Reply to topic  [ 26 posts ]  Go to page Previous  1, 2
Anyone do source modeling? 
Author Message
User avatar

Joined: Tue May 08, 2007 1:08 pm
Posts: 1045
Post 
no response from Thoth cause his bluff got called ^__~

[edit]
C4k3 do you really need c4k pics? Just send the files already
[/edit]

_________________
Image


Last edited by That Annoying Kid on Thu Aug 19, 2010 4:00 pm, edited 1 time in total.



Thu Aug 19, 2010 1:36 pm
Profile
Site Admin
User avatar

Joined: Tue Feb 20, 2007 12:50 am
Posts: 1415
Post 
:P

VNC'ing the California rig from London is not the most dependable way to keep on the internet. >_>

Quote:
My question to modelers -- How far does the game need to be before you'll help me fix the animation/hitbox/collision hull issues with *1* ported HL1 model?

Thought I made that clear when I said playable with placeholders. Seriously though, how is 1 ported model holding up an entire mod for two years? I assumed ye had no models at all, and even then, that shouldn't stop you. But one? :\ That, just doesn't happen. Hell, I can fix it myself, if it's just one, and my modeling skills are pretty half-assed. (Usually more of the convert and process / QC editing type of thang.)

You've got a map (IDK if it's original, or if you have some magic wand up your sleeve that allowed you to port it that I wished to hell I had - or the Tanith source), you've managed to import models... So... What could possibly be stopping you from having a working alpha, but code?

I am skeptical. I've just seen this sort of thing too many times, and I've seen too much talent wasted. (Too many checkerboarded hopes, as it were.) I suppose I shouldn't complain, as I am the f*cking lord and master of cannibalizing artwork from dead mods for use in other mods... However... The four modelers I have access to (FER, Gaz, Orochi, and Karoda) are all working on good games that are either already out or going somewhere (MSC, MSS, PVKII, and a sekrat Chinese Room project)... And modelers really aren't that hard to come by anyways. They're a dime a dozen compared to coders. If the coder can't come through, they can't do anything for you. You need a functioning world for them to put their stuff in.

Link up the model in question with a description of the issue, and I'll either fix it myself or get it to someone who can. You may get a bite. Link up a working alpha, and I'll fraps it, and send it to everyone I know who can model. You'll have an army.
_________________
"Order! Order! Do you kids want to be like the real UN? Or do you just want to squabble and waste time?" -Principal Skinner
Thothie


Thu Aug 19, 2010 2:34 pm
Profile YIM WWW
User avatar

Joined: Tue May 08, 2007 1:08 pm
Posts: 1045
Post 
nice quote

_________________
Image


Thu Aug 19, 2010 4:04 pm
Profile

Joined: Tue Feb 27, 2007 8:30 am
Posts: 78
Post 
zomg, PvK2


Thu Aug 19, 2010 4:16 pm
Profile
User avatar

Joined: Fri Jan 04, 2008 7:08 pm
Posts: 324
Post 
That Annoying Kid wrote:
no response from Thoth cause his bluff got called ^__~

[edit]
C4k3 do you really need c4k pics? Just send the files already
[/edit]


Image

Got it, working again. (not)

/edit, nevermind my fucking wife is screaming, brb real life. Gook luck with your army of modelers. I'll be there killing shit when done.


Thu Aug 19, 2010 5:08 pm
Profile
User avatar

Joined: Tue May 08, 2007 1:08 pm
Posts: 1045
Post 
L337 K4mP3r Kr3W wrote:
zomg, PvK2


That is also another thing that ran thru my mind when I read Tothies post


[edit]
it's a shame they nerfed the damn keg
[/edit]

_________________
Image


Thu Aug 19, 2010 11:23 pm
Profile

Joined: Wed Aug 18, 2010 2:41 am
Posts: 2
Post 
While I would love an army of modelers to do custom models, my first priorty is the same as it has been before I started on the project -- get 1 fully functional skulk model so I can test the Source engine to see if the game can even run properly on it before I waste the many hours (that I already have) writing game code. I had given up on the original modeler and decompiled the NS1 models and recompiled for Source. The problem is all .QC commands (afaik). If I can get them fixed for 1 model, then I can apply the same to the rest of the models. The problems with the models are as follows:

1) No collision hulls (I read that I can use the model itself as a collision hull, but haven't been able to figure out the proper .QC commands to make this happen)

2) No hitboxes (again, the hitboxes are there, but they need defined in the .QC file and I don't know how to do it properly)

3) Animations are bugged (structure animations are fine, but player animations are quite strange. I believe the problem is related to Source's 9-way blending, which again is a .QC issue)

My test with respect to the Source engine is simple -- test the absolute basic game play to see if Source can handle the fast moving players (in terms of lag compensation and prediction). The basic game play is celerity+leap+wall walking skulks vs lmg+shotgun+pistol+resupply marines. The wall walking works, but needs improvement, which can't be done without collision hulls. I didn't want to work on any game code beyond this unti I was able to verify Source is capable of handling the game, but due to lack of modeler I have gone far past this point against my own better judgement.

What does it mean to be an Alpha or playable? UWE released an NS2 Alpha that isn't playable imo. So what features need to be implemented in order for it to be considered playable (and subsequently worth your time)? Most of the game functionality is present, but without a commander interface it's not playable. I guess I could spend a bit of time and make a combat game mode, which wouldn't need the comm.

The original, decompiled, and Source skulk models are available here: http://evolvens.com/temp/skulk.zip


Sat Aug 21, 2010 11:15 am
Profile
User avatar

Joined: Mon Feb 19, 2007 8:51 pm
Posts: 1011
Location: Texas
Post 
or we could just wait for ns2? Not to troll, but why is this even being made? Will this be the fortress forever of NS2?


Sat Aug 21, 2010 3:38 pm
Profile
Site Admin
User avatar

Joined: Tue Feb 20, 2007 12:50 am
Posts: 1415
Post 
TheDoc wrote:
or we could just wait for ns2? Not to troll, but why is this even being made? Will this be the fortress forever of NS2?

Fortress Forever coulda saved TFC for me, if it hadn't decided to go out NeoTF style in the end. *sigh* Was a shame to see TFC die the way it did with no proper successor.

I better not see a repeat of said in NS2, but so far it doesn't seem likely. Though I've not hardly had a chance to play.

Quote:
1) No collision hulls (I read that I can use the model itself as a collision hull, but haven't been able to figure out the proper .QC commands to make this happen)

2) No hitboxes (again, the hitboxes are there, but they need defined in the .QC file and I don't know how to do it properly)

3) Animations are bugged (structure animations are fine, but player animations are quite strange. I believe the problem is related to Source's 9-way blending, which again is a .QC issue)

1) The phys-model determines collision hulls in Source. I *think* you can just define it in the QC as the reference model, and it'll have barebones functionality. Then just give it a general definition in the QC. eg. (from the headcrab-spider thing for another Source mod I just happen to have open):
Code:
$collisionjoints "phymodel.smd" {
   $mass 5.0
   $inertia 5.00
   $damping 0.01
   $rotdamping 1.50
   $rootbone "headcrabclassic.bodycontrol"

   $jointconstrain "headcrabclassic.thighl_bone" x limit -37.00 12.00 0.04
   $jointconstrain "headcrabclassic.thighl_bone" y limit -39.00 46.00 0.04
   $jointconstrain "headcrabclassic.thighl_bone" z limit -44.00 46.00 0.04

   $jointconstrain "headcrabclassic.thighr_bone" x limit -30.00 19.00 0.04
   $jointconstrain "headcrabclassic.thighr_bone" y limit -39.00 47.00 0.04
   $jointconstrain "headcrabclassic.thighr_bone" z limit -44.00 46.00 0.04

   $jointconstrain "headcrabclassic.calfr_bone" x limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.calfr_bone" y limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.calfr_bone" z limit -77.00 57.00 0.04

   $jointconstrain "headcrabclassic.upperarmr_bone" x limit -37.00 64.00 0.04
   $jointconstrain "headcrabclassic.upperarmr_bone" y limit -53.00 23.00 0.04
   $jointconstrain "headcrabclassic.upperarmr_bone" z limit -61.00 55.00 0.04

   $jointconstrain "headcrabclassic.forearmr_bone" x limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.forearmr_bone" y limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.forearmr_bone" z limit -84.00 39.00 0.04

   $jointconstrain "headcrabclassic.upperarml_bone" x limit -64.00 35.00 0.04
   $jointconstrain "headcrabclassic.upperarml_bone" y limit -23.00 39.00 0.04
   $jointconstrain "headcrabclassic.upperarml_bone" z limit -44.00 53.00 0.04

   $jointconstrain "headcrabclassic.forearml_bone" x limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.forearml_bone" y limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.forearml_bone" z limit -70.00 48.00 0.04

   $jointconstrain "headcrabclassic.calfl_bone" x limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.calfl_bone" y limit 0.00 0.00 0.00
   $jointconstrain "headcrabclassic.calfl_bone" z limit -95.00 57.00 0.04

[yadda... yadda...]
}


2) Hitboxes are handled with almost the exact same formatting used in GoldSrc. Only difference is you have to define a hitbox set. Eg.
Code:
$hboxset "default"
$hbox 0 "HeadcrabClassic.BodyControl" -3.500  -6.000  -7.170  13.500  4.000  7.170
$hbox 0 "HeadcrabClassic.ForeArmL_Bone" 0.000  -0.500  -1.000  8.000  1.500  1.000
$hbox 0 "HeadcrabClassic.UpperArmL_Bone" -0.500  -1.000  -1.000  8.500  1.000  1.000
$hbox 0 "HeadcrabClassic.UpperArmR_Bone" -0.500  -1.000  -1.000  8.500  1.000  1.000
$hbox 0 "HeadcrabClassic.ForeArmR_Bone" 0.000  -1.500  -1.000  8.000  0.500  1.000
$hbox 0 "HeadcrabClassic.CalfR_Bone" -0.250  -0.500  -1.000  4.250  1.500  1.000
$hbox 0 "HeadcrabClassic.CalfL_Bone" -0.250  -1.000  -1.000  4.250  1.000  1.000


Vs. Gold Src:
Code:
$hbox 3 "Bip01 Pelvis" -9.360000 -4.000000 -6.540000  5.050000 8.860000 6.560000
$hbox 6 "Bip01 L Thigh" 0.000000 -4.680000 -4.220000  16.680000 6.370000 2.020000
$hbox 6 "Bip01 L Calf" 0.000000 -2.550000 -1.620000  14.200000 1.380000 1.450000
$hbox 6 "Bip01 L Foot" 0.000000 0.000000 -0.940000  5.140000 7.114834 1.620000
$hbox 7 "Bip01 R Thigh" 0.000000 -4.690000 -2.030000  16.680000 6.370000 4.200000

Same diff. You just need to match any bone name changes your conversion may have caused. Beyond that, it's copypasta.

3) Not sure why the animations would be bugged save for the orientation is different (but I'm sure you know that - player models measure from center, everything else measures from ground), and then there's various possible controller assignment issues.

Source QC commands:
http://developer.valvesoftware.com/wiki ... C_Commands

I'll forward this model onto a few folks with your concerns, assuming none of that helps.

Ya could define temporary hit/collision boxes code side via the root joint or entity location/angles, however, so you at least have something fraps'able to show. Tis what we wound up doing on an older mod, although, downside in that case was that it sorta became half-permanent for the melee system the game used, causing us more problems down the road.
_________________
"You're only given a little spark of madness. You mustn't lose it."
Thothie


Sat Aug 21, 2010 4:54 pm
Profile YIM WWW
User avatar

Joined: Tue May 08, 2007 1:08 pm
Posts: 1045
Post 
what do you think of this approach thothies

Quote:
My test with respect to the Source engine is simple -- test the absolute basic game play to see if Source can handle the fast moving players (in terms of lag compensation and prediction). The basic game play is celerity+leap+wall walking skulks vs lmg+shotgun+pistol+resupply marines. The wall walking works, but needs improvement, which can't be done without collision hulls. I didn't want to work on any game code beyond this unti I was able to verify Source is capable of handling the game, but due to lack of modeler I have gone far past this point against my own better judgement.

_________________
Image


Thu Aug 26, 2010 11:53 am
Profile
Site Admin
User avatar

Joined: Tue Feb 20, 2007 12:50 am
Posts: 1415
Post 
Eh... Yeah... Read that.... Tis *why* I told him how to make hitboxes and collision hulls. :?

A collision hull made with the ref model is not going to be *optimal*, because it's more complicated than a collision hull should be (may also get some non-convex error warnings), however, as HL1 models are much lower poly than Source models, you should be good until you get more optimal one. The hitboxes should be good as gold from the getgo though.

TL;DR: *smack* That's what I just responded to!

I got one bite from one of my modelers, but he needs to wait until the end of the month as his current mod's team is aiming for a patch on Sept 1st, and he's gotta get a bunch of p_models done before then. ...That is, assuming you've not already resolved your problem with the info I gave you here.
_________________
"All patterned behavior tends to go unquestioned gathering destructive momentum." –Darwi Odrade, Dune
Thothie


Thu Aug 26, 2010 12:15 pm
Profile YIM WWW
Display posts from previous:  Sort by  
Reply to topic   [ 26 posts ]  Go to page Previous  1, 2

Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software.