Search the Community

1st: GrotFest mod, low poly 2nd: Broken Sky, medium poly (game) 3rd: Illustrations for a book ( not mentioned in the previous list) Besides that I have a school project to take care off. BUT *if* the model is interesting enough I could give it a try, for both. Cobra 6

Where would you guys think this model is best suited? I had originally thought the 'Mechanical' folder (because it's clockwork) but would it be better in the generic 'Furniture' folder?

Still have a version with one object. i'm now using an older version of maya (5) which i know works with exporting..but not yet for this model I'm getting some kind of "getpolygonUV failed" error which i gues means it has to be something with the uvmap but i can't figure it out...yet. Other models work just fine. "fixed: finally exporting "

Not as far as this model is concerned. There is no creative aspect to the coding. Once I put it in the game, the code justhas to make it turn in certain vectors when you press a key. Hardly creative.

Perhaps you'll tell me who esle created the player model except me? I'm responsible for the creative side of the model, and that includes the name. You're only responsible for the technical side, and that has nothing to do with the name. I don't try to get involved in coding, and you should stop trying to get involved in creative matters.

I think I read something like this on Doom3World. You can scale models and the likes, but I'm not sure wether simply scaling an AI down or up a bit really makes variety, because it can look strange. Shorter people are not neccessarily thinner as well, but simply scaling them down will result in a model that is overall smaller, not just the height.

Yes, several people do really hate the name Daerek. How are you writing the code that makes him come alive, it's the animaitons that do that. I could implement him as the default player model using the existing D3 code. I think we've already established that your grasp of what constitutes good or bad sounding names in English, is laughable at best. You're the only European on the team who needs to deal with measurments.

I'd personally like to see Farrell be the same height as the beautiful fair maiden lady: The Beautiful Fair Maiden Lady *sigh* *batting eyelashes* With all the enemies being so large, in general, though, I have a feeling it might seem odd when Farrell goes toe-to-toe with *any* enemy and he's looking at their chests. On the one hand it could be cool that everyone will look so intimidating, but on the other hand, we might get the feeling that he's just a little kid looking up to a bunch of adults. The latter isn't so appealing and I won't know until I play it if I get this feeling. Just a hunch that I/we will, though. Critiquing... I think the player model's boots look too big/tall. They look to almost cover his knee-caps. Proportionally, something just looks haywire there in the lower legs. I know you love hearing about your boots, oDDity. I rescinded any vote I had for the name 'Daerek,' or whatever, once I found out how it was going to be pronounced. I originally thought it would be pronounced more like "dark," but later heard otherwise. LOL, what are you talking about? How did you deduce that he had pagan roots just because Fearghal means "heroic"?? (Please, no need to explain ) An analogy, though: my name means "tough," ergo I have hammerite roots. WTF?? LOL I think we would have been fussy. None of us liked the lack of creativity for the paintings in Thief 3. Springheel could've gone too modern, too nouveau, or too many other things. Similar to how we have to get the music & sounds just right.

Why don't we wait until the player model is in-game? Then we can see what kind of name would best suit the way he looks (and Odd can do his best to make him look like a 'Farrell').

I think it's pretty clear that there has yet to be a consensus about the name. I personally think the thief's height is fine, though I agree the elites are a little tall. I'm 6'5", and I don't often run into people my height. If people are making maps based on our player model, they could easily wind up with doors too small for elite guards to get through.

I think nobody will complain if you help out Oddity a bit, and add additional models for variety. As long as we have only one usefull human modeller, we must see that we get a good number basic models, and then we can start to create a variety. Even so it is a lot to model for Oddity and we also want to have different heads as well.

This is where procedural detail generation kicks in. Of course, it would take a modeller (or a team of modellers) an enormous quantity of time to create a forest of trees, and the mapper if they were to place them in a map by hand. But tools like http://www.speedtree.com/html/speedtreert.htm make populating a map with thousands of unique, individual trees a doddle for mappers. Same goes for a lot of animation - manually keyframing a statue to fall when hit by a rocket is a lot more time consuming and less realistic than simply placing a statue model that has physics properties that will mena it falls over when struck by a rocket. You can apply procedural content creation to a huge chunk of game design by building tools that model the laws of nature (this of course means much more initial work for programmers, but it accelerates content creation immensly). EDIT: here is another example: have a look at this http://www.theprodukkt.com/ go to .kkrieger (you may have seen it before) - a full DirectX9 3D game in 96 kilobytes. All content procedurally generated on the fly - models, normal maps, sounds everything pcked into a tiny file. I was gobsmakked when I downloaded this tiny file and saw how much detail was generated procedurally (granted, the game itself is ordinary, it is a tech demo after all, and it has fairly heavy system requirements, but you get the idea).

hahaha its just an interesting topic for me, I cant help it. But to answer your question, it allready is on CGtalk, Military-Meshes, EWnet community, Deviantart, Heroes in Feldgrau, Axis history Forums and a load of others. But I share my work with everybody, not only two forums. Cobra 6

Lucky that it's true then, isn't it! IIRC any arguments otherwise have pretty much been dealt with and thrown out in two threads on this forum already. Care to suggest a better one? The 'oDDity is King of all' model where no one but you gets a say?

Well, I have been away for quite some time. I am very busy in school and some other projects I'm working on, but I thought I let you see me newest models for those who like them SdKfz251 "D" Tiger Tank interior JU87 "D" Stuka cut-away Well its about finished now, only a couple of things left to do: -Model the bomb release -Model the attachement of the wing to the fuselage -Some minor color touchups maybe -Adding a background And that is about the max my machine can handle right now http://www.guuspeters.demon.nl/101.jpg http://www.guuspeters.demon.nl/103.jpg http://www.guuspeters.demon.nl/104.jpg http://www.guuspeters.demon.nl/105.jpg Cobra 6

For camera reference and shadow they don't have to be very refined. Blocking in the intial animation is quick, but refining it to a finished animation takes 80% of the total time, and it's really only for mirrors that a refined animation is needed. SO is the camera going to be attatched to the player model's head or what? How will the lean animation affect the first person view? I'm not sure what this animation is supposed to look like, becasue leaning is performed against a wall at a corner or around a pillar etc, you dont' lean while standing in the middle of a room. There is no generic leaning action, it depends on the circumstances.

We may also want to allow AI to lean, right? So we could always just reuse animations for the player model, since they don't have to be particularly unique.

Doing an ultra detailed high poly model for a normal map, then doing a highly detailed diffuse map to fit it is a lot more work than making a 32x32 texture. THe difficulty doesn't really change, it's just a different kind of difficulty, but the amount of time it takes is getting longer. I can make a low poly character in a few hours, but making the high res version for the normal maps is weeks of work for some characters.

That would sound more convincing if you were a master animator, but you're only an average animator. You'd get far superior results by tweaking motion capture than animating by hand from scratch. Once you're a master, then of course your skill can make your animations stand up to motion capture, but for average mortals, using motion capture saves a lot of time and looks better. With terrain, you can either sit down and learn all the geology about rock formatons, how river beds form, about erosion patterns etc, and then learn how to model them, or you can use DEM data and tweak it for your own purpose.

WHat animations are needed for the player model at this stage.

Daerek's player model will never be in game, but Farrell's can be in as soon as you like. I can just knock up a low poly version without bothering to texture at this stage. I guess it just the dimensions of the model itself you need.

Oddity, when can we expect to get Daerek's player model ingame? I think this would certainly help in many cases, if we could get it.

Something I'd like to know (since I've started playing with Blender): How do we know how many polys is too many? Mostly in terms of Doom. Right now for instance, to replace a round column I have, I could simply make a round column in Blender. But, does that get me any advantage at all? In fact, I have a concern that the Blender model will actually be less efficient/more polys, than the simple D3 patch. I'm not sure how to check (mainly in D3). Do the same rounded shading effects that work on patches apply to models? If not, that would make patches preferable for many things. And, using a model might end up actually being more work, because I imagine getting a texture to look right on the damn thing is certainly far more difficult than applying a texture to a patch and saying Fit/Normal. I also imagine D3 is optimized to fudge it's patches as much as possible so they are very efficient. Since I'm going through tutorials and learning how to make some stuff, I find myself asking, "but should I be doing this at all?"

Yeah. Explaining things like the meshflow and how it affects the final model is quite helpfull, even if you are using a different app.

Yeah. I wanted to do the hires model for normal mapping. We will see. At least these human shapes are rather stylized, so they hopefully are easier to cover up, than a real human. I have additional shots from all the relevant parts from closeup. Ebay is quite usefull for that, because the seller photographed all of the individual sections in very good detail. Yeah. That's what it is mostly about. I wanted to do something that challenges me, to get a bit better at it. Don't know if I really finish it, but at least I should learn something from it.