)
I was sure this would compile with the minigame now, but it didn't. What doesn't make sense is after reducing the main map's entdata to 132%, then compiling with that extra 11%, it wouldn't compile lighting. This doesn't make sense because 132+11=143. 143<157, which was the previous entada, WITHOUT highjump. Nothing else went over the limit, but waterindices went from 74% to 86%, despite not having any water in highjump. I don't really know what waterindices is though.
Could anyone explain what the hell s going on here?