This a theory I wanted to try out. It attaches the players view in front of the head, so there would only be the playermodel required, no viewmodels. I saw this in a couple of games, and I think more games should use it.
Yes, it has a new model, so it's not just the old ranger model, which wouldn't give the point across properly.
No, it's not a good model. It is only a test model, to prove my point.
Please read the readme first, for a couple of recommended cvars.
I also forgot to mention that cl_stairsmoothspeed 1000 or above is recommended, for it to not feel glitchy as you walk on stairs.
The code is included for anyone who wants to see, it was based on an old version of a mod called RemaQe's source, which was based on the scratch tutorials. It's not anything clean whatsoever, I just wanted to see if this would work, and it did.
It was made for DarkPlaces, but it may work in other engines supporting md3 models, namely FTEquake.
Big thanks to LordHavoc for hosting this file.
http://ghdigital.com/~havoc/hosted/Urre.supertorso.rar
Theory in action
Theory in action
I was once a Quake modder
Argh?!
http://ghdigital.com/~havoc/hosted/Urre.supertorso.rar wrote:Access forbidden!
You don't have permission to access the requested object. It is either read-protected or not readable by the server.
If you think this is a server error, please contact the webmaster.
Error 403
ghdigital.com
Tue Nov 23 13:41:02 2004
Apache/2.0.49 (Linux/SuSE)
Improve Quaddicted, send me a pull request: https://github.com/SpiritQuaddicted/Quaddicted-reviews
Urre:
Very nice. I agree that this is something that should be done much more often. It gives a more correct feel, looks better (provided you have a decent model, heh), and really simplifies things because you only have one model for each weapon (and weapons are visual to the other players as you see them).
But that model you used needs to be put to death now. And I don't even mean mercifully.
Personally, I think the theory is a good one.
Very nice. I agree that this is something that should be done much more often. It gives a more correct feel, looks better (provided you have a decent model, heh), and really simplifies things because you only have one model for each weapon (and weapons are visual to the other players as you see them).
But that model you used needs to be put to death now. And I don't even mean mercifully.
Personally, I think the theory is a good one.
When my computer inevitably explodes and kills me, my cat inherits everything I own. He may be the only one capable of continuing my work.
Yeah, the codebase was untested before I added the playermodel handling, I have no idea what's in there. The purpose was only to show this theory.Sajt wrote:...Also, there seams to be some weirdness with falling damage or something, because I took 38 damage from running into the start.bsp torch stands funny
It has it's ups and downs. Some say it's just generally glitchy, unnecessary and ugly. I personally think it's just generally cool to see things exactly as it would be seen from the characters eyes. Seeing the legs as you look down, the arms, and all that. Also with all the realtime lighting stuff going on, I think it's cool to see the shadow match the stuff you see in your view.
The downsides are the obviously higher use of entities. This one is six entities, could be five. It is currently client, legs, torso, arms, head and camera. The legs are attached to the client entity. The camera is attached to the "eyes" tag on the head model, and the client is using SVC_SETVIEW on the camera entity. It could be done so, that the playermodel is a dummy separate from the client entity, and the client is simply attached to the eyes tag, no camera entity used. That would mean five entities instead of six. I did not do this because DarkPlaces currently has no option for separating orientation on tagged entities from their parent, meaning that any angle changes would be relative to the parent that the view is attached to. This would cause some funky stuff to happen if you simply tried to turn or look up/down. Adding a flag for separating those things is on the DarkPlaces todolist. I'm also not sure how I would have handled the movement in such a case, though I'm sure I would've figured it out soon enough.
So in general, this does require a bit more work to look good, but once that work is done, I think it looks so good that it beats the traditional viewmodel way any day.
And yes Wazat, I shall brutally slaughter the model now.
I was once a Quake modder
Oh yeah
I should also mention another downside, which may be obvious to some, which is that this doesn't support polygon hit detection on the model, because of the tagged models. Might not seem as a huge loss considering that it doesn't work on any tagged model setup, it's simply a limitation of the md3 tag system. I've been discussing it with LordHavoc, and it seems that the main issue is collisions from the tagged entity, with all it's rotations and all, rather than against it. I'm personally only after being able to do polygon hit detection against it, for bullets and stuff, don't really care about other things. Don't see much of a reason for it to collide with the world or anything.
I was once a Quake modder