Tales of the Rampant Coyote
Adventures in Indie Gaming!


(  RSS Feed! | Games! | Forums! )

Wednesday, October 26, 2005
 
Will I ever NOT suck at 3D Modeling?
I used to fence - you know, the sport with the swords (well, foils and epees and sabers). It's a great sport, and EXHAUSTING. It's ridiculously fun, too. The only problem is that I suck at it. Oh, against other beginners I do okay... but man, the sport is so fast that even when I'm called upon to judge a bout I have a tough time seeing the order of the actions. Who attacked first and had the "right of way". That's an important distinction in foil fencing - he who established the attack, or correctly countered his opponent's attack and then attacked, gets the "right of way" and gets the point in the event of simultaneous touches. Which happens a LOT.

So I acknowledged that I suck at fencing. But that was okay. My goal, I told everyone, was to someday "suck less." I didn't harbor any illusions about being some great championship fencer. I'm not a teenager anymore, and I am not about to make fencing my life. But I could still improve to the point of being competent. I kept going to the club in an effort to suck less. Unfortunately, the club closed down and I haven't hunted down a new one yet, so I still haven't achieved that state of less suckage. But towards the end, I could see a very definite improvement in my abilities.

So now I'm learning 3D modeling. I'm using Blender. I've done some small amounts of modeling in the past - mainly very simple objects in Milkshape (I was very proud of my floating wrench pick-up in Void War. ) I managed to get a BOX modeled, textured, surrounded by a collision volume, and (with great pain and gnashing of teeth) exported from Blender into Torque. THAT experience nearly made me throw my game away and start over with another 2D game. At least in 2D, a picture is a picture. You don't have to worry about polygons with skewed textures, or the collision volume being inside-out, or all the crap you have to go through to export. It's a lot of work. But once I did it, I found that while it was a pain in the butt, it wasn't difficult. But all I had was a box. And not a particularly brilliant-looking box, either. I still suck.

But after that experience, I suck less. I think. So now I want to tackle something trickier. I have a TON of content requirements (SCARY!), most of which I'm going to have to out-source to friends, associates, and hired help. But I thought I could try something kind of simple and start learning the ropes of adding animation. I don't dare start with a human-looking character. But I do have need of having a giant mosquito. So I figure I can do that... insects are simple, right?

So I spend a few hours trying to model a mosquito. I'm still having to battle blender getting textures to "stick" - I think I'm doing something wrong there. But I get the model together, and I throw together a colorful texture as a stand-in, and I arrange the model with my newfound technical knowledge of how to export an object from Blender to Torque.

With great joy and excitement, I put the mosquito in my game. Stand-in texture, no animation, just to see how it looks and works.

And he resembles nothing so much as a giant Cootie. You know, that preschool game where you have to build a bug? If I was doing "Cooties: The Computer Game," I'd be golden. But my giant mosquito was supposed to be all menacing and stuff. Maybe it was the colorful stand-in texture, and maybe all it needs is some tweaking. I sure hope so. But I was turning up my nose at some content packs earlier because they look "amaturish." Compared to my cootie - I mean, mosquito - they look awesome.

*Sigh*

It's a good thing I earn a living with my programming skills. 'Cuz I still suck at 3D modeling. But I'm going to keep at it in hopes to someday suck less. And in the meantime, I'm going to continue to be insanely jealous of you guys and gals who can whip out a beautiful, complete human character model in a couple of days.

FEAR THE GIANT MOSQUOOTIE!!!




Did you enjoy this post? Feel free to share it: del.icio.us | Digg it | Furl | reddit | Yahoo MyWeb

Comments:
Cootie bug rocks! Fricking hilarious. You got mad cootie bug modeling skillz - don't let anyone tell you different. ;)
 
I'm lamenting this -- we don't have a large graphics (programming or art) budget for our games, but I'd like them to look nice. I think that a hyper-unrealistic style will go a long way in helping us here (such that we're not competing head-on with Doom 3 or Half Life), but there are certain instances when a solid block artwork for interface and website design would come in handy.

What's a small studio to do, Jay?
 
Well, one thing I've tried recently (we'll see how it goes) is exchanging code for artwork. It's kinda hard to pull off, because programming is RARELY convenient and modular like artwork is. But I did find one recent opportunity. That kind of horse-trading has some tax benefits as well.

There are SOME low-cost alternatives out there. You can try some of the guys (like the artists at GarageGames) who have talent but not much experience, who are looking to put together a portfolio, and will be willing to work for cheap. The problem is that it's a big risk working with them --- and of course, once you help them build their portfolio, they are likely to go outside your price range in the future. I guess you get what you pay for.

Stock content packs are another option - but that'll only take you so far. You'll end up with content of different styles which might not work together, a chair that has the poly count of a main character, etc. And STILL you'll need to have a lot of custom, game-identifying artwork done to give your game it's unique look, feel, and personality.

Beyond that - man, I don't know. I'm good at programming, so in a sense the time I'm spending learning to model and design levels and so forth is *wasted* time I could be spending programming. It'd be nice to not HAVE to outsource to get some stuff done - but it's really raising my costs in terms of TIME.

Long-term: We need better, easier (and CHEAP) tools that are maybe a lot more limited in features, but much faster for getting the job done. Imagine a "furniture builder" that lets you import textures, lets you mix-and-match styles, and designate a polygon budget. Or maybe one that'll throw together a 3D model of various insects based on simple, drop-down menus. There are a few tools like that for making trees and plants out there right now. That's a possible solution.

Better yet - how about having a web-driven version of the same? You choose your parameters online, and pay a fee to download the model.

Another option which has been attempted but hasn't really caught on yet has been the attempt to create kind of a code & content co-op. The problem is that while people are happy to take content FROM the co-ops, it's harder to get donations (at least those of any quality & usability).

Quite the puzzle!
 
Post a Comment

Links to this post:

Create a Link



<< Home

Powered by Blogger