Nosslak
-
Posts
679 -
Joined
-
Last visited
-
Days Won
3
Posts posted by Nosslak
-
-
Great art! Very inspirational.
If you want something to do you could make a zombie concept for me as we need to replace the existing ones with our own. It would need to be somewhat similar in proportions, naked and somewhat decomposed. If you do make one I'll need at least an orthographic front-view and preferably a side-view as well.
The "planning" pic is my favourite by far - ones like that tell a story more than even the most elaborate cityscape I feel <3
That's my favorite as well. When I were making the plague doctor mask I considered making a similar hat but wasn't sure if it'd look good. If there's interest I can try to make one after the zombie.
-
Looks great!
That's a rug, so unless it has been in place long, it could look shiny new.
In that case it should probably have a trim and if possible it should be like a short cylinder so it'll have some height as well.
-
It's the model, the rigging, and the animations that are the hard part. Well, that's the problem...the number of team-members who have the skill to do good quality animations is quite low. I wouldn't be able to do a good enough job even if I wanted to.
I am really terrible at animation so I won't even attempt that, but I guess I can try to rig it with Blender. I would need to know the technical limitations of the engine before I do that (can it handle IKs, constraints, wight-paint, should the eyes be separate, etc.). Also is it possible to export rigs from Blender so that other people can open them with Maya, Max, Lightwave or whatever and do the animations?
The number of people who have offered to make zombie models in the past (and never delivered) is also somewhat disheartening.I'm quite busy in school so I don't know how soon I'll be able to finish it (might not be until summer in the worst case scenario).
Well, you could use the existing ones as a reference, as they would have to be pretty close to that. I'm personally a fan of the Walking Dead zombies. http://blogs.amctv.c...hirt-zombie.phpI haven't played all that many missions (even less with zombies), but in the ones I have played they haven't been bloody or anything. Should I take this as you want zombies that are a bit more gruesome? Also do anyone have any good shots of the current zombies or will I have to fix some myself?
-
Well, we're currently using the Doom3 zombie and skeleton, along with the 20+ animations that come with them. That would be a good start, since at least a third of our missions use them.
I'm no expert at making character (here's my latest attempt, a mudokon) but I guess I could give it a try. Someone else would need to do the rigging/animations though as I'm pretty bad at it and I'm not sure how well I can export anims from Blender and then pass them on to other members (and I won't do it in Maya as I despise that program). Some kind of concept art would be appreciated but not necessary if there is none.
-
I don't think a for is warranted - let's first see if we can replace the obviosu things. If v1.08 comes close to 95% of D3 free than that is a huge step - replacing a lot of things and then whittling the list of things down to a manageble version.
Rhere is no sense in debating a full fork (which will break a few missions) when people can't even do the work in replacing art assets. A for is VERY complicated to maintain - and who is gonna maintain that if the team+community (no offense!) can't even go ahead and replace dds/guis/assets/white.dds?
If we have come along to the 95%, then you can always still do the fork if you really want. But my guess is there will be a lot of "debating" in the forum (as there is always), but the actual work will be done by the usual few suspescts like Springheel, Arcturus, greebo (and so on).
Now is the time to step up and help
If someone compiles a list of the needed art assets I am willing to do some of the work (I'd mainly do some models and textures).
-
I'm not a native english speaker but you're welcome to help
.I'm not a particularly skilled writer either nor a native speaker so I think you're better off getting help from someone else.
-
The screenshot looks extremely washed out (otherwise it looks good) and the briefing is littered with typos and grammatical errors. Also the briefing doesn't sound like something a Victorian/Edwardian person would write, mainly because of the references to games (hunting game, game called life) and shooting fish in barrels (there are no guns in TDM).
Sorry if I sound harsh, but I feel that the briefing could be improved a lot without changing the meaning of it.
-
Thanks, I attached it below, just rename it to .py.
-
Hey Noss,
Just took a look at the ASE's, the exporter is using a single material and then multiple sub-materials. This is rather messy, while some games support sub materials, others don't. Generally they are avoided as they can become a pain - as they were a very late addition to the already loosely bound format. The exporter also does not correctly assign references to the material - something that is likely a result of the developer only supporting one game, and that game assuming that it will always use submaterials in some order, urgh.
I'd take a look at the exporter's options, there might be something related to sub-materials.
Anyhoo, just fixing up some other stuff and then I'll make the edits by hand.
Alright, I think I'll keep using the .ASE exporter for Blender 2.49 instead in that case. It's not really a big deal to export it from 2.49 but I hoped I could eliminate that extra step.
-
I'll have to give Blender another shot. I've heard the interface is much more intuitive these days.
Yeah, you really should! The basics haven't changed all that much though, so if you were unhappy with the hotkey-based interface chances are that you'll not like it this time around either.
Yes I have got 3ds Max 2012 and you can not import .ase files. You can export .ase files though. I don't get it why.. This was the case in 2011 too, but there were plugins available (which aren't available for 2012). So I can export models but can't import them which is annoying since you could export entire rooms from Dark Radiant for 3ds max. You could but you can't ^^
I think the same problem's present in Blender. I haven't looked very hard but I haven't been able to find any script to import .ASE either.
-
Hey Noss,
Sorry — I've been house-sitting for some people, laptop is limited to playing terrible 2D RPG's and watching stuff, so I kinda forgot to take a look at your stuff!
Do you still need me to take a look?
Yes, please do! I hadn't imported anything to TDM for a pretty long while when I tried importing this model so I might be doing something wrong, but to me it all looked right. I were using a new exporter that works right out of Blender 2.61 and I think that might be the reason for the problems, but if possible I'd like to use that exporter rather than the 2.49 one as I don't need to launch a new program.
-
I'd say so.
One thing that you may be having an issue with is that Doom 3's reliance on normal maps has two troubling consequences:
1) Since they eat Gallons of texture memory, your budget for unique textures is lower and hence more repetition
2) With older static lit engines, you could use hand-drawn "bridge textures" to add nuance and detail to material transitions...
....with normal maps these transitions would need to be made as high-poly models and baked to a model that is a copy of the brush geometry (unless you've got an eye from drawing normals by hand...)
Problem 2 is something that I've been considering awhile now. The answer to how "solvable" it is comes down to:
1) Is there a tool that can convert Normal Maps back to geometry (Oneofthe8devils had a doom3world thread that implied there was...) ?
2) Might it be possible to modify the "renderbump" command to work with brushes and patches?
3) Are there any tools that can convert normal mapped brush prefabs into "virtual high-poly models" in real-time to "sculpt normals"
Blender plugins (etc).
I'm surprised that such solutions would never have been discussed before given how widespread the use of Normal Maps are but
I guess that since UE3 is heavily biased towards Modelers (as the core content creators) that the topic has not really been broached that
much by the industry...
If DR ever gets the ability to blend between textures using vertex colors on patches and brushes (as opposed to only models with ready-made vertex colors) this article from the Amnesia guys might be worth a look:
http://frictionalgam...n-textures.html
About 1/3 down the page (close to the articles end) they discuss a nice way to blend texture that looks a lot better than the regular linear interpolation. Their tecnique requires that the textures that are to be interpreted between will need an additional alpha so there would be an initial extra effort for that but I think it'd be worth it.
-
Would you use the existing skeleton in this case or would you rather use the motion capturing variant, if there comes out some cool concept art that enspires the modelers?
I'm pretty bad at rigging and animating (have barely done any of them) so I'm definitely no authority on this subject, but I'd probably use the existing skeleton in the hope that someone will come along in the future and make some new animations for that instead of using the mocapped rig which might not have all the animations we'd want anyway which would then force us to make two sets of animations every time we want some new animation.
-
But I don't get the sence of making concept art without realising it (e.g. models, maps, textures, ...)
If he makes some cool concept art some of the modelers (Baddcog, Arcturus, Springheel or myself) might be interested and able to make it in a suitable quality for the mod (as opposed to someone who's making his first model).
-
So are you showing of new character models then? If you don't like Blender, then use 3ds max
I don't know if it's really a good idea to try to push him to learn to model (especially characters) when as far as I've understood he haven't modeled anything at all before. Learning to model, unwrap and texture at a somewhat acceptable level takes time (I'd guess it'd take at least a year until you could make something good, but obviously that's different for everyone). If he enjoys making concept art and he's good at it then that's probably what he should stick with.
-
I honestly should have checked previous concept are because I was not aware of the elite guard.
I would love to do that too, but I have no idea where the concept art is stored (except for the very limited amount shown under the media section that is).
Machines you say? Like anything In general mechanical? More specific.I'm not a mapper so I don't know for sure what they want but AFAIK they want bigger machines that are somewhat general purpose (we've just got about 5 machines ATM). The mod is steampunk themed so it might be a good idea to make somewhat exposed mechanics for the machines you make such as gears and pipes, etc. (while still making it possible to model without too many polygons or texture without extreme resolutions).
So long as it pertains to the Dark Mod and I get as constructive criticism as this? (figures the one person I draw I dont use the golden ratio on has fucked up proportions XD)I'd be happy to give some constructive criticism next time as well. So yeah, good luck next time!
-
More importantly is the concept good for about a hour of work?
I think it's pretty good. I'm no expert (far from it) at proportions or anatomy but I think the arms look very short, the head looks a bit to big (might be a big helmet though) and the legs and feet look too thin as well. Detail-wise I'd say it's pretty good. I think the shield could use some kind of trim on the top like on the bottom, albeit not as big as the one on the bottom. The greaves looks a little feature-less/boring. For an hours work I'd say it's good, but it could be better (it's a lot better than I could've done, but I'm mainly a modeler). Also it'd be beneficial for a modeler to have as many orthographic views as possible (at the very least a frontview and preferably a sideview too). I do realize that this probably won't go further than this (as we have the elite guard already) but I just wanted to give some pointers for the future.
Modeling characters is a lot of work and we don't have many members who can do it (I think it's only Springheel ATM, I could do it but I'm a horrible skinner so I don't know if I'd be able to finish it), but something that's easier to make would be machines. There was talk about making cameras and some other stuff a while back but I don't think that really took off but more general machines would probably be appreciated.
-
Are there even any ASE scripts that work for the latest Blender 2.6x series? All I can find are 2.4x and 2.5x versions, which don't seem to work (at least they don't show up in the interface when installed).
There is this one:
http://campagnini.net/2011/04/16/blender-to-udk-ase-export/
The exporter for 2.49 felt a lot better as I didn't need to combine all meshes and it didn't triangulate the meshes in Blender either, even if you wanted it to triangulate the resulting files. I tried using it on my bed model but couldn't get it to work so it might be a bit borked (or I just failed somewhere along the line).
-
Hm, I am pretty sure I already converted all TGAs to DDS, leaving only the normal maps, and some TGAs that were not worth it (like anything under 100Kbyte). Can you point me to the files in question?
I think I had just extracted an outdated pk4 and assumed that it was still the case but after looking again in the pk4 I see you were right.
And I agree with the "colored" materials, I think as long we change the texture (and leave the material shader with the same name) it won't break missions - they use the material name, the actual texture file name is irrelevant to them (unless the missions pack their own material shaders which is rare).
Great!
-
I think some of the cloth textures need some fixing as a lot of them are TGAs instead of DDS's. Also I think the texture "cloth_baize_green" should be a colorme material (not sure that's what they're called) as it didn't lose anything IMHO by having it desaturated in Photoshop and then colored with the Hue/Saturation again, so it should work just fine (though it'd probably have to be added as a new texture so it won't break existing maps) with colorme in DR.
EDIT: The whole cloth_plain_creased_ texture family could also use a b/w colorme texture as well.
-
Well, You can try materials with blends, then just use an Ao texture, and blend the color diffuse over it. Though it's really tough to get the colors you want from my attemtps.
In that case I'd need two UV-sets. One for the AO and one for the diffuse (which is ATM scaled bigger than the 0-1 UV range). I also think the AO is simple enough in this case to just use vcolors.
Or you can just use an AO, and it can be colored by RGB in materials. Probably the best bet, good shading in ao and authors can pick any color they want. Plus there's no texture bloat for the mod.Either way you need a material made, it's super easy to copy/paste. My flowers are done that way. AO texture, rgb blend.
I want mappers to be able to choose any texture they want for the bedding and then make matching pillows with that and make it possible for them to match it with mostly any other piece of furniture so that's why I wanted to choose a good method where it wouldn't be necessary to make a new skin for each variation. Ideally it'd be possible to modify an existing material inside DR to change diffuse-, normal- and specularmaps (and more) so that mappers could mix however they wanted instead of only relying on ready-made skins and materials and DR would automatically generate a new material for that. That'd be such a great system for mappers (I think, I'm not one of them but it'd potentially create a lot more variation) and artists (doesn't need to make a 100 variations of a material), but I'm not naive enough to believe it'll ever get implemented (if it would've been as easy as it sounds, it should've already been made).
Would be much easier for you to experiment if you just took a little time to get stuff in game yourself.
I know and that is what I were going to do this time (even though I find it very tedious and boring). I was just unsure about how I should approach this issue.
-
Oh that looks much more convincing Noss! <3
Thats looks very nice there young Noss.
Thanks!
I'm working on the material right now for my bed and I'm unsure if I should make a material for the bedding as it is just using a tiling texture that supposed to be mixed with 10+ diffuse textures (would have to make a lot of materials for all of those) while I also want the vertex colors multiplied over the diffuse texture (essentially AO). I also want the same for my pillows (but without the quilt texture) what should I do?
-
A video would've been nice. I'll check it out later.
-
@noss, the more the merrier dude! having lots of pre-made models is always much faster than having to make patch-based func_static prefabs etc which Melan will attest to.
Wasn't me this time around. Pusianka is the one deserving the credits.
It is a nice Roman portal, but we need in first place multi purpose, common models. Too recognisable models can't be copied among one map. Do we have a list of "must do" objects?
This one's a bit better:
http://wiki.thedarkmod.com/index.php?title=Model_Wishlist
Essentially they're the same but I updated the wiki one by removing some of the things that have already got made and removed some non-working links. Keep in mind that it is still a very outdated list so do check if the models already been made if you decide to make something.
Zombie concepts
in I want to Help
Posted
I will give it a try, but as I said in the other thread I haven't made all that many characters (even fewer decent ones). I will give it a try never the less though, I do think school will be a bit hectic in the coming months so I can't promise that it'll be quick, but I will try to make one. I'll see if I can get any response from Dominus otherwise I'll just base it upon a muscular/skeletal set of imageplanes.