How do I get func_group to work in GTK for Quake? I read it somewhere, but I can not find it. I copied the func_group thingy from the Quake3 entities.def file installed with GTK, and pasted it into the Quake.def file, as I thought this was what I had read, but it did not seem to work. 

Quake II does, though.

(What does func_group do, anyway?) 

I remember that mapcon was able to move func_groups back to worldspawn, but I don't know what to change in the def file to make func_groups.

For q1 mapping in gtkradiant all it does is allow you to handle brushes in a group, which can be handy. 

Fat Controller
From what I read, although Quake does not have func_group, you can use func_group, in GTK Radiant as an in editor only tool, which as pushplay says allows you to handle brushes in groups

pushplay
"For q1 mapping in gtkradiant all it does is allow you to handle brushes in a group, which can be handy."

This is what I am after.

I either read about it hear, or at the old QMap, in a thread about Quake mapping and GTK Radiant. I have searched here but could not find it, and can no longer access the old QMap site. 

since qbsp doesn`t know what a func_group is, it will create an actual entity which the game will claim has no spawn function. Does anybody`s map converter push func_group brushes into the worldspawn? 

Yes, SleepwalkR's does resolve func_groups, you just have to put '-groups' as a command line parameter. Go download it and read the included readme.txt for further information.

http://rem.fov120.com/?show=projects&action=display_project&id=12 

or u can bother manually ungrouping all the stuff once map is done :/ 

anyway, since there is discussion on how to remove/revert the func_group back to worldspawn etc, I think it is reasonable to assume you also know how to get func_group to work in the first place, anyone care to enlighten me?

anyone?

anyone?

anyone?

:( 

:) 

care to share? 

:) 

Now, this is how I got it to work, there is most likely a better way to do parts of it, especially selecting the group after it has been made, but at this time I do not know it.

First you got to copy the "funk_group" section from your Quake3 "entities.def" file which gets installed with GTK, just paste that into your quake.def file. (This, as I mentioned above, is what I thought I had read.) Then when you are in GTK, you just select the brushes you wish to group together, then either right click and select funk/func_group, or, hit the N key and bring up the entities dodad, select func_group from the list, double click it.

Then to select it, hit the L key to bring up the entities list, expand func_group, select it, click select, and it will be selected for you to work with. (If you have multiple func_groups, you could give each a name or something to identify them, I just put "name" in the key feild and then call it whatever the "value" feild, Then you can identify them individually in the entities list (L key) window). I do not know why I can not just select the group by selecting 1 brush, but I do not seem to be able to do it that way

When you want to change it back you just select one of the brushes in the group/or the group itself, right click, select "ungroup entity" and the brushes will be changed back to worldspawn. Or you could use the java version of sleepwalkers map converter and do it as mentioned above. I use the old non java mapcon which does not have the -groups switch.

It is a lot quicker than it sounds, and besides that, it is the only way that I know, so until I find a better way, it will do for me. 

Thanks, much appreciated. 

is there any way to get faster compile with tyrlite ? takes way too long already (no -extra and only ~50% of the light are there)
I`m on slow pc ;/ 

Besides getting a faster PC and or having less lights or faces in your map I don't think there's a way... ;)

Well, if it takes that long you can always ask someone in #tf to compile it for you I guess, there are some people with quite fast machines, aren't there? 

ok, i need some dudes to beta test nesp09...

you`ll need a custom engine on this one, any one that has the edict limit bumped will do (PanQuake, TomazQuake). The engines that I know that don`t have bumped edicts are: FitzQuake and MHglquake.

i suggest the use of Panquake -- practically identical to the original glquake (just make sure pan is disabled).

if you need any help getting these, just ask.

so if your interested, email me:
necros -at- planetquake -dot- com

nb: serious people only please... 

Does anyone know why the upper part (there are two parts) of a Q1 custents func_spawn entity is visible before it is triggered ? It is then non-solid but otherwise works as expected when triggered (i.e. becomes fully visible and solid). 

it`s because the model is getting set, but the falltofloor part hasn`t been executed at this time.

not sure though, i`d have to check, but i know what you mean.

incidentally, is there any particular reason why sometimes when a post gets eaten, you can go back and it`s still there? (like now) but other times, the post is no longer in the box? 

incidentally, is there any particular reason why sometimes when a post gets eaten, you can go back and it`s still there? (like now) but other times, the post is no longer in the box?

We Fremen have a saying: metl created Func_Msgboard to test the faithful. One cannot go against the word of metl. 

Tell me of your home world asshole. 

lol. 

We Fremen have a saying: metl created Func_Msgboard to test the faithful. One cannot go against the word of metl.

Right Kell, but also have to remmeber the prophecy:
One will come, the voice from the coders world, bringing the holy bugfix, an addslashes() function, which will cleanse the bugginess and bring us out of darkness. 

...but YOU also have to REMEMBER the prophecy...

Ahhh, they're all over! Someone make the typos go away! Release me from this pain! 

Thanks for trying to help, although I still cannot figure out what is going on. I tried to follow your droptofloor (which I assumed you meant) lead, but there are too many things happening in the qc code for me to get the actual sequence of events.

What I do not get is why this particular two-part model does not work when a one part model like the soldier works fine. It is probably some error in the original qc code that has never been looked into and I am too inexperienced to locate the cause.

If you (or anyone else) figure out the cause of (and hopefully the cure for) this error, I would appreciate the information. 

what are you talking about? :)

i`m not sure what you mean by `two part model`. what makes a soldier.mdl a `one-part` model?