Posts by Spherix

Fuck me that's pretty. And new trees.. Oh my.

As for making it less clean, I don't see any (detail) props, overlays, decals etc. Trashbins with trash around them, stains on the floor, etc. etc.

Correct, one of the many things I'm addressing over the week , haha

I've all the texture/lighting issues sorted, even a LDR compile is no longer an issue Had to align hundreds of surfaces, also made some alternative VMT's without the normalmap to match the rest of the wall segment textures. I've let it run a full HDR+LDR compile in one go, took a good 13 hours only to find out I made an error with an areaportal; yay.

Adding a little extra holiday flair now I have to recompile anyway, this is it's current state:

[Blocked Image: http://i.imgur.com/uGcRM7Z.jpg]

[Blocked Image: http://i.imgur.com/n9rsIgK.jpg]

More here; http://imgur.com/a/QRATj

All I need now is a modeler to help me with the arctic avengers as the version I have of them is incomplete and therefore buggy

I'm fixing them..I must've altered over a 100 surfaces already.. It's a mess. They also have a lot of _bottom textures without a normal map, and then the top part has one, which makes the oddities shown in the other screenshot I've posted before. Might be intentional, but it sure looks odd when you take the time to look at the map properly.

Quote from Spherix

How would I get rid of these lighting errors? Compile options used for light_exe: -ldr -final -textureshadows -StaticPropLighting -StaticPropPolys -game $gamedir $path$file [Blocked Image: http://i.imgur.com/9ZqkMj1.jpg]

So... the official version of de_dust2 is full of texture sloppyness... My lighting and compile settings just made them more obvious.

[Blocked Image: http://i.imgur.com/12NdVhv.jpg]

More found in 30 seconds: http://imgur.com/a/cKBr9

Decompiling doesn't actually read a func_instance, since it's contents are incorporated into the same file upon compiling. Instead, the contents of the instance are in the normal vmf file without losing data.

I've also checked a number of brushes' coords in the raw vmf to see if there were any rounding errors due to the decompile, that doesn't seem to be the case.

What exactly do you mean by collision meshes in the 3d skybox? It's unchanged from the original as well.

The second image seems to have a different scale for those texture for some reason (probably the -ohnoyoudidn't- decompile that caused that). So that should be an easy fix.

The top building however, the scales, coords etc. all match up. They have a lightmap scale of 8; the textures are stock and I haven't changed those from the original dust2. All the instances should be fine, but I'll have a check once I get home.

How would I get rid of these lighting errors? Compile options used for light_exe: -ldr -final -textureshadows -StaticPropLighting -StaticPropPolys -game $gamedir $path$file [Blocked Image: http://i.imgur.com/2Xfju9f.jpg][Blocked Image: http://i.imgur.com/9ZqkMj1.jpg]

Ahh how could I have forgotten, Destroyed village from MOHAA was one of the maps I surely have spent most teenage FPS hours in aside from CS... I actually attempted to recreate it in CS:GO as a method of practice to get back in to mapping after years of absence, only going by reference screenshots until I later found out I could just export the brushes and scale them down to CS scales to use as perfect reference Abandonned it since it's not viable for cs:go gameplay anyway.

Found this progress screenshot

[Blocked Image: http://i47.tinypic.com/11jpsnl.jpg]

Maps like as_crazytank for sure. Those maps took the seriousness out of the game and added fun for me;

[Blocked Image: http://csarmy.ru/img/maps/as/as_crazytank3.jpg]

I got carried away.

[Blocked Image: http://i.imgur.com/IR683Mi.jpg]

It was linked to from Reddit with every map hype.

Quote from Nakroma

You are the func_vehicle guy? Holy shit, nice work.

What I don't quite get, how are you supposed to play? With knifes it doesnt work very well, so we gave us mp9s each round

Yarr. As for supposed to play: It was/is a knife map. Knife, push people in the toxic water, or run them over. There is a 'hidden' weapons room though.

It's ka_roadwars :>

Not really WIP anymore but I made silly stuff:

The map ain't restarted fully afaik, I've noticed that when I was testing a map where I used an env_texturetoggle so change a scoreboard; a round end kept the texture as it was last set.

Managed to do some small tests and it seems to be either broken or just not parsing the team ID's at all. The game event is always round_end. I noticed a round_officially_ended as well (which fires a splitsecond before the round is actually restarted), but a logic_eventlistener didn't even pick those up.

What you could do after the generic round_end is detected, is do a count of the amount of players per team left. That should be done quick enough for the script to figure out which team should be awarded points as it has or hasn't got any players left.

One of the options that comes to mind is to have a script constantly check for the presence of a CT/T on the map; as soon as the result is 0 for either one of them, that means the opposite team would still have a member left.

For example;

while (Entities.FindByModel(variable, "models/player/tm_phoenix.mdl") != null) {
  printl("Terrorists are still alive");
}

If that returns null, then all the terrorists should be dead; in which the CT's have won the round.

I'm not sure if going by models alone is doing the trick though, dead bodies could share the shame model and that would cause issues. Another way is to check for player team ID's and health, and go by that.

All in all this feels a bit too dirty, I was pretty sure the eventlistener should do the trick. I'll do some testing when I get home to see what is causing the issue.

Edit: Another entity-based option would be to have a trigger_multiple over your entire map with a teamfilter, and have it trigger a script event as soon as nobody from that team is touching it anymore (meaning they're dead).

Edit2 (Solution): Was able to do a small test during my lunch break, it seems to work fine after I killed the last bot outside of warmup.

Firing: (game_playerdie)
Firing: (game_playerkill)
(22.92) output: (NULL,NULL) -> (round_end_script,RunScriptCode,0.0)(RoundEnded())
+++++Bot Keith with Max Difficulty 1 will become Bot Chad with Max Difficulty 2
+++++Bot Ron with Max Difficulty 1 will become Bot Joe with Max Difficulty 2
+++++Bot Ian with Max Difficulty 1 will become Bot Xavier with Max Difficulty 2
Achievements disabled: cheats turned on in this app session.
Achievements disabled, ignoring achievement progress for SILENT_WIN
(22.94) input <NULL>: round_end_script.RunScriptCode(RoundEnded())
======================ROUND ENDED==========================

The ==ROUND ENDED== is the Vscript outputting a line after the logic_eventlistener has been triggered by round_end.

Attached the VMF and .NUT files for the heck of it, you'll notice I haven't done anything strange.

round_end.zip

Do the players spawn;

- once

- in waves

- instant respawn (deathmatch)

Does the round end when:

- All players on a certain team have died

- The time has run out

- An objective has been achieved

Check which apply so I know what the options are

I read/replied to your question on Facepunch; if you would be so kind as to explain what gamemode and such you have in mind, then I can probably write the Vscript needed for you. A simple snippet won't be helpful if I have no idea how you plan on ending rounds.