Well, let's have a try than...
Just an ordinairy level shows:

edictcount
num_edicts : 373
active : 373
vieuw : 253
touch : 326
step : 109

I know I can count to three, as well as to 600.
But does this mean 373, or the whole lot? 

the one you want to look at is active or num_edicts.

the others... i don't really know what they are. 

I noticed your sm31_speedy map and I was wondering what light tool (version and especially parameters if you recall that) you used to build that? 

question is: have you seen the map (original sm31, or some later version, or just screenshots ?)

its tyrlite and delay 2 lights


I guess you could decompile it and look yourself :) 

Orkan is a bit hard to see, since the textures are very dark and contrast themselves. 

I have the original sm31_pack (I think), are there more than one?

I've tried to re-light that bsp with TyrLite and my Light but I can't seem to get it like it was originally. AFAIK, you can't get this information from decompilation, it's all in the entity data plus invisible command line options.

Thanks, I'll try the -delay 2 and see how that goes. 

the lighting is way darker than the original. Also, I get an error about missing models which makes TyrLite abort. In my Light I've changed those into warnings but the lighting is still too dark using -dist 2.

What gives? Have you somehow manipulated the entity data after using TyrLite (e.g. with its "-compress" option)?

Wait a minute, I just noticed that you wrote "delay 2" and I mistakenly interpreted that as "-dist 2". What do you mean, can the delay be controlled from the command line? Isn't that just a key for each light entity? 

..is set to each light entity individually, like brightness (only not essential). In tyrlite it effects the attenuation, and a value of 2 sets the falloff to inverse distance^2 for the light entity it's assigned to 

about dealy xen said it all..

probably it was lit with old tyrelight with -nominlimit option, which actually made light 2x brighter
I can give you the .map if you want
Why do you care anyway ? 

info and yes, I'd like the zipped map if possible.

The reason why I care is partly that I find the map fun to play, but more important is that I'm trying to learn more about how different maps are lit and how the "-range" and "-dist" parameters affect lighting in general.

And this map consistently resisted in revealing how it was actually lit up originally. Most maps are easy to find the right combination of parameters (if any at all).

Have you any idea why the model error occurs when trying to re-light the bsp? 

about using TyrLite 0.8 with the "-nominlimit" bug. When retrying with that version and that option I immediately hit the spot with the same light distribution pattern.

Xen also reported on QMap having this problem with his xnq1002 map.

Old bugs die hard ... and there's still no explanation of the missing brush models. 

using fitzquake075.
testing out a map. map title shows, then:
got a NaN velocity on monster_shalrath
*crash* to desktop.
GLquake just crashes.

(software iterations load the map fine though)
looked for missing shalraths (vores) didn't find any.

since the only engine to actually tell me anything before crashing is fitzquake, hopefully metl might know something about it? 

is fitzquake also the only engine that crashes? 

i need to read more carefully.

better question: does this happen consistently? 

i'm looking into all the vores on the map, but i don't really see anything out of the ordinary with them. 

it happens ALL the time. like i said, software works, but it seems all gl engines crash.
(i only tested with fq, glquake and tq) but it screws up the mouse acceleration and i stopped testing gl engines because it was pissing me off. 

okay, does it happen when the map loads, or during gameplay?

Care to send me the bsp? 

when i lauched the map from the command like with +mapname, the game loads fine, and almost right after the map's title is printed on the console, i get got a NaN velocity on monster_shalrath then the game seems to continue loading for a few more seconds then crashes to the desktop.
i'll email you the current compile. 

Metl, you mentioned in your email that winquake crashes as well.

i used tronyn's winquake and it didn't crash (which is why i assumed that regular winquake would work).

so tronyn, what makes your engine so special? :) 

& one difference is that the max edict count is increased. 

sorry Tronyn.
working on the same project + having similar names = bad for necros' memory 

Has anyone got this to work for Quake editing? 

but you make it sound like it's more difficult in this version. i was planning to get it a little later, so info would be appreciated... 

I got it working, long story, but it will basically only run once on my machine, the next time I go to start it, it will crash with a windows error "GTKRadiant exe has caused an error.......................and will have to close bla bla bla" you know the one. And basically the only way I can run it again is by clearing the registry when the option is there, but then the second time I start it, its the same thing (Groundhog Day) It happens whether or not I load it for Quake, Quake II or Quake III, so I guess that rules out it being a prob just with trying to edit for Quake. I also tried uninstalling my previous version of GTK, but that didn't help any. So after many installs/uninstalls, I decided to go back to the previous version of GTK. Probably something on my machine it doesn't like, Oh well, I'll try again next release.

Give it a go, you may have better luck 

...except for the clipping bug and annoyances with the surface inspector.