 News | Forum | People | FAQ | Links | Search | Register |
Log in |
|
|
I mustard mitt that a map jam along the lines of "set grid to 32, and no non-orthogonal angles allowed" could be quite the lol.
Hii i have a litle problem.
If i turn off RT World my lighting Map is too bright And if i turn on its ok. http://www.noelshack.com/2015-05-1422718655-2015-01-31-00001.jpg http://www.noelshack.com/2015-05-1422718653-2015-01-31-00002.jpg http://www.noelshack.com/2015-05-1422718653-2015-01-31-00003.jpg http://www.noelshack.com/2015-05-1422718656-2015-01-31-00004.jpg How to revise that. Thanks ok, I'm guessing those are external textures and way brighter than quake textures, so they looked all burned out under normal quake lighting?
Looks like the same textures in both shots
Fitz/ QS don't even have rtlights so maybe it's a problem with the actual light maps versus the rt approximation not matching. What is the purpose of modelheader->size in the
The engine multiplies it by this:
#define ALIAS_BASE_SIZE_RATIO (1.0 / 11.0) // normalizing factor so player model works out to about // 1 pixel per triangle GLQuake doesn't seem to use the resulting value at all; Quakespasm definitely doesn't. Software uses it in R_AliasCheckBBox: https://github.com/id-Software/Quake/blob/master/WinQuake/r_alias.c#L238 Maybe someone who knows the software renderer can tell what's going on there, I can't make any sense out of it! software rendering has two ways to draw a model.
1: either using spans 2: just subdividing each triangle and drawing a singleish point in the middle of it. method 1 is more correct. method 2 is faster when its further away. method 2 has the side effect of rounding the model outwards resulting in (most noticably players) being larger when distant. this makes them easier to spot than in gl (when running at the same res), and is imho one of the two reasons that people used software rendering over gl in quakeworld for so long. The engine needs some easy way to determine which method to use, hence that header field. I know synctype is almost entirely unsupported, and any static entities with an autoplayed framegroup will be visibly synchronized.
If you were to use a model with framegroups on a thinking entity, would the group cycle start at 0 when yes, I am actually wondering how viable it is to pack extra frames into an mdl, how did you know My next attempt to convert Q3 mdl files to Quake1.
This time I took the Dragonite model of Pior, a french artist that made the model. The hard part is to combine the three parts of the model back to one and import the base dxf into an editor. From there the trick is to give the model bones and animate them to the basic Quake1 frames. Only thing left is to give it a good flame throwing breach to attack the player, which I can only append to the Quoth mod. Wish I could expand the s_explode.spr better in qc. dragonit Okay this is wierd
Some times ago I was able to Import the nailgun moel into Blender thanks to that mdl script Now I want to mess with Q2's chaingun model, so I saved it as a Anyone got a solution to this? you could try using another model editor such as QuArK. Open the
|
|
| You must be logged in to post in this thread. |
|
| Website copyright © 2002-2024 John Fitzgibbons. All posts are copyright their respective authors. |