 News | Forum | People | FAQ | Links | Search | Register |
Log in |
|
|
More proof!
<goetz> http://www.celephais.net/quake/ <goetz> metlslime is that u ? <goetz> i thought u were male <metlslime> i'm a hot japanese girl <Auhsan> woa <Auhsan> I love you metlslime <metlslime> this explains why i never get laid, after all, who wants to have sex with a hot japanese girl? <metlslime> nobody. I wish i was one of those skinny white computer geeks. They get all the sex. <Auhsan> from their hands <metlslime> heh * metlslime cries I am just trying to see the difference between using _sunlight or placed light entities in a map.
For some reason, light does not seem to work on my trial, 2055 brush, map. Here is the readout: - 4361 planes 87220 4252 vertexes 51024 2387 nodes 57288 6 texinfo 240 3888 faces 77760 13244 clipnodes 105952 1202 leafs 33656 4808 marksurfaces 9616 17010 surfedges 68040 8506 edges 34024 2 textures 130652 lightdata 0 visdata 0 entdata 176 File: ter2.bsp 3 entities read, 1 are lights, 3888 faces Light: 0.0%, Elapsed: 0:00 Light: 100.0%, Elapsed: 0:01 lightdatasize: 0 0 switchable light styles Elapsed time : 0:02 BspBuild [Light complete] The numbers mean nothing to me. Can anybody spot my (probably obvious) error? like you only put on light entity:
3 entities read, 1 are lights, 3888 faces sunlight needs: _sunlight # for intensity, and _sun_mangle x x x for direction. there's a phantom brush which has cropped up in my map which i'm trying to get rid of...
the problem is, the brush doesn't appear in gtkradiant, but when i compile the map, it is there, in quake. there's nothing wrong with the brush, other than it is a phantom... it's not super deformed or huge or any of the common problems... any suggestions? Look in the readme for sunmangle details and some recommendations to start with. The third number, the "placeholder", isn't required at all anymore in my tools (but is accepted for compatibility).
In your log printout, lightdatasize is 0, which means that there is no lights generated at all (the map is probably pitch black). There is one light entity (sunlight isn't counted as a light entity), but for some reason is doesn't generate any measurable light. Send me the zipped map+wad and what tool arguments you're using and I'll try to find out why it doesn't work. I believe texture randomisation in HL is done with a naming convention, so to randomise between 3 textures you would name them
-0texname -1texname -2texname and so on. They probably have to have the same dimentions for this to work. Dumb question probably, but you're sure you don't have anything either filtered or hidden in-editor?
If not, try brush-cleanup (either plugins or a button, depending on what version you're running). You could also try hiding all brushes in the area (select one-by-one and hide) and then doing a "select touching" in the empty space that's left and see if you get anything selected? Necros, sunlight only needs the intensity value to be activated, the sunmangle is default 0,-90 (pretty boring zenith sunlight, though). Check out the readme for this and other related information.
As for the phantom brush, is it solid, i.e. not penetrable with rockets/player? i believe it works as preacher said, but as for the implementation, i think it subdivides the face into polygons (just like quake does, actually) and each polygon of the original face gets one of the 3 textures in preacher's example.
like you only put on light entity
necros: _sunlight and _sun_mangle are keys, not entities. They both go in the worldspawn entity. I have two test maps: one is lit via the _sunlight key in the worldspawn entity, and the other is lit by one light entity with no entry in the worldspawn.
I wanted to see how _sunlight lit the map as I saw some advantages when considering how many light entities my map would require (it is not a big map in terms of depth and width, but is fairly high due to some pesky mountains. I will take up aguirRe's offer to see what he can tell me. Thanks everyone. You mean it would be possible to write a bsp compiler for Quake that does it, then?
AguirRe, Riot, whoever, get to it! :P pjw: yep, i'm sure it's not just filtered or something :)
aguire: It appears to be a completly solid brush, but it doesn't appear to be a extension from another one, because i've removed all the brushes surrounding it, and this brush is still there, in the middle of nowhere... metlslime: yeah, i meant to say: "it looks like you only put one light" -- forgot the e. As I am working on a PAK file, and adding all kind of Quake1 bsp's into it, I got rather confused of the size of these maps.
As I really got Zombie by betatesting, this strange appearance overcame to me. First I had some maps I laid my hands on, and I was quiet happy to play them. Then I decided to make them attached to eachother and turned my gloomed eyes to this board to find some knowhow. At first the maps were small enough to be played under the safe surroundings of Windows. But as I learned a lot of you and was pinching the maps to a higher level, it suddenly freaked me by getting the Alloc_hunc error. It wasn't worrying me, because I was in the opinion that with the -winmem16 option it would go as smooth as possible. Now experimenting with these PAKfiles I came to the conclusion I had to make a choice. Or I would make a PAKfile what would play under every circomstances in the Window surrounding, but than it would be the one with the smallest maps and therefore the crappy one. But at least noone would get stuck on this Alloc_hunk error, what is in my opinion the worst buck I could imagine. Or I would be as stubborn as I am, and just load all the maps with their sizes of some even 2.2Mb in the PAK file and think that when one would like to play it, one ought to size ones computer to the DOS standard and get along with it. This last one is one of my own appealings. Still I have the idea it is a reasonable question to ask, because I have found myself to much times grumbling behind this alloc_hunc error after having a lot of fun with a pak of files. In fact it is the same question as trying to make the levels to ID standards, and by doing so making the chance of distroying it under its own cummalating weight. You may find I make to many words for a small question as this, but after sending a lot of maps to all the ones who were willing to take a look at it, where I am still gratefull for, I wanted to know what you single super mappers out there would think of this question from this poor little sod out in the Quake Universe. And please don't ask if I was drinking again, because I begin to see every pack of ammunition for a 6pack of beers... because I begin to see every pack of ammunition for a 6pack of beers... :)
anyway, if you're asking if it would be ok to make players use the -winmem 16 (or -heapsize 16000) command line when playing your maps, then the answer is yes. in fact, almost all my maps have needed -winmem 16 or even -winmem 32 to play. don't worry. just include a little note in your readme about how to do it and it should be fine. What necros just said. That memory error message isn't a problem at all for most Quake players today, don't worry about it.
There are probably other more important issues you might worry about in your pak, e.g. visuals, gameplay, flow etc. Necros, you can send me the zipped map+wad and I'll take a look at it. You'll probably have to tell me where the brush is (and isn't supposed to be) as well.
Really, why is anyone bothered about edict limits anymore? There are plenty of engines out there that abolish them.
|
|
| You must be logged in to post in this thread. |
|
| Website copyright © 2002-2024 John Fitzgibbons. All posts are copyright their respective authors. |