News | Forum | People | FAQ | Links | Search | Register | Log in
Mapping Help
This is the place to ask about mapping problems, techniques, and bug fixing, and pretty much anything else you want to do in the level editor.

For questions about coding, check out the Coding Help thread: https://www.celephais.net/board/view_thread.php?id=60097
First | Previous | Next | Last
The Only Way 
is to increase the size of the texture itself. A lot of mappers increase the size of monster spawn room textures. 
I Forget The Reason 
There's a reason world polygons can't exceed a certain size in the quake engine - anyone remember? 
Surface Cache? 
 
QBSP 
cuts Faces greater then 224 units. This is afaik not changeable. 
 
I think it's a lightmap issue isn't it? They have a max size, therefore the surfaces have a max size ... which I think is what Lunaran said. 
Hmm,, 
Lightmap size sounds likely 
14588 
not quite. It cuts faces that size when the texture is 1x1 size.
It is relative to the texture size. I forget why though.

So scaling a texture to 10x10 will give you larger cuts. In the old days, you had to do that to sky textures because the compilers would cut up the sky into tons of faces. New ones do this automatically now though. But yes, if you are looking to get some extra faces, scaling up your textures will reduce the amount of cuts qbsp does.
It's rarely an issue these days though. 
Oh 
an easy free one is to make all your triggers and other invisible brushes have scaled up textures cause you never see those anyway. 
 
240 units is the default, this can be raised in some compilers but then you Brel comparability with most engines. Dark places will still load it though.

It's has to do with light maps but also, the way software renderer does lighting calculations.

Anyway who cares about poly count anymore? 
 
r_speeds be damned! 
 
"Anyway who cares about poly count anymore? "

Careful, I've tried that argument before and gotten brutally chewed up around here. :) 
 
High poly counts break vanilla Quake. If you don't care that it won't play on GLQuake (which has the same limits as software Quake but who even uses software Quake anymore) and make sure that that fact is in the release notes then I agree, who cares? Even an old (10 years or so) POS PC should be able to do GLQuake using its on board graphics chip. 
 
I had a POS PC in 2004 and it was very much capable of running GLQuake. ;) 
Hmmm 
who even uses software Quake anymore

I imagine vanilla WinQuake still has a following because of that very special je ne sais quoi that the software renderer provides.

Can't imagine many people are still playing vanilla GLQuake though. Ugh. 
 
I endorse Fitzquake with overbrights and GL_NEAREST for all your je ne sais quoi needs 
 
Also, if a lightmap is 16:1 and 8x8 it doesn't quite cover a 256x256 square. You need data points at the edges to interpolate from, but 9x9 is abhorrent and nonbinary, which is where surfaces split at 240 comes from. 
No Vanilla! 
Yea Fritz and Spasm for me with Nearest filtering for me. I just like to see how pretty we can make the game look with these most recent additions without getting lost in all the nextgen madness. 
 
I imagine vanilla WinQuake still has a following because of that very special je ne sais quoi that the software renderer provides.

Nah, that's what the qbism super 8 engine is for. ;)

Except in extreme cases, you're safe to go to the Fitzquake doubled map limits at the very least. 
Indeed 
I endorse Fitzquake with overbrights and GL_NEAREST for all your je ne sais quoi needs

Oh absolutely, but for me it doesn't quite tickle the nostalgia glands in the way oldskool 8-bit rendering does, the way it uses the colormap and whatnot.

Be nice to see an option in the fitz-type engines that uses a shader to simulate the 8-bit rendering.

necros - for whatever reason, I can't run super 8 on my pc. 
 
Except in extreme cases, you're safe to go to the Fitzquake doubled map limits at the very least.

You won't get any argument here from me on this. Way back in the early 00's I was saying that r_speeds were going to be less and less of a concern in the future. Of course I really had no idea that people would still be mapping for Quake as much as they do over a decade later, but I did know that the common PC was going to be able to handle a much larger and more detailed level design without even breaking a sweat.

Moores law and all of that stuff. 
Moving Objects? 
I have a question regarding doors and platforms... from what I have seen in quake 1 they tend to move one direction and set distance correct? Or is it possible to have them follow paths / move one direction and then another? Have a door shift backwards and then down into the ground as an example. Or a platform on a looping track? 
Doors Not Func_Trains... 
Unless I have to use that as a hack? 
 
Outside of secret doors, no. Everything moves in a linear fashion from A to B. func_trains are the only option for anything fancier. 
 
Func_train and func_path are what you will need to do. Group the brushes that you want to use into one object then give it a func_tran attribute. From there place a func_path at each point where you either want it to turn or want it to wait.

Keep in mind that odd shaped platforms may need some tweaking with the func_paths to keep the platform from disappearing into walls or whatever. The func_train is also just like a func_wall in many respects (no shadows cast by the train platform itself for example) and that whatever shadows fall on it when it spawns into the world will always be on it so be careful about where you start it off from so that it has the right lighting cast on it from the start. 
Danget Map Limits! 
I thought I saved it but I know some helpful person on this forum made a map for quake1 that was the maximum size possible that you could load up in any editor to check if your level was outside the world limits. Anyone know where that link is? :) 
First | Previous | Next | Last
You must be logged in to post in this thread.
Website copyright © 2002-2024 John Fitzgibbons. All posts are copyright their respective authors.