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
Probably... 
You probably could, I'm not organised enough to set that up first. The only problem with it is that there is already an info_notnull in the fgd, and it's a point entity, not a trigger. I'm not sure how well worldcraft would cope with two entities with the same name but different properties. You certainly wouldn't want to hardcode that function into the info_notnull, the best thing about them is that they can do all sorts of different things. Which reminds me, back to writing that up... 
:) 
yeah maybe one could differentiate if the other is a point entity and the other a brush, dunno, maybe i'll look into it some year... 
Regarding The Entities, Preach 
Thanks for the tip about this. I've used it now for a targeted trigger_changelevel by simply setting the .use key to trigger_changelevel, which works because trigger_changelevel() doesn't precache anything.
Worldcraft does NOT like it if you put a definition for both a solid and a point entity with the same classname in the fgd. You'll have to make two entries for it, one being solid and called info_notnull2 or something.

However if you're using aguirre's tools, you can automate the process a bit by setting up a batch file like this:
updateents.bat
@echo off
bspinfo -copyents %1
echo Please go ahead and edit %1.ent
pause
qbsp -onlyents %1.ent
@echo on

And then run it like updateents mymap, edit the generated .ent file in your text editor of choice and then go back to the bat file and "press any key to continue...".
If you've got the knowhow to do it you could probably have a commandline tool to do a regex search on the generated ent file automatically and replace all info_notnull2 with info_notnull. You know, something like grep or the likes. 
Actually After Thinking About It For 10 Seconds More 
It would perhaps be best if you didn't have any info_notnull entries at all in the fgd file.
WC will not complain if you type in your own classname in the entity properties, and then doesn't care if the entity is solid or not. The only drawback is you won't get the benefit of SmartEdit, but for info_notnulls there are so many different fields to be set on them depending on what you plan to use them for that using SmartEdit would just be a clutter. 
The Grep-like Tool 
is often called sed (stream-edit) and for the simple entity name replacement, you don't even need regex (which can be a bit tricky to set up). 
Preach.. Which Function? 
and finally add

self.frame = 198;

to the bottom of the function. And that's all...


Which function? 
Nevermind 
I missed the bit about copying the function and renaming it, got it now. 
Preach Again... 
Well, I looked over everything and I'm certain I did things right. You forgot to mention I need to work with the ZOMBIE.MDL by putting it in a progs directory.

So, in the map - the zombie is completely stationary and not in crucified pose. Yet he moans like he does when in that pose. Also, I triggered him and he didn't move to any path_corners I set. I even removed the trigger_once and walked up to him, but nothing happened either. 
CrAzYCoW... 
...if you want a hack for multiple explosions using one button then tie sequentially named func_explosions (fe1, fe2 etc.) to trigger_counters with increasing count numbers but the same name. 
OK... 
...that might've been a bit confusing.

Make the launch button target a series of trigger_counters named "missile", but give each trigger_counter a higher count value than the last. Now make each trigger_counter target a different func_explosion... to make things easy, get the trigger_counter with count 1 target the func_explosion with the name fe1, trigger_counter with count 2 targets the func_explosion with name fe2 etc. 
Moaning 
The misc_zombietrain shouldn't moan at all, unless that's some bit of code you've added. There are no sounds that automatically play with certain animations, and the misc_zombietrain doesn't even precache them, so they'd only be loaded if you had other zombies in the level. Could you make a quick testmap with just a misc_zombietrain and a few path corners, and see if you get the same result? 
Preach 
What I originally didn't realize I needed to do, was rename the monster_zombie to misc_zombietrain and set the fields. I just tried and it works well.

Though, is there anyway to have him in his crucified, animated pose while retaining his train functionality? Or at least how might I have him stationary but in the crucified pose? (He's just stood straight up). 
Animation 
I may not have been clear enough in my explanation of how to edit the zombie model. You need to duplicate the crucified animation, and only make the second, copied version of it into a framegroup. If you do that, the framegroup should be frame 198 of the zombie model, and so the line that sets the zombietrain's frame to 198 will make it animated while crucified. I'm guessing that the problem is your modified zombie doesn't have a frame 198, so it loops back to frame 0. The first crucified animation should remain as individual frames, as the normal crucified zombies need those frames to be seperate. 
Yep 
I'm pretty sure that's how I did it, but I'll try it again... 
Ok 
Is this what you mean? Cause if I do it how I understand it, the total frames goes to 204, not 198:

http://www.phait-accompli.com/q/s4/pre/qme1.jpg 
Two Things 
One is that the total framecount by QMe counts all the frames in a framegroup seperately, so the last frame in the grouped crucifixion is number 204 according to QMe, but that's not how the quake engine will see it. Since the last frame is 204, the first frame should be 198. Since quake treats all the frames in a framegroup as one frame, they will all have the number of the first frame, 198.

Except....There are four other animations that are grouped in your copy of the model, painb through to paine. Each one of these sequences was 10 to 20 frames long in the original model. However, since they are framegroups now, they are each only one frame long. Which brings the total count of frames as quake sees it to about 140.

If you're using some other mod which requires them to be grouped, then you're going to have to work out which frame the grouped crucified animation is, and use that value instead of 198. Otherwise, ungroup them and it should work out.

Oh, and one more thing I should have said before but forgot, rename the second crucified sequence, it's possible it could cause a problem. I really should have put that in the first post, as it's not obvious you need to do it. What usually happens is the two sequences with the same name end up merged into one, but if one is grouped and one is ungrouped it may not happen. Still, best not to take chances and give it a different name. 
So Yeah, An Updating Batch File Using Sed: 
bspinfo -copyents %1
sed s/info_notnull2/info_notnull/ %1.ent > %1.edited.ent
copy %1.edited.ent %1.ent /Y
del %1.edited.ent
qbsp -onlyents %1.ent


The stupid copying thing is because I couldn't figure out a 'overwrite old file' switch for the rename command, and if you make sed output to to same file you input from you get a blank file and if you mke sed append to the same file you input from you get a doubled file. Not foolproof, but it works.
Get sed here: http://www.cornerstonemag.com/sed/ 
Thanks CZG 
Should prove to be useful.

On an entirely unrelated not, I'm sure I remember once seeing a compiler for Q1 that supported a nodraw texture. But now I can't find it, did I just imagine it? If you know where it went please let me know, it would be handy for a mod I'm doing. 
Preach 
That Was The One 
Thanks! 
Weird Fish Behaviour 
Is there any obvious reason why performing a QC "monster in wall" check (i.e. !walkmove(0,0)) for fish should cause them sometimes getting moved way outside the entire Q1 addressable grid, e.g. (100000 10 5)? I've used this extra check for some time in vanilla Q1 without any obvious problems, but doing the same in Nehahra seems to be bad.

In this particular map (neh2m5), fish often swim through walls both with and without the extra check, but it appears as if the check sometimes cause them to really move outside the grid to completely invalid coordinates making them impossible to kill.

Oh, and has anybody been able to run the Nehahra QC build tools in WinXP? I can only run them in Win95, they are some Borland DPMI apps. 
CZG 
If you delete the original ent file first, you can rename the new one without problems. It's the same number of batch lines, but it's slightly faster as there's no data movement. Not that it matters much with small ent files though ... 
Quake Demo Editing 
I can't seem to find a working link for "Remaic" - a Quake demo editor allowing camera setups and different perspectives, etc. Might anyone have Remaic Studio, or know a working mirror? 
Perl, Win32 Binary, C++, We Have Them All! 
Ah Thank You 
:) 
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.