when is someone going to add extra8? ;)

Get dem high def shadows 

when you use -extra4, it's like using a raytracer to render a 400x400 pixel image, then scaling down to 100x100 in photoshop, versus rendering at 100x100 directly. The final resolution is the same, the -extra4 version just has imperfections smoothed out.

For actual HD shadows, you'd need a new file format, like a high-res version of .lit, and engine support :(. I was just asking spike yesterday, no one's build such a format for q1bsp as far as he was aware. 

.lit2 go! 

That's quite interesting to know, I thought it was genuinely higher res. :) 

I meant just make it less controllable to avoid weird numbers messing up the distribution.

But using the levels of 'extra' makes much more sense. 

But using the levels of 'extra' makes much more sense.
it does and it doesn't. decoupling sun count from -extra means you can do tests with higher number of suns without the lighting pass taking 30 minutes. 

http://www.quaketastic.com/files/screen_shots/5thlighttest.jpg

This is with dirt mapping and extra 4. 

how is the tree casting a shadow if it's a model? 

_shadows 1. 

Could be anti-lights, or skip textured invisibrush 

it's a func_wall with a skip texture and _shadows 1. 

I wonder would a radiosity solution work better for sunlight than just spawning multiple fake light entities. It needn't have bounced light (that's not radiosity, that's indirect lighting), just surfaces emitting light.

Using visdata (if present) to accelerate lighting seems to be something that nobody has ever done in the Quake tools. The Quake 2 tools do this and it should work quite well.

High res lighting could in theory be done by just taking the output of a -extra (etc) pass and using it directly. Would need engine support, dynamic lights and animated lightstyles would become slower to update, and shadows would start becoming harder-edged with more noticeable stair-stepping. I'm not sure it's a great idea. 

How hard would it be to add an option to up the resolution of lightmaps? That alone I think would improve Quake visuals a ton. I know it would require some engine level fiddling and tool changes, but I wonder how much REALLY. 

The biggest problem with these is updating of dynamic lights and animated lightstyles. It's not a question of work, it's a question of performance. 

@mh
Just thread it? 

Nice idea for software but in GL it all happens in glTexSubImage2D. 

Sorry for noob question, but I saw some comments in the Beyond Belief in one map thread that made me wonder if I can add AO to maps by recompiling them with TyrLite? I saw some AO screenshots and it looked really nice. In short, can I recompile the vanilla game's bsp's (and other ones as well of course) and get those nifty effects - or is it more involved than that? 

Nope, it really is that easy. Extract bsp from pak, relight with -dirtwhatever, play and ogle.

Should distribute a batch file that does all the stock maps. end.bsp might even look half decent. 

When you have a chance to play with it, curious how the _sunlight2 in the latest build compares to what you were doing!
The actual sun positionong I took from the "skylight" feature in q3map2 

i redid the quake maps but found dirt mapping isn't as awesome in those as you would think.

the real benefit comes from outdoor areas that use the modern _sunlight method because that yields completely flat light on its own, and the dirt mapping comes in and makes it awesome.

id had done all their lighting with point lights which already gives a bit of a dirt mapping look.

so yeah, dirt mapping does improve the lighting in oldschool lit maps, but the real benefits will be maps that relied a little too much on minlight or had big outdoor _sunlight areas. 

eric, will test, recovering form hard drive replacement 

eric, will test, recovering form hard drive replacement minor surgery. 

Get well soon! :-) 

Cool, thanks! 

I may be talking out of my arse here, because I don't know the tools code very well, but I was thinking....

You know how face splits due to fiddly details (thin bars, steps, walltorch holders etc.) can often cut across lots of geometry, and it sucks?

I was wondering - if it was possible to influence the order in which faces get split, could it improve the situation?

To wit - imagine a key which you could put on a func_detail (or func_group) - i.e. "_split_priority" or something - then bsp uses this to order the way faces get split so you can constrain fiddly details to splitting just the immediately adjacent faces by making them func_ and giving them a lower split priority. You could even have more than one func_ butting into each other, and use different _split_priority on them to really micro-manage the order of splitting.

Caveat - just the rumblings of someone who doesn't know much about quake's BSP algorithm, but does any of this sound interesting / make sense?