Post here if you volunteer to make a certain prop/static model.

As with the other "asset" threads, this one is not for discussion, but merely a place to let people to know what you want to work on, so we don't have redundant work happening. Also, the first post will be updated with the latest information and the old replies removed to keep the thread organized and easy to read.

G

=======================================

Static Model List

- Cars, Trucks and other Vehicles

---- Volvo 760 - gw

---- Mercedes E200/E280/E320/E320 CDI

---- Mini Cooper

---- Saab 9 5

---- Renault Espace

---- other?

(Want to make a car for the project? Check out http://www.suurland.com/blueprints.php? ... order=name for a large selection of reference blueprints to get you started.)

- Post/Mailboxes - obsidian

- Newspaper stands - obsidian

- Trash cans - dfloss

- Street Lamps - Goliath

- Doors

- Windows/Window frames

- Parked (stationary/non-interactive) vehicles

- Road signs on posts - Goliath

- Road signs, hanging

- Traffic lights - obsidian

- Fence: Wood

- Fence: Metal

- Bench: Wood

- Bench: Metal

- Road barriers

- Road rubble (broken slabs, of concrete, potholes, etc. that would be easier to model than create with brushes)

- Traffic cones (these may already be in D3, I think... we can re-texture them though)

- Tables and Chairs: Cafe

- Tables: Picknick/park

- Trees and Bushes - insectattack

=====================================

These are just the things off the top of my head and a few things that I picked up from reading other posts. Post replies to sign up for something and a mod will update this list accordingly. Of course, it's fine to have people working on the same type of object... I'm sure we'll want more than one kind of road sign, for example. But we also want to avoid 5 people making similar traffic lights, and meanwhile no one making any tables.

G

Don't forget free 3D models which can save a bit of time, on this page there's a useful park bench and two street lanterns that look fairly decent (dxf format) :-

http://www.3dcafe.com/asp/househld.asp

I also don't see much point in making models of bog-standard street signs that consist of little more than a pole with a metal sign on them. A standard cylinder with a flat patch mesh on it would suffice. The flat mesh would be textured with a two sided decal.

Its good to have bog standard light posts cause if a change needs to be made its easy for it to be made to all of them and also when each mapper is making their own they don't have to worry about little things being slightly different. guarenteed identical lamposts and signs for everyone.

colganc wrote:

Its good to have bog standard light posts cause if a change needs to be made its easy for it to be made to all of them and also when each mapper is making their own they don't have to worry about little things being slightly different. guarenteed identical lamposts and signs for everyone.

Fair enough if everyone is making a different map, but this is one project to form a London style city square with surrounding buildings - it's not like we're going to throw signs up higgeldy-piggeldy wherever the fancy takes us - it has to make sense. Traffic lights at intersections, no parking signs in logical places (in London that would be everywhere

), one way signs on one way roads.

But if people reckon models for what essentially constitutes a pole with a piece of metal stuck on it, fair enough - I just think it's overkill.

What something to stick on your pole ?

http://www.highwaycode.gov.uk/signs04.shtml

All the UK signs you need.

Yeah, the reason I listed things as simple as road signs on the static model list was to ensure uniformity. I think it'd be easiest to supply a few pole models and a bunch of sign models (it'd be easier to get the correct and consistent shapes in a 3D app than in the editor) that have their origins lined up such that they're a piece of cake to add to the map...

Just right-click, add euro_pole01. Right-click, add euro_sign_oneway. Line up the origin points. C'est tout.

If we have them created within the editor, then we might have signs with different thicknesses or different sizes for the signs, using too many or too few subdivisions for the "pole" cylinder patch, etc.

Plus, with the few signs I modeled last night, I ended up putting a base at the bottom of them that would be a pain to make in the editor.

G

I'm sold !

BTW, I guess your going to "bash" the signs up a little - wear and tear and all that ?
You could maybe even have one or two with bent edges, or poles at slight angles.

Yup. The reference images I have for the signs are waaaay too spotless. We will definitely need to weather them up a bit.

G

On pg 18 of "D3CDIT"Project thread I posted a few images of a building entrance I had modelled. I'm highly intrested in being involved with the project.

http://www.doom3world.org/phpbb2/viewto ... &start=340

I seen that shot revender but I'm wondering what the intention is. Are you planning to model an entire building?

rich_is_bored wrote:

I seen that shot revender but I'm wondering what the intention is. Are you planning to model an entire building?

I could do that. But at this point it was just to put together something that could be turned into normal maps.

No, textures are great. We need that for sure even if all you can contribute is normal maps.

I thought you were starting on a building because you had a whole entrance modeled.

I wasnt quite sure where to post this, but this thread seemed like a good place for it. I have been trying to learn lightwave so I can model some stuff for the map. Im a Max man myself, so there is currently a lot of teething going on.

I have been working on a street lamp for us to use, so here is a shot of it so far. I have not modelled the luminaires yet, becuase I am still trying to get texturing working. This could take a while :roll:

so I thought Id upload a pick of the untextured mesh anyway, and see what people think of it.

If you're doing this in lightwave, check out my lightwave/static model tutorial... it covers how to get stuff properly textured in both Lightwave and Doom3.

http://www.doom3world.org/phpbb2/viewtopic.php?p=17647

G

Well, your tutorial is what I was basing most of my lightwave learning from, since it was relevant to Doom3. I just went over it again for the third time, and it still doesnt work. I am only trying to generate a diffuse map at this point, to keep things simple, I am pretty sure that cant be whats screwing it up, because I am not specifying any normal or spec maps in the material.

Any suggestions? I am about ready to give up

Hey... head here so we don't clog this thread:

http://www.doom3world.org/phpbb2/viewtopic.php?p=68686

G

Great project! I've read the whole board, and I'm impressed with the progress so far. As a Doom 3 modder, I'm also excited about the PR potential of the success of such a project

Anyhow, to the task at hand: I wouldn't mind modelling a 2CV or Smart Car for D3CDIT. I could texture it too (except for normal mapping [1]).

I think I'll do it for kicks anyways, and post some screenies to see if youse[2] are interested.

EDIT: Should I keep anything specific in mind if this model were to become driveable?

-Glee

----------------------------------------------
[1] As I can't generate a Doom 3 normal map from Blender, as I understand it.
[2] plural

Gleemonex wrote:

EDIT: Should I keep anything specific in mind if this model were to become driveable?

-Glee

I'd check out the bones from the id's buggy and try and match their bones structure and names, that should make life easier for making it drive able.

I've been meaning to help out with this project for some time now, but couldn't find the time...

To start off, I guess I could start off with a few models.

These look fun:

-Post/Mailboxes/Newspaper Boxes
-Traffic Lights

Added you to the first post. I also removed my name from a few things because... well... I just don't think my modeling abilities are up to the level of quality in this project, and I don't want my noob-ish work to bring it down. No worries though, I have plenty of other areas to which I really want to contribute, so....

G

Did this yesterday as a test... took a couple of hours. What do you guys think? There's one of these on the corner of my street, so I went outside and took some rough measurements in the snow. I'm pretty sure that all my neighbours now think I'm crazy, since I was standing in a pile of snow, measuring and photographing a newspaper stand.

Edit:
If you want, I could model this plus the mailboxes to actually open and close. For the traffic lights, I may need some help if you guys want the lights to change through scripts (or maybe just the material file will do?).

I'll also model a fire hydrant. Just make sure nobody puts a dog in this map, I don't want them pissing all over my work.

obsidian wrote:

For the traffic lights, I may need some help ...

I thought you were going to say you need to borrow a ladder to get the measurements

BTW, Nice model, the map needs more models on the street.

Dante_uk wrote:

obsidian wrote:

For the traffic lights, I may need some help ...

I thought you were going to say you need to borrow a ladder to get the measurements

me too....

I made a tree, maybe you could use this.
Polycount is 567 triangles, i'm not sure if this is justifiable, considering a character model is about 2000 polys.
Also, the trunk's uv map is really el cheapo...

And what could we do about textures with alpha channels not casting shadows?
People have suggested to use decals as fake shadows, i didn't like the idea either, but what are the options then?
No shadows at all?

download (598 kb)

Those are some great looking models guys.

I wish I had a solution for the shadow problem. I suppose you could try using forceShadows on the leaf material. It would only cast shadows from the geometry but perhaps that's good enough.

One thing that I think we should do is set up some sort of polygon count limit for some of these models. We don't want to overblow our polygon budget in the scene by adding a few small models. Of course, it all depends on the model, but a rough polylimit estimate would be helpful for us modelers. So for the typical D3 model, what is the general polycount for both high poly and low poly versions. Some sort of rough ratio would be very helpful.

I haven't much experience with renderbump on models yet, since everything that I modeled previously was with the Q3 engine. So I've been experimenting with various LOD's trying to get decent results. It's hard figuring out just how much detail can be "faked" using normalmaps and how much actually needs to be modeled. Any tricks and tips would be greatly appreciated.

The listModels console command will list all the models currently in use aswell as their polycounts.

I've been using it as a rough estimate.

It seems like Doom 3 doesn't really mind polycount too much. Most everything I've made up to this point has had a polycount around 2k.

Well... polycount does come into play a bit... but not nearly as much as previous engines. It really does all go back to lighting the level well, though because high poly counts in a poorly-lit map just make the performance problems worse. The good news is, for something like a tree, it won't be lit by too many lights anyway (hopefully just one, the sunlight), so I don't think it'll be too much of a problem.

Put it this way... in my tests (many of which I do in the latest test-build because it gives me a decent baseline for performance) I see no real slow-down when using medium poly models without normal maps and low polys with normalmaps. Then again, I have a 256MB card, so....

G

I think that's because the calculations for normal specular and diffuse stages are done regardless whether there is a image assigned to that stage or not.

I'd like to help out because i think this is an awesome project. I finally figured out how to import models into doom 3 so if no has taken it yet, I can start small with a trash can (or trash bin should I say?). I'm not from the u.k. so can any people who are more familiar with london tell me if this is what a typical london trash can looks like?

or is this a bit closer:

I think either of those would fit in. It's up to you which you want to do.

thanks rich,

well here is what i came up with so far... wrapped it up in a pak file and threw it in a practice map i made. it's getting late so i added the pedestrian symbol from the pics before. I don't know if it's too simple for the project, but I think it fits the clean lines of the d3cdit project. I'm very new to textures (i just read some of the tutorials here today to do this trashcan). I added a specular map, but no bump mapping.

Looks really good, dfloss. I think the proportions look slightly off... Maybe it's the size of the "slot"? If you make the slot larger by reducing the vertical size of the top part of the can, I think the proportions will look a little better. Excellent work!

Be sure to send me the model when you have a chance and I'll include it in the current test-build.

G

Goliath - it only takes me a few seconds to change the size of the holes or the size of the object itself. I did that this morning:

I feel like I'm way taller than the zombie, but I scaled it to his size. If you notice I'm having some problems with the textures stretching etc, plus I have the trash symbol on the inside of the can as well as the outside. Let me know what you think of the overall proportions of the trashcan and I'll work on the textures tonight

Hmmm, I think it looks good, just needs to be much shorter. About waist height (if that means anything in a game).

Maybe put yourself in 3rd person mode, change the viewdistance and stuff, then see what it's like compared to your character. It needs to be, maybe just above waist height at most.

Yeah, do a:

pm_thirdperson 1
pm_thirdpersonrange 200

in the console and test things out.

Also, you can get a .obj-format playermesh to import into your 3D app here:

http://dev.rosebrough.com/d3/euro/asset ... player.rar

It's "headless," but it should still help w/ proportions while you're in whatever 3D app you like to use.

Goliath

Some new models:

Mailbox

It's based of this photo that I found online

Fire hydrant

I started working on a car last night, based on my old 86 Chrysler Fifth Avenue. It may not exactly fit into the whole european thing, but maybe it got shipped over

. About how many triangles should a car be, right now it looks to be ending up somewhere between 1k to 2k.
Here's a shoot of what I have going now http://img129.exs.cx/my.php?loc=img129&image=car2hu.jpg
It is still very much a wip but you should be able to get a feel for the look of the car. What do you think?

If the low poly is going to be around 1 or 2k then you'll be fine.