So, what are you working on right now?

[Baddcog]

Jewelery Box/Silver with matching key. uses existing textures. 583 tris for box and lid, shadow is 64 tris combined.

 

Key is 545 tris with a shadow of 114 tris.

 

 

wine crate and worn (still need shadow meshes-forgot between making them and coming home and uploading)

 

window bars (need skins)

 

open debate

[Springheel]

Very nice. Love the window bars.

[ungoliant]

you seriously made that key in DR???? I know you probably used the scale function, but still... thats some fancy patchwork right thar. Not to mention, how did you texture it?! hmm, looks like the top of the box and the key handle use the same texture?

Edited February 19, 2012 by ungoliant

[Baddcog]

Naw, the box and key were made in Max, but DR is great for things like the bars and crates.

[Diego]

Naw, the box and key were made in Max

ah.. My eye was twitching already.

[Nosslak]

Can't find a screenshot right now, but I still have the map with the crates build in DR.

Yeah, you should show it so that we can judge whether the rejection was just or not.

 

wine crate and worn (still need shadow meshes-forgot between making them and coming home and uploading)

I think you should add a couple more segments for the rope handles otherwise they look pretty good.

 

window bars (need skins)

Looks good though I think the bars feel just a little bit too thick right now so if you could scale them down they'd look great.

[Baddcog]

Yeah, that's the main issue modelling in DR, just hard to get stuff small enough. I was working at a pretty tiny grid to get those patches that small.

[Nosslak]

Why are you making them with DR anyways? I've barely used DR but to to me it feels like a modeling app would be a lot faster and grant you more control.

[Baddcog]

Just on a laptop seeing what I could make in DR. Plus all the materials are right there, no need to hunt them down, extract them...I needed some bars for windows in my map so...

 

The crates were easy enough as there isn't that much to them anyway. Just using stock textures so no need to bake normals or anything.

 

But yeah, I wouldn't do anything too complicated in DR. But I've used it enough for mapping I'm comfortable with it anyway. New editors really are getting much more like 3d apps, You could never have even done simple stuff like this in Dromed (well, straight bars... but no bends)

 

Part of it too was just to show people what kind of stuff can be made

[Tels]

Yeah, you should show it so that we can judge whether the rejection was just or not.

 

http://forums.thedar...post__p__224924

 

I can't find the model, prefab or map, tho. Maybe it did not survive my spring cleaning last year (run out of HD space)

 

@Badcogg:

 

Making small stuff in DR isn't impossible (but I wish it would be possible to simply zoom in more), but you can use a trick. Make it oversized, then scale it down before exporting. (Hm, maybe the ASE exporter should have an "scale factor" input box, the scaling down is quite cumbersome in DR).

 

See here:

 

http://forums.thedar...post__p__264060

 

esp. the last picture in this post, reposted here:

 

 

 

(Oh, and if I said I never did any modeling, I was wrong. Obviously, I modelled the keys last year. And they were added simply w/o consulting any one

[Nosslak]

http://forums.thedar...post__p__224924

 

I can't find the model, prefab or map, tho. Maybe it did not survive my spring cleaning last year (run out of HD space)

You were right, IMO it should've been added to the repository as either a prefab or model. I can see how it would clutter up the list but I think it's still good enough and would save mappers enough time to justify it's inclusion.

[Springheel]

Nobody said it couldn't be uploaded as a prefab; in fact, that seems to be what most people wanted. But instead it turned into a one-sided argument about why it should be a model.

[Bikerdude]

Nice windows bars BC, are there in SVN already..? along with Tells models..?

[Baddcog]

yeah mine are up.

[Tels]

Nobody said it couldn't be uploaded as a prefab; in fact, that seems to be what most people wanted. But instead it turned into a one-sided argument about why it should be a model.

 

I love how you skirt around the issue - if I want to add a model, you say it should be a prefab instead - completely ignoring the fact that a model and a prefab are two completely different things. One is an ASE, the other is just a bunch of func_statics and brushes.

 

I wouldn't mind adding the original DR work as prefab in addition - but lets fce it, nobody asks any other modeller to convert their model into a prefab, so why should I do it? Plus there are the technical issues with func_statics-as-prefabs that can be avoided if it is a model + entity. But I stated my case and won't repeat it again. The work seems to be lost, anyway.

 

(And for the record, nonsense replies like the above are why I am getting frustrated instead of doing any TDM work. You put in all the hours and just to get stalled by silly "oh why don't you do X instead" where X is something completely random.)

 

Edit: It occured to me that not everyone knows the difference between a prefab/func-static and a model, so here are the key differences:

 

* func-statics are stored inline. That means if you duplicate them in DR, you duplicate the loading time, the storage requirement, the parsing time, and the memory they occupy (50 func statics take 50 times the space). Models, OTOH, are only loaded and stored once and then referenced.

 

* although you can prevent faces from rendering by caulking them, func statics will still store these faces in the map. They get still parsed, loaded, then ignored. For a model, faces that are invisible are simply non-existant.

 

* you can do some thing with models (like def_attach them to entities) that you cannot do with func_statics. It would never be possible to define a pre-defined "pile of wine crates" except by manually copying them around. See also next point.

 

* SEED has problems with func statics. After many hours, it works some what, but there are still things it cannot do with func statics that it can do with models. (mostly has to do with the fact that you cannot refer to a func static by its name alone). This has to do with the anonymous nature of func statics, which are from the point of the engine, just a bunch of random primitives.

 

* If we ever implement instancing in the engine, it would very probably not be able to cover func_statics - the engine can't know that one func_static is a 1:1 copy of another func_static - with models, however, this is easy. Making things like crates as prefabs prevents that optimization from every taking place (so each crate would be rendered by itself, instead of all of them being rendered by the GPU in one pass and w/o additional data uploads).

 

It makes perfectly sense to use prefabs for world geometry, or for some types of things (mostly things that are used only once, anyway, or things that the mapper will modify a lot after loading them). It makes, however, no sense to use prefabs for cookie-cutter type of things, like crates, bottles, flower pots etc. Basically everything we have as a model is better as a model.

 

And that's enough forum posts for me for today. I already wasted even more time on this (and the result is just as zero as before). So excuse my while I now go to do some real work - there are fonts that wait for me.

[Springheel]

I love how you skirt around the issue - if I want to add a model, you say it should be a prefab instead

 

Actually, if you want to be really accurate, it was Sotha, Sneaksie and Badcogg who were arguing for it to be a prefab. All I said was that people shouldn't start uploading every model they make in DR just because they can--something I still stand by.

 

but lets fce it, nobody asks any other modeller to convert their model into a prefab, so why should I do it?

 

Firstly, your model was made in DR, so it was made of brushes and patches to begin with, while models made in modelling programs are not. Big difference. Secondly, that was your first model, which you admitted "looks quite simple" and "misses a few details" and was primarily "a learning experience". In general, someone's very first model is not going to be the kind of quality that makes it an immediate candidate for the mod. This isn't some special rule that only applies to you--it's the reason we ask for modellers to show examples of their work before giving them something to work on.

 

Badcogg, by comparison, has been modeling for TDM for almost five years; that puts him in quite a different category. After contributing dozens (if not hundreds) of models, he's more than earned my trust that his style and quality will suit the mod. If you're arguing that anyone who makes their first model in DR should be given the same trust, then I don't know quite what to say.

[AluminumHaste]

LOL talk about quality control, let's start with the massive difference in quality between the blackjack (which looks pretty amazing) and the horrible sword model+texture. That should have been rejected and a new model and texture made. Take a look at the huge array of weapons in Skyrim+Oblivion and tell me why something like that can't be made for Doom 3. It's not like the engine can't handle a ton of polies.

 

EDIT: I'm not trying to make the author of said model cry, I appreciate the contribution we all do, but as SH pointed out, Quality Control is important and things like player inventory items and weapons should be on the top of the quality list.

For example, the looking glass overlay is low res, and looks like crap. Is there a technical reason it's like that?

[Springheel]

and the horrible sword model+texture.

 

I don't see any problems with the quality of the sword (both the sword and blackjack were made by the same person, incidently). Nor have I ever heard anyone else complain about it in the past seven years that we've been using it.

[AluminumHaste]

So I guess then I'll take a stab at making a better one lol, put up or shut up right?

[Springheel]

Well, replacing the player sword would require re-rigging and animating the first person attacks, which I'm not keen to do unless there is a lot of demand for it. We could always use more variety in our general weapon models though.

[AluminumHaste]

Well not necessarily right? If the handle is the same length and the sword same width and length and thickness and the joint attach point is the same, can't you just drop a replacement in?

[New Horizon]

The sword, blackjack, bow, and player arms were all modeled, textured and animated by oDDity. Maybe I haven't looked closely enough over the years, but I've honestly never had a problem with any of them. oDDity's work was always of the highest quality.

[AluminumHaste]

I don't see any problems with the quality of the sword (both the sword and blackjack were made by the same person, incidently). Nor have I ever heard anyone else complain about it in the past seven years that we've been using it.

 

I want you to take a good hard look at this model+texture and tell me everything is fine. I mean was it made on purpose to have a huge dent in near the bottom of the blade? Is that a uv issue? Plus the model itself is low poly and it's just not an attractive piece.

 

 

Now take a look at the blackjack:

 

 

Very nice, smooth normal, lots of clean detail very attractive item. Compared to the sword I'd say the sword was whipped together in a day and the texture made the same way.

 

No compare that with something like this:

which was made for Skyrim. Or this:

 

[AluminumHaste]

Now I know that obviously the guy making swords for Skyrim is INCREDIBLY talented, and our modellers need more experience.

[Baddcog]

The sword does appear to have some wobbly/normal map issues.

 

I guess I never noticed because I rarely even equip and when I do it swings so fast I can't see any details.