.:I recently bought a XFX 8800GTX 768MB and Crysis the game. The visuals are so brilliant it's freakin ridiculous. This got me wondering. Many people are hobbyists in the 3D world with an intention to go into it professionally but have never had the opportunity to do so and this game smacks people like me in the face! The fact is 3D is moving so quickly unless you're in that field you don't stand a chance with your single desktop and 4 hours a week (nevermind 4hrs sometimes I only get to some 3D months later) Right or wrong? I'd like some hope here. Do we stand any chance of keeping up?:.

It all depends on what area of 3d you are working in. Movies, animation and the gaming industry are only a portion of it. 3D and digital art is actually involved in any of the areas that traditional art is found in.

A game like Crysis (which I'm also playing) only is daunting if you're a hobbyist trying to get into the high paced world of gaming production. Then it could seem hopeless.

But for any of the other avenues for our craft.....I don't think we have to worry and a desktop will still do the trick.

Nope, no chance of an inidividual keeping up with Sony, or ILM. Best you could hope for in that direction would be to get yourself onto a dev. team or into a studio.
Still,
There are many indy film makers. They are not Universal Studios; yet that fact has not stopped them from making movies. Same with games; OK, the game is not EA, but it's still pretty good :)
You can make money too. Not all clients/customers (by a long long shot) are in the position to demand ILM level output. They can't afford it either.

Well, if you mean make an entire game on par with the big boys...by yourself...no. If you mean create characters that look as good as (if not better than) those you see in games, yes. Yes you can. You just need time and once you have things down, you'll start speeding up so that things that took you days to do, will only take you hours. You can hope to develop characters or even environments on par with games of today but you'll be hard pressed to create an entire game by yourself on that level.

.:Ok I'm with you guys in many respects. So how do you break away to do the things you love without any formal training? I already have a pretty OK job but no time left in a day for any 3D / Art / Design. I don't quite have enough for studies + it would need to be full time study, time I don't have with any job good enough to pay for the studies. Ah fart! (lol) Suggestions?:.

Well, I never had any formal art training at all, I taught myself. Whenever I had a moment to be in front of the computer, I practiced.  Max, Rhino, Wings, Hexagon, Silo, XSI - whatever I could get my hands on. I practiced every free moment I had. It meant I had some sleepless nights but it was worth it. Same for 2d art. Every moment spent idle (riding a train or taking a bus...even lunch), was spent drawing. There's always time if you make it. Like I said, it may mean sacrificing a few hours of sleep a day but one or two hours a night less won't kill you (unless you're in the military - sleep's important for anyone who shoots as a proffession).  Just dedicate yourself to the goal and it will be done. It won't be a quick thing. It takes time to be good at anything - unless you're a prodigy of some type. 

Your first step should be deciding what you want to concentrate on. Do you want to be a modeler? If so, do you think you'd prefer modeling characters, props or environments? Down the road you'll learn to do them all but you'll want to start with one first.  Do you want to be a concept artist? Programer? etc. Ask yourself what are your interests  and then ask yourself what are your strengths. If you can find a strength to match an interest, then you're off. If not, it'll be a slightly longer ride.

Budget your time, to make time for working on things. Set aside an hour each morning, or each evening, or however you can work it into your schedule. Maybe you can set aside saturday mornings, or thursday evenings, or whatever...
"...but if we never take the time, how can we ever have the time?"
 - The Merovingian, Matrix Reloaded

I don't have near Teyon's experience; but am completely self taught also. Most apps. have just about all, if not all, of the information that you could need available through help files and online forums...and you can always ask, if you can't find what you're looking for. But I think that 'hands on' experience is the most important.

.:Yes. Time... Thank you for your valued input:.

I was lucky enoguh to have some formal training for 3ds Max (with v3.1 about 7 years ago) but before that I had been completely self taught starting with POVRay in the early nineties, then progressing through bryce / truespace. The formal training really was a more how to do it in max then about the ideas behind it.

One of the really important things is to think aboput what you want to do,

Great character stem from good biology!
Great scenes stem from an understanding of light / nature / architecture
Great animation from an understanding of motion

None of us can be Pixar in a single person - studios have teams of specialists - the modellers, the riggers, the texture artists etc all we can do is attempt to make our own little piece of it. If your idea is to switch careers in the future to work in a studio then think about whether you should reflect a little bit of a pipeline in your show reel....

Above all whatever you do enjoy it!!!!

Something to consider. King Kong only had 50,000 polys. The rest was hair and skin shaders. He can laugh, cry, beat his chest, run, jump, fight other nasties and produce in very realist ways a bloody big monkey. The Battlestar Galactica comes in at around 3 and a half million polys. It has a few guns that can go round and point up and down. And it has bits on the side that go in and out. It sits there like a lump, or moves across the screen, like a lump. It kicks Cylon butt, but that is the extent of it. 

My point is, if KK only needs 50,000 polys to produce, where are we going wrong when our pet project hits 100,000. Are we looking at the modelling side the wrong way? Are these projects do-able, we are just going about the basics wrong? 

very true cybastorm, take a look at how much emotion Pixar got a desk light to evoke, very low poly but excellent motion......

No, it's not wrong and while the model used in film was about the size of your average Poser character, it used bump and displacement mapping taken from a model in the millions of polygons. So both a high and low res version was made (this is the case for most 3d these days - games or film).

The trick is, after you've completed your model or scene with a super high poly count, take a good long look at it. Try to see the things that you could have done with fewer polygons or fewer isoparms (if working with nurbs).  One of the things working with nurbs taught me is to capture a silhouette with as few isoparms (and I would later translate that idea to polys)  as possible. Once you have the volume and form of the silhouette down, detailing it becomes easier and will require less information. That's my take on it. See how low you can go, efficient meshes will lead to faster modeling for you and easier use for the end user.  If something can be done with a texture or displacement/normal map instead - do it. Save yourself time in anyway you can that doesn't sacrifice the quality of the work being done.

Currently, my method of modeling goes like this:

Create a base mesh (I have one biped base, one head base, and one creature base) in Silo.
Sculpt the base mesh into the model I want it to be using ZBrush or another sculpting app.
Take this high res model into Silo and trace over new, low-res  topology.

Working this way, I can concentrate on form and volume, while keeping the creative juices flowing the entire time. It's easy to lose interest if you're just working with polys or nurbs, trying to complete the mesh while keeping it as animation friendly as possible.  Working with a sculpting tool, you can finish your idea fairly quickly and then come back to it to get a better, more useable mesh from it. I've found many artists are starting to use this approach with fantastic results but the method isn't for everyone. Just because it works for me won't mean it'll work for you.

So there's not a real  right or wrong method for modeling in general but there is if you have a specific goal in mind or time table to fill. Animation requires light, clean meshes for anything good to come from it, while still images don't. Some jobs will allow you a very limited time to get a model done, so being efficient with your poly flow is important, while if you're modeling just for your personal hobby, you can take as long as you like.

Ultimately, it takes time and dedication no matter what approach you take or why you take it.  KillerBean, was/is a project being made by one man. At first, it was just a short done in his spare time. Now, he's putting the final touches on its full length feature. All one man. However, it took him years - more years than an average film (I think 8 or 10)- to complete and much personal sacrifice.  Is it Pixar quality? Not really. Is it good? Probably, if the short is any indication but that's the trade off you have to expect when working alone.

Anyway, I'm hoping this forum will become a place people can turn to for learning such things. I've already started recording myself working (with voice yet!) and plan to do some examples of my workflow. I hope it will spark others to do the same, since, everyone works differently. With a repository of different approaches and ideas to the same basic things, I expect it to be easier for people to feel comfortable with themselves as modelers once they see there's so many others with the same desire and all with different ideas of how to reach their goal.

Teyon, when you say trace over new topology I'm not sure if I am fully understanding, are you placing the high res topolgy over with a mapping of some form?

I wasn't meaning to strongly suggest people shouldn't use high poly models, but I for one have a tendency to work very quickly at poly levels that aren't neccessary (when sculpting), I do believe that it's worth emphasizing that the base low poly mesh should have time spent on it before upping the sub-div levels.....

Finally I for one have learnt a huge amount from looking at some of the items you've posted and mini-tuts you've posted in this forum, can't say thanks enough for that!

I'm glad people can get something out of what I do. That makes it worth doing!

You're right that the base is the key. If you screw that up, everything else falls apart - when modeling straight. When modeling the way I described earlier, it's not as bad, because you can fix mistakes in the sculpting process.

So, to answer your question, Silo (and now ZBrush) gives you the chance to use a marker tool to basically, draw - like you would on paper - on your model, a new topology. Some folks swear by it. That is NOT how I do it though.

What I do, is load in the high res model in Silo. Turn on Surface Snapping and lock the high res mesh (while switching it to a ghosted mode so I can see through it).  Then, I start modeling over the high res mesh. With the Surfac Snapping on, my edges or my polys - depending on how I'm working at the time - will stick to the surface of the high res model. This helps me keep the volume and form, allowing me to concentrate only on the topology.  Handy for converting a  high res mesh into a more useable low res mesh.

This is an example of what I meant. The image on the left is what it looks like while I'm working on the low res. The image on the right, is the wires for both the high and low res mesh.  As you can see, I keep the basic form and look of the high res-Zbrush sculpt without the wasted polys.  To be honest, I could cut it down even more but I chose not too, for the sake of smoothness when rendering.

Hi Teyon,,, Hello everyone... 
I have seen better models in renderosity 3d modeling forums than in some games out there including Crysis.. which I like playing..   That's because high poly models are forbidden in games.
The key with making good efficient games touches on what Teyon is illustrating...  making high polys into low polys...  
The other very important elements for the visuals is the texturing,,,  and lets not forget the game engine... 

you can lease it from your favorite production house...  you can downloaded free if you are just messing with it... 

Thre are quite a few good free 3d game engines around.
http://panda3d.org/
http://www.ogre3d.org/
http://irrlicht.sourceforge.net/
http://www.rage-game-engine.za.net/
OGRE is the most widely used and supported; but Panda (Disney/Carnegie Mellon) is really top notch too.

Then check out this site too:
http://www.delgine.com/index.php
Free OGRE tools (and other engines too), and a free 'lite' game engine, Nuclear Fusion. The Package for the full version Deled (graphic editor) and Nuclear Fusion (engine and language) is only ~$100 US.
You could make games with the lite versions, but it would be harder to do.

Thanks for those links.... 

sad as it is I find myself pondering this in bed whilst failing to get to sleep last night.

My mind started drawing parallels between software development and 3d work. When I started software development (a long time ago :) ), my first professional job was hand coded and I hab a limit of 2k of memory space (out of 64k) to create a sprite switching routine, now I have all sorts of helpers - create software using models and auto code generation -  it's all much larger and faster but I think having that understanding of how to do it in an efficient manner can give an edge.

With 3d I started out creating scenes by hand writing the code for the renderer (POVRay - a long time ago) now I have all sorts of helpers and stuff  it's all much larger and faster but I think having that understanding of how to do it in an efficient manner can give an edge.

:)

That said I rather be playing Crisis then the stuff I created  for the amstrad CPC464........

Finished part one of my vid tutorial for silo2. It's not, by any stretch, on the level of Gnomon DVD's or anything. lol. I'll post it in a few. It covers making an eye and making a really bad mouth, lol. Okay, not as bad as all that but not exactly the way I normally make 'em either. Serves me right for doing it late at night.

Looking forward to the tutes teyon, are you guys saying that I should moderate my wild expectations and not put out a hodge podge of terrible characters, movies and games to con the massess into believing that I am a brilliant artist????
:b_funny:

Now I will hit my head with a reality check and go back to learning how to join the poly dots!

For my 2 pennies worth. I reckon unless a heap of modellers her unite and show those 'high enders' some real art that you guys do then everyone will just wander along. Jen

My issue is I have heaps of partially completed projects at the moment, so if anyone would like to do a tut on how to get off my lardy backside and choose which one to finish.......

"Silo 2 Face Modeling Excercise "

This is what the model looks like after a mirror and a tweak of the mouth area again within Silo2. For this image, I did not add or change poly flow, I simply re positioned the verts for a better look in the mouth area.

so if anyone would like to do a tut on how to get off my lardy backside and choose which one to finish.......

Hmmmm....any one modeled a size 15, we could do a tute on getting the boot to swing at Gog's backside. :lol: and one for mine too while we are at it.

Then it would be an animation tut huh.hehehe