Warning: Too Many Direct Light Styles On A Face

kawak · October 2003

<div class="IPBDescription">virtual memory problem?</div> Hello!

Again..i have another problem with ns_agora <img src='http://www.unknownworlds.com/forums/html/emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif'>
This isn't a big problem because the compile ends up succesfully (after a long time..)

The Rad is going very slow with the "Makescales thingy" (it takes age to end up).
Then the 2 Bounce Gatherlights's operation are going crazy (very looong too)
(i use zhlt 1.7 with the (awesome) XPCagey's tools and batch compiler)

After each bounces, I have the message:
"Too many direct light styles on a face"
and a Windows's dialog Box says me that my virtual memory is going very bad (i dont remember the exactly term).

This is a bit annoying cause now i have a pentium 3Ghz, 512mo ram, 2 hard drives, ati radeon 9800 pro, and windows XP.
With my old pc (750mghz,128 ram) the makescales was endless..

Here is my log:
</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>BuildFacelights:
(1290.61 seconds)
BuildVisLeafs:
(817.50 seconds)
visibility matrix : 90.8 megs
MakeScales:
(2710.98 seconds)
SwapTransfers:
(1421.39 seconds)
Transfer Lists : 273234480 : 273.23M transfers
Indices : 103913748 : 99.10M bytes
Data : 1092937920 : 1042.31M bytes
GatherLight:
<span style='color:red'>Warning: TooW maarnnyi ndgi:r eTcoto lmiagnhyt dsitryelcets loing hat fsatcyel(e?s, ?o,n? )a face(?,?,?) </span> (<-- What's the problem here??!)
(1434.88 seconds)
GatherLight:
(1014.83 seconds)
Warning: Too many direct light styles on a face(?,?,?)

Warning: Too many direct light styles on a face(?,?,?)

Warning: Too many direct light styles on a face(?,?,?)

Warning: Too many direct light styles on a face(?,?,?)

FinalLightFace:
(28.92 seconds)</td></tr></table><span class='postcolor'>

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Object names Objects/Maxobjs Memory / Maxmem Fullness
------------ --------------- --------------- --------
models 118/400 7552/25600 (29.5%)
planes 36432/65535 728640/1310700 (55.6%)
vertexes 15503/65535 186036/786420 (23.7%)
nodes 9200/32767 220800/786408 (28.1%)
texinfos 10594/32767 423760/1310680 (32.3%)
faces 12627/65535 252540/1310700 (19.3%)
clipnodes 25825/32767 206600/262136 (78.8%)
leaves 5024/8192 140672/229376 (61.3%)
marksurfaces 16975/65535 33950/131070 (25.9%)
surfedges 53592/512000 214368/2048000 (10.5%)
edges 28099/256000 112396/1024000 (11.0%)
texdata [variable] 7660/8388608 ( 0.1%)
lightdata [variable] 5540076/6291456 (88.1%)
visdata [variable] 187353/2097152 ( 8.9%)
entdata [variable] 50986/524288 ( 9.7%)</td></tr></table><span class='postcolor'>

The only problem that i've seen in the .bsp was 1 world brush very very brighted (due to the "too many direct.." ?).

When the compile is finished, i have to reboot my pc cause it runs very slowly.

Is there a way to resolve this problem?
512 mo ram is not enough? (i really hope that this is not the reason <img src='http://www.unknownworlds.com/forums/html/emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif'> )
Is Windows XP the bad guy?
And i also want to know if this kind of compilation isn't "recommended" for the computer?

Tanks in advance for your help

Vahn_Paktu · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>CODE</b> </td></tr><tr><td id='CODE'>Warning: TooW maarnnyi ndgi:r eTcoto lmiagnhyt dsitryelcets loing hat fsatcyel(e?s, ?o,n? )a face(?,?,?)</td></tr></table><span class='postcolor'> This is an odd error message. Maybe something with the tools themselves.

What compile settings are you using.

DarkATi · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Too Many Direct Light Styles on a Face</td></tr></table><span class='postcolor'>

Self-Explanatory, where do you have alot of different colored lights?

~ DarkATi

EDIT: Grammar

devilblocks · October 2003

IIRC this is from too many lights that can be turned on/off shining on the same face. I believe the limit is 4.

kawak · October 2003

sorry, i certainly did not explain clearly my problem (my english sux).

In fact i know how to resolve this error ( i already have it with a cs map),
But "normally" this error looks like:
Warning: Too many direct light styles on a face <u>(592.000000,-768.000000,-255.500000)</u>

For me, i dont have any "Coordinates"
it's (?,?,?)
And i'm not using anywhere in my map more than 1 lights with a special fx at the same place.

So i'm not sure that i really have this error..

I just want to know why windows says me that my "virtual memory" is going bad (just after the "makescales")?
And if the "virtual memory" could "affects" the RAD's compiler (and then some stranges errors appears)?

ps: i really have that in my log:
</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>GatherLight:
Warning: TooW maarnnyi ndgi:r eTcoto lmiagnhyt dsitryelcets loing hat fsatcyel(e?s, ?o,n? )a face(?,?,?)</td></tr></table><span class='postcolor'>

DarkATi · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (devilblocks @ Oct 14 2003, 11:33 AM)</td></tr><tr><td id='QUOTE'> IIRC this is from too many lights that can be turned on/off shining on the same face. I believe the limit is 4.  </td></tr></table><span class='postcolor'> 
That too DD but I get this error alot and I have no on/off lights. Texture lights can cause this as well as entity lights. I usually get this when I have more than three lights of contrasting colors (Red, Green Blue etc.) within about 96 units of each other.

~ DarkATi

DarkATi · October 2003

What version of Cagey's tools are you using? P10? p12? etc.

~ DarkATi

WolfWings · October 2003

You either have bad RAM, or a corrupted copy of the compile tools. It's really that simple. :-) This is the fullproof, but perhaps not as quick as possible, way of solving this problem.

If you have a CD burner, download <a href='http://www.memtest86.com/memtest86-3.0.iso.zip' target='_blank'>this MemTest86 .ISO</a> and burn a copy. Leave it in the drive and reboot your machine. It will automatically boot the CD, and start testing your memory very exhaustively.

To guarantee that it completely tests all your memory with every test, after the program has loaded, press C, then 2, then pick the entry for 'All Tests' and let it run overnight. If it detects any errors... guess what, you have bad RAM. =^.^=

If it doesn't detect any errors, then you have a corrupted file. That means either your HD is going bad, or your copy of the file is either bad or virus-infected. Go to <a href='http://housecall.trendmicro.com/' target='_blank'>Trend Micro's HouseCall free online virus scanner</a> and let it run through your ENTIRE system. Be patient, it's worth it.

If it detects any viruses, take care of them as appropriate. I can't really help any more than that with a virus problem.

If it doesn't detect any viruses, you now either have a failing HD, or simply a bad copy of the file.

Double-click on 'My Computer'
Right-click on the drive you have the compile tools installed on
Pick 'P<u>r</u>operties' from the pop-up menu
Go to the 'Tools' tab
Click on the '<u>C</u>heck Now...' button beside 'Error-checking' or 'Surface scan' or something similair
Make sure 'Sca<u>n</u> for and attempt recovery of bad sectors' is checked, along with '<u>A</u>utomatically fix file system errors' and click on the '<u>S</u>tart' button.
Be patient, let it completely run through the drive, it may take a while.

Now, regardless of if it detected any errors, re-download the compile tools and re-install them, as your copy was corrupted to get to this point in the first place. The previous steps were to help make sure the corruption didn't spread further, and that any corrupted files were marked as such. The files are still, likely, corrupted, so you need to re-install them. :-)

DarkATi · October 2003

I just noticed something. That "jumbled sentence" spells out the error.

Displace every other letter and you get two sentences.

<b>EDIT: Not always EVERY OTHER letter but just omit the ones that don't spell "Too Many Direct Lights Styles ona face."</b>

Too Many Direct Light Styles on a Face

&

Warning g: Too light dircetling ht... etc. (Too lazy to figure out the rest.)

Kind of creepy. Are you the anti-christ trying to recruit us through subliminal messages? Dircetling! Dircetling! Just kidding man. <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'> Wierd error.

~ DarkATi

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Warning: <b>Too</b>W <b>ma</b>arn<b>ny</b>i n<b>d</b>g<b>i</b>:<b>r</b> <b>e</b>T<b>c</b>o<b>t</b>o <b>l</b>m<b>i</b>a<b>gn[b]h</b>y<b>t</b> d<b>s</b>i<b>t</b>r<b>y</b>e<b>l</b>c<b>e</b>t<b>s</b> l<b>o</b>i<b>n</b>g h<b>a</b>t <b>f</b>s<b>a</b>t<b>c</b>y<b>e</b>l(e?s, ?o,n? )a face(?,?,?)</td></tr></table><span class='postcolor'>

Kage · October 2003

Update/redownload your compile tools. I think Cagey added or fixed the location indicator a couple of versions back.

It's not only the number of lights that hit the face, but also their names. The targetname of the light gets involved at some point; try getting rid of unnecessary targetnames or consolidating several lights under one targetname.

watch_me_die · October 2003

Again, tommy is your friend <img src='http://www.unknownworlds.com/forums/html/emoticons/smile.gif' border='0' style='vertical-align:middle' alt='smile.gif'>

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Warning: Too many direct light styles on a face(coordinates given)

You have too many different kinds of light shining on a face(s) of a brush(es): reddish, yellowish, blueish, bight, soft, dim - too many types. Or too many switchable lights. You can have more than one light shine in an area, but it should be the same light type/color/shade/name(if any) if possible. If they are texture lights, how the texture looks is unimportant, it is the type/color/shade of light shining that is important. The new switchable texture lights count as one dynamic style if they are all the same type/color/shade/name.
I am not certain about the new shadow coloring control, but I do not think that applies.....
(Each face can have shining on it 4 light styles, 3 switchable/dynamic light styles and 1 reserved style for the static/unchanging lightmap. The lightmap is sub-limited in turn to 8 static/unchanging types, some total combination of one env_light, some light entitys and some spotlight entitys, and texture lighting all shining on one face of a brush. That is because 8 static light take 64 times as much RAD calc/bsp mem as 1 light, times the number of brush faces involved. There are also 8 possible combinations of the lightmap and the switchable lights, making for more long RAD work.)
choice to fix:
1. remove a bunch of lights in the area.
2. change more lights to the same type - type by light properties.
(Recent info from Cagey about how the RAD code works revealed that it looks at EVERY entity with a Style property and counts it in as a light style! This includes entities such as buttons that do not give off light!)
3. So also go thru your entitys with Smartedit off an remove Style stuff, or open up your .map of your level with an editor like notepad and remove the Style values from entitys like buttons that do not give off light really....</td></tr></table><span class='postcolor'>

Maveric · October 2003

There are too many entitiy-based lights on a face.

HL doesn't like having 20 differant lights hitting a side or even a portion of a room, for some reason...

NerdIII · October 2003

This 'two in one' message is beyond my understanding of code. Ok unicode text uses two bytes for a single character, but on english systems the first byte is the letter and the second one is 0. How can these messages merge?!?
Hey have you guys heard of those PCs that crash when another person than their master trys to use them? I have seen with my own eyes how a friend of my little sister made two PCs crash after 4 minutes playing Age Of Empires. They never crashed before on this game <img src='http://www.unknownworlds.com/forums/html/emoticons/wow.gif' border='0' style='vertical-align:middle' alt='wow.gif'> . Maybe some people's aura can modify electronic signals in the frequency of the used RAM to create texts that use letters from two memory adresses!
kawak, do you know if you are a gouvernment experiment perhaps?

DarkATi · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (NerdIII @ Oct 14 2003, 09:13 PM)</td></tr><tr><td id='QUOTE'> This 'two in one' message is beyond my understanding of code. Ok unicode text uses two bytes for a single character, but on english systems the first byte is the letter and the second one is 0. How can these messages merge?!?
Hey have you guys heard of those PCs that crash when another person than their master trys to use them? I have seen with my own eyes how a friend of my little sister made two PCs crash after 4 minutes playing Age Of Empires. They never crashed before on this game <img src='http://www.unknownworlds.com/forums/html/emoticons/wow.gif' border='0' style='vertical-align:middle' alt='wow.gif'> . Maybe some people's aura can modify electronic signals in the frequency of the used RAM to create texts that use letters from two memory adresses!
kawak, do you know if you are a gouvernment experiment perhaps?  </td></tr></table><span class='postcolor'> 
In some cases Vice Versa. My friend's PC feezes all the time, when I use it, it never freezes. Hmmm.... anyway did you see the two messages though Nerd or was it just me and you're playing along?

~ DarkATi

Cagey · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (kawak @ Oct 14 2003, 08:09 AM)</td></tr><tr><td id='QUOTE'>Is there a way to resolve this problem?
512 mo ram is not enough? (i really hope that this is not the reason <img src='http://www.unknownworlds.com/forums/html/emoticons/confused.gif' border='0' style='vertical-align:middle' alt='confused.gif'> )
Is Windows XP the bad guy?
And i also want to know if this kind of compilation isn't "recommended" for the computer?

Tanks in advance for your help</td></tr></table><span class='postcolor'>
512 MB of memory should be enough for it to run OK, but you're definitely getting into virtual memory before RAD finishes:

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Indices :? 103913748 :? 99.10M bytes
Data :? 1092937920 : 1042.31M bytes</td></tr></table><span class='postcolor'>

That's over 1 GB of memory used by RAD during your compile--I'm guessing your hard drive chugs pretty hard when you're running RAD as it serves virtual memory...

Regarding the wierd error message: I'm guessing you're using at least two threads when you compile (by default the number of threads is equal to the number of CPUs on the compile machine, but you can specify more at the command line).

I don't believe the console output functions are threadsafe, so if two threads tried to write error messages at the same time, it's possible for the messages to get intermixed like DarkATI noticed. You're basically looking at two different error reports that happened "simultaneously" on two threads:

Warning: Too many direct light styles on a face(?,?,?)
Warning: Too many direct light styles on a face(?,?,?)

They are being sent at the same time to the same place and the letters end up shuffled together like cards from two halves of a deck. The fix for this problem is adding a guard around the print functions so that only one thread can enter them at a time--once that's done, the second warning won't print until the first one has finished and you end up with two readable messages.

The question marks were introduced in the texlight code, which didn't originally provide error coordinates when the error was copied and pasted from its original location. The latest copies of my tools should always report a location, but I suppose it's possible that I missed a spot when I was adding the information. Let me know which one of the "p" series you were using, and I'll check my source if it was a recent build.

kawak · October 2003

ohh, I thought that I used the last version but in fact not ( i was using p10)

I'm downloading right now the p12 and launch a compile tonight to see if it's ok (if not, i would come back <img src='http://www.unknownworlds.com/forums/html/emoticons/tounge.gif' border='0' style='vertical-align:middle' alt='tounge.gif'> ).

Thank you for your help.

EDIT:
</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>Regarding the wierd error message: I'm guessing you're using at least two threads when you compile (by default the number of threads is equal to the number of CPUs on the compile machine, but you can specify more at the command line).

I don't believe the console output functions are threadsafe, so if two threads tried to write error messages at the same time, it's possible for the messages to get intermixed like DarkATI noticed. You're basically looking at two different error reports that happened "simultaneously" on two threads:</td></tr></table><span class='postcolor'>

Like i said before my english isn't good..(it's hard to understand the "technical term") but i think u are talking about that:
</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>-= Current hlrad Settings =-
Name | Setting | Default
--------------------|---------------------|-------------------------
threads [ 2 ] [ Varies ]</td></tr></table><span class='postcolor'>
i'm using the default setting here.. (i have a "pentium® 4 CPU 3.00 GHZ".. so..should i specify 4 threads?)

WolfWings · October 2003

No, actually specifying '1' instead might actually speed things up, as the speed-gain by a hyper-threaded CPU is less than the speed-loss from the extra memory bandwidth needed. Hyperthreaded CPU's like your Pentium 4 'pretend' to be 2 CPU's, but don't perform as well as two actual, seperate CPUs, so the gain with something like the HL Compile Tools would be minimal, and might be offset by the extra memory bandwidth used.

Of course, all that is moot since you seem to be using more than your half-gig of memory for the compile, so CPU speed is a dead issue when you're limited by how fast your virtual memory is. :-)

Try -sparse though I doubt it will help much, and try raising your -chop value to 80 or 96, that will likely help much more.

One last thing that surprisingly might help, if most of your textures are scaled to less than 1, is -notexscale. This is <b><i>usually</i></b> a bad thing, but since most Natural Selection maps use down-scaling for detail instead of upscaling for terrain like most HalfLife maps, it can often lower the lightdata used, I've found.

Yamazaki · October 2003

Let's dispel this myth here and now.

Colour has no impact on light styles, none whatsoever. You can have a thousand lights of different colours in a room and it will not generate this error. This error is only generated when the appearance/custom appearance/entity-name have different variations.

Cagey · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Yamazaki @ Oct 15 2003, 12:30 PM)</td></tr><tr><td id='QUOTE'> Let's dispel this myth here and now.

Colour has no impact on light styles, none whatsoever. You can have a thousand lights of different colours in a room and it will not generate this error. This error is only generated when the appearance/custom appearance/entity-name have different variations. </td></tr></table><span class='postcolor'>
Yeah - need to bug tommy14 about that to make sure the misinformation doesn't propagate further.

EDIT: I'm gramatically challenged today.

WolfWings · October 2003

And that reminds me... Cagey, a quick note your way for something that would be VERY useful in your toolset. Some way to disable scaling DOWN the lightgrid, but allow it to still be scaled up. Basically, texture scalings less than 1 are clamped to 1 for the lightgrid, but if they're greater than 1, they're still used. So you can still use massively scaled-up textures for lowered lightgrid size on terrain, but also trim (generally useless) lightgrid detail from detail-textured surfaces like computer terminals and buttons. =^.^=

NerdIII · October 2003

@DarkATI: I didn't notice it until you mentioned it, but now Cagey cleared that for us <img src='http://www.unknownworlds.com/forums/html/emoticons/smile.gif' border='0' style='vertical-align:middle' alt='smile.gif'>
'threads' was actually spooking through my head when I saw the messages merge, but I was so damn sure that the tools only use one thread that I didn't bother about that. Silly Microsoft functions. Shouldn't they notice that two outputs in the same console will look weird and frighten all too superstitious. <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'>
@WolfWings: I have a strange feeling that Cagey already impelented this. He said that he did a lot of optimizations with ... forgot what it was, but something about the light patch generation or so. *waits for official answer*

Cagey · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (NerdIII @ Oct 15 2003, 05:58 PM)</td></tr><tr><td id='QUOTE'> @WolfWings: I have a strange feeling that Cagey already impelented this. He said that he did a lot of optimizations with ... forgot what it was, but something about the light patch generation or so. *waits for official answer* </td></tr></table><span class='postcolor'>
I sped up the inner loop math functions for HLRAD (speed optimizations), then spent three revisions cleaning up the mess <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'>. I haven't intentionally done anything with the algorithms themselves... if everything's working as I meant it to, p12 should look like a 1.7 lighting pass but compile (not run in game) faster.

kawak · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (kawak @ Oct 14 2003, 06:18 PM)</td></tr><tr><td id='QUOTE'>And if the "virtual memory" could "affects" the RAD's compiler (and then some stranges errors appears)?</td></tr></table><span class='postcolor'>

My new compile has just finished.
I'm still using the p10 version.. (i forgot to install the p12 :s )

I made only one change in the RAD's settings. I increase the -chop value to 96 like (as?) WolfWings suggests me (was using 32 before).

The result is very cool:

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> </td></tr><tr><td id='QUOTE'>BuildFacelights:
(1338.63 seconds)
BuildVisLeafs:
(103.36 seconds)
visibility matrix : 13.5 megs
MakeScales:
(297.53 seconds)
SwapTransfers:
(6.92 seconds)
Transfer Lists : 32954054 : 32.95M transfers
Indices : 21442084 : 20.45M bytes
Data : 131816216 : 125.71M bytes
GatherLight:
(5.52 seconds)
GatherLight:
(8.13 seconds)
FinalLightFace:
(5.08 seconds)</td></tr></table><span class='postcolor'>

It takes "only" 350 sec with the makescales, swaptransfers,..bounces..etc.. (check my old log on first page to see the amazing changes).
No more windows's dailog box about the "virtual memory"
And no more weird error message "too many direct.." ("whereas" i didnt make any changes in the map).

And i didn't see any problems on the bsp.. (the lighting seemed to be "normal")

So the virtual memory's problems can cause some strange errors...

WolfWings · October 2003

Gah... no, it you had your -chop at 32 before, you were INSANE. Just raise it to 64, hon. Having -chop at 32 was the whole source of your problem, that's more than 4 times the detail half-life lighting is supposed to have. =^.^=

It'll take about twice as long to light as it does now, but will look noticably better. :-)

NerdIII · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (XP-Cagey @ Oct 16 2003, 03:12 AM)</td></tr><tr><td id='QUOTE'> </span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (NerdIII @ Oct 15 2003, 05:58 PM)</td></tr><tr><td id='QUOTE'> @WolfWings: I have a strange feeling that Cagey already impelented this. He said that he did a lot of optimizations with ... forgot what it was, but something about the light patch generation or so. *waits for official answer* </td></tr></table><span class='postcolor'>
I sped up the inner loop math functions for HLRAD (speed optimizations), then spent three revisions cleaning up the mess <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'>. I haven't intentionally done anything with the algorithms themselves... if everything's working as I meant it to, p12 should look like a 1.7 lighting pass but compile (not run in game) faster.  </td></tr></table><span class='postcolor'> 
I was thinking more about your rewrite of the tools. In the current versions the tools don't bother if there are too many light patches on a face because the light patches are scaled down to much. HLRAD would stop with an error message then. The new tools could automatically adjust a too small value to save the compile. Sorry, I forgot the error message, but if I remember right there were numbers like 16 or 17... <img src='http://www.unknownworlds.com/forums/html/emoticons/sad.gif' border='0' style='vertical-align:middle' alt='sad.gif'>
The faster lighting algorithm is amazing though now that it (seems to) work! Now you can start making it slower again with fancy stuff like the texture-colored reflections <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'>

Redford · October 2003

Attn: XP-Caegy is my official mapping diety. That is all.

DarkATi · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Yamazaki @ Oct 15 2003, 02:30 PM)</td></tr><tr><td id='QUOTE'> Let's dispel this myth here and now.

Colour has no impact on light styles, none whatsoever. You can have a thousand lights of different colours in a room and it will not generate this error. This error is only generated when the appearance/custom appearance/entity-name have different variations.  </td></tr></table><span class='postcolor'> 
I stand corrected. <img src='http://www.unknownworlds.com/forums/html/emoticons/smile.gif' border='0' style='vertical-align:middle' alt='smile.gif'>

~ DarkATi

kawak · October 2003

</span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (WolfWings @ Oct 16 2003, 10:15 PM)</td></tr><tr><td id='QUOTE'>Gah... no, it you had your -chop at 32 before, you were INSANE. Just raise it to 64, hon. Having -chop at 32 was the whole source of your problem, that's more than 4 times the detail half-life lighting is supposed to have. =^.^=

It'll take about twice as long to light as it does now, but will look noticably better. :-)</td></tr></table><span class='postcolor'>
ahhh ok <img src='http://www.unknownworlds.com/forums/html/emoticons/tounge.gif' border='0' style='vertical-align:middle' alt='tounge.gif'>

i've never understood what was the chop for..i use it whith the -sparse cause i had seen somewhere (on this forum maybe :s ) that these settings resolved the MAX_PATCH error (and it worked.. so i was sure that the -chop 32 was only "friendly").

I'm going to decrease it to 64, Thanks for the advice ;-)

Cagey · October 2003

Tommy14's page has been updated with the correction <img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'>

<a href='http://www.slackiller.com/tommy14/errors.htm#lightstyles' target='_blank'>http://www.slackiller.com/tommy14/errors.h...htm#lightstyles</a>

Warning: Too Many Direct Light Styles On A Face

Comments