Hello, Here's a quick model I drew up in half an hour. This is a simplistic model in my team's latest game; I'm using it for educational purposes. I'd like to help my team out more in the art department, especially since there is a lack of 3d artists these days. A brief description of the model and a few pics of em' are below: [Artillery Piece] "Nuke Cannon" This Russian artillery peice can launch a 100lb nuclear charge every 45 seconds. It is designed for short range attacks on targets up to 5 miles away. The Nuke Cannon's full power can be seen when used in a battery, or grouping of several Cannons. A FULL VIEW OF THE MODEL THE TOP THE BASE I'm working on: a.) Lighting b.) Rendering c.) Texturing I welcome all comments, whether positive or negative. Suggestions as to further improving this model are welcome as well. Thank you, [Edited by - D Shankar on June 24, 2006 8:32:36 PM]

The question I have to ask is, when you qualify your art with a statement like:
"Here's a quick model I drew up in half an hour,"
why should we spend 5-10 minutes (a third of how long it took to model) critiquing it? I'm sure if you spend another hour on it, you can come up with something good. The design is extremely bland and doesn't make sense, that's the real problem. You're modelling work should be more efficient (basically, use 8 sided cylinders, but this is secondary to the design. Think about a weapon that launches missiles, or research a weapon that does, think about how it is reloaded, how the rocket is fueled, is it mobile, etc.
If you showed me that model, I wouldn't guess its a nuke. Are those fuel cannisters? A barrel for the artillery gun? Honestly, you owe it to us and yourself to spend more than a half hour on something you expect critique on. In college, my professors didn't yell at us if we didn't do work, they just refused to critique our work... if I don't want to put in the time, why should they work harder to cover for it?

Quote:

Original post by Professor420 The question I have to ask is, when you qualify your art with a statement like: "Here's a quick model I drew up in half an hour," why should we spend 5-10 minutes (a third of how long it took to model) critiquing it? I'm sure if you spend another hour on it, you can come up with something good. The design is extremely bland and doesn't make sense, that's the real problem. You're modelling work should be more efficient (basically, use 8 sided cylinders, but this is secondary to the design. Think about a weapon that launches missiles, or research a weapon that does, think about how it is reloaded, how the rocket is fueled, is it mobile, etc. If you showed me that model, I wouldn't guess its a nuke. Are those fuel cannisters? A barrel for the artillery gun? Honestly, you owe it to us and yourself to spend more than a half hour on something you expect critique on. In college, my professors didn't yell at us if we didn't do work, they just refused to critique our work... if I don't want to put in the time, why should they work harder to cover for it?

well, i didn't want to say anything becuase i didn't want to be as harsh as you. but you asked all the right questions.
my question i would like to contribute is, are you studying to be an artist or are you a programmer who is trying to do art?
as far as the actual model goes, as said it is not design very effeciently, cylinders are the most poly sucking models in 3d and those are way to high poly, reduce the amount of sides.

[Edited by - Jarrod1937 on June 24, 2006 10:48:10 PM]

Quote:

Original post by D Shankar [Artillery Piece] "Nuke Cannon" This Russian artillery peice can launch a 100lb nuclear charge every 45 seconds. It is designed for short range attacks on targets up to 5 miles away. The Nuke Cannon's full power can be seen when used in a battery, or grouping of several Cannons. I welcome all comments, whether positive or negative. Suggestions as to further improving this model are welcome as well.

How does it shoot? Your design right now has it so it would only fire straight up in the air, and that 100lb nuclear charge would land right where it was fired from. You need to mount it on a turret for it to make sense. That is unless it fires a self guided missle, to which you would need to make a silo-based model rather than a gun-based model.

Next, its really designed more like a building than a WMD. Weapons require a lot of enegery and support for them to work, it looks like one small tank cannon would take out that structure. It needs more beef to it! If it were a mobile structure, then a lightweight design would fit how you are describing it, but you say its full power can bee seen in a cluster.

Finally, a "Nuke Cannon" is something that is supposed to strike fear into the enemy's heart. They'd at best get a little giggle if they saw that. It needs to look formadible and scary, which goes to the design of it.

So, it's not that it's bad, you just have some work to do. For an artillery piece, especially a Russian artillery piece, use Russian weapons to model from or rework your description. When I hear of Russian artiliary, I think of their mobile missile launchers (modern day Topol M). So take this post with a grain of salt. Good luck! [smile]

Lol my rating went down by 14 points after that post ;)...Really messed up that time.

So how do I model "efficiently"? How do I create a cyndrical shape without cylinders?

As for the design aspect, it wasn't intended to be popped into the game immediately, it has a lotta work left to be done. Research wise, thanks for the suggestions. I'll follow up on that.

Is there any book I can buy that will tell me in detail how to efficiently do stuff and basically how to model the right way? Thanks.

EDIT: Thanks for the harshness. It reminds me not to be noobish on the forums, and to only post stuff when I'm completely satisfied. I was actually trying to get help on improving it 'cuz I wasn't satisfied. My lighting was off (i looked up tutorials, not too helpful), I had no idea where to begin in texturing. And I wanted to clone and make it into a battery type thing. But I didn't want to get too intricate before I knew how to do the other stuff. For example, WHY make it detailed when you guys pointed out that cylinders sucked? I would have wasted a lotta time...

(Just so you know, I'd never rate anyone down unless they actually were negative or unhelpful... just being n00bish or foolish isn't enough reason)

Anyway, I suggest that you go to www.3dtotal.com or some other site and go through a bunch of tutorials. Basically you are asking us to help you on your first or so model, you have something specific in mind and you're asking us how to make it. First, learn the fundamentals of modelling, of your 3d program, get some research for design, and in a month come back to the model with the skills you've learned.

Also, forget about lighting and texturing for the time being. There are a couple good books on lighting for CG, but its basically no different from lighting for film or anything like that. Texturing is going to be difficult (I'd say impossible) without any artistic training and studying (formal or self-taught), but you definately can't begin your road to texturing having bypassed the art and design and fundamental/traditional aspects of it.

Quote:

Original post by Professor420 (Just so you know, I'd never rate anyone down unless they actually were negative or unhelpful... just being n00bish or foolish isn't enough reason)

That's good to know. ;)

Quote:

Original post by Professor420 Anyway, I suggest that you go to www.3dtotal.com or some other site and go through a bunch of tutorials. Basically you are asking us to help you on your first or so model, you have something specific in mind and you're asking us how to make it. First, learn the fundamentals of modelling, of your 3d program, get some research for design, and in a month come back to the model with the skills you've learned.

I've been working with Max 7 for over 6 months now (Sad, I know). Without good training/tutorials, I'm finding it difficult to learn exactly what a chamfer is or a fillet and so on. I think a lack of formal training and years of experience is to blame when I produce inefficient junk like the Nuke Cannon..

Furthermore, this isn't my first model (although it seems so). I modeled a Ferrari Testarossa (no texturing or lighting) a month ago, and a (simple) Church (complete with church windows, altar, pews etc.) before that. Aside from those two, I did many practice models which I have saved in a "Practice" folder.

Quote:

Original post by Professor420 Also, forget about lighting and texturing for the time being. There are a couple good books on lighting for CG, but its basically no different from lighting for film or anything like that. Texturing is going to be difficult (I'd say impossible) without any artistic training and studying (formal or self-taught), but you definately can't begin your road to texturing having bypassed the art and design and fundamental/traditional aspects of it.

Alright then, I guess I'll put my time and money into modeling for now.

Thanks

Hi. This is my first post on this site as i am new to the gamedev world (few weeks). I hope to learn many skills from other and in turn pass skills on I have learned. So I feel i can start my end of that deal right here as I am a 3d artist.

So about this "Nuke Cannon", all the comments you have had so far I would agree with, even the more harsh ones. Since I started down the path of game development about 3 weeks ago, all i've read about in beginner tutorials is "don't expect to be programming quake 4 anytime soon!". Well this has been pretty easy for me to accept because I've already gone through all this with 3d graphics 9 years ago!
"don't expect to be animating Toy Story 3 anytime soon!" Man were they right. Fortunatly I wasn't put off. Any skills like this require not hours but years of your time. Seems like forever when you start and you just can't wait to be doing the "good stuff". Well I'm just about to start my forth week in programming. I've just finished reading "C++ in easy steps" and even though all i have is a lotto number generator to show for it(plus all the basic skills of course) I am more excited than ever. Learning new things is the best part of it.

Anyway. Wow. I went on alot there! Anyway on to your model. First thing is i'm not going to talk about your design. Everyone else has pointed out great ideas. The only thing i will say about your design is....i hope you've got one! If you don't have something drawn in front of you, even if it's just a good ol' scribble on a napkin, then you are destroying your chances of producing a great model. Not to mention making it about 100 times harder for yourself.

Some of the things you've asked for help are rendering and lighting. Well this confuses me a little because I assumed this was a model for a real-time 3d engine, which if it is, shouldn't the game engine cover the rendering and lighting? If your wanting to "bake" your lighting to your textures then I can understand the need for it but in that case i think your jumping to far ahead. But i know how you feel though. You've just started your model in 3d but it looks a bit rubbish in your viewport because there is no texture and the default lighting does not do it any favours. But thats ok though. Models don't always look that great with no texture and lighting. The final look of the graphics need ever element to complete the look. But for now you should just focus your efforts on the model, then when its time to texture and light it, the extra effort early on will pay off no end.

You've asked,
"So how do I model "efficiently"? How do I create a cyndrical shape without cylinders?"

Well a cylinder can have as little as 3 sides giving you a very low poly count. You need to spend a bit of time thinking about how this model will be seen by your audiece(the gamer). If by the time this model reaches the game it only gets represented by about 50 square pixels then all that geometry is going to be eating away at your resources for nothing! Also look at your model as it is in these pictures. If you half the polycount on the cylinders on and around the barrel then doubled the polys on the base you would still be left with much less polys than you had to start with but you would notice almost no difference on the reduced cylinders and the base would now be really smooth!!

This does not just go for games. Pixar have just released Cars and they will be thinking of exactly the same things. If i car is only visible from way down a street and never gets close to the camera. Do they bother modelling a fully detailed windscreen wiper or do they just represent it with a simple box that matches the basic size of the wiper!

Look around you at objects on your desk even and start to think how can i represent that shape in 3d with the least amount of polys. Its much easier if you can see the object in front of you. Thats why when you don't have a "nuke cannon" outside your window to look at you must have at least one drawing on paper. I like to have at least 2 views of the model i'm working on. Try to draw views that show detail! If for instance your modelling a human head your going to need refrences to work from. The best 2 views are a front view showing all the detail of the face and a profile(side) view. There is not much need for a top view in this case because that won't really help you. Think about it, all you will see is the top of the head with the nose sticking out a little. But a car you would have the top view, along with a side view as they show the most information. If you can have more views, even better!! And if you can get a 3/4 veiw then your laughing!

Sorry to go on so much! I hope this has helped a little. What 3d app are you using? if its Max then I should be able to help you with that as well. Work some more on the model now, using the advice everone has posted here, don't worry about texture at the moment.

Play some games!!! Have a real good luck how its done in your fav games. C&C Generals is a good one to look at. Thats a game that needs lots of detailed units on screen at once but cannot comprimise on speed! If you zoom in on the units you will notice that the models are very basic indeed. You'll be supprised how little geometry is needed to represent a tank or an airport. (have a look at chinas nuke cannons!!)

Above all have fun!!

Thanks for listening to me go on for so long. Hope to see more of you all as I go struggle to learn more about programming and gamedev.

Thanks

Dan

Dan -
I really appreciate your help here.

About the "engine doing the texturing & rendering" --> I just wanted to be able to show something nice in my design document. The Nuke Cannon is based of the turret here:
http://commandandconquer.filefront.com/screenshots/News/21813/1
Funny thing is, its based of C&C, just like you told me to! I drew up some "concept" art on this turret, and I'll scan that later. It looks quite similar to the nuke cannon itself with the exception of the circular base (it has a hexagonal base). There is more detail on the concept.

I primarily use Max 7, but I borrowed Max 8, Softimage XSI (advanced) and Maya 7 off a friend of mine. He's in the CG (3D) art business but he's on vacation in Germany. He let me borrow a lot of his software, and I thought I might as well make use of all this. The only software I've touched so far is Max 7 (as I said, I've been working with Max7 for half a year.) I haven't even touched the others, yet. I also have Zbrush 2 (borrowed from the same guy), but thats texturing, and I'm not going there until I'm able to model a full character I have in mind.

I bought a book on human modeling, but I have one big problem.
Drawing all those views - side, top, front etc..I'm only able to draw one of them to perfection (usually an angle view). I can never replicate the picture because I'm a perfectionist and I'm concerned over having the head height be perfect. 'cause when I use blueprints (like for the Ferrari I drew), they all matched up. That might be the first step I need to overcome.

And about the polycount: I'm not too concerned at the moment because I decided (sorry for not mentioning before) that I'd model some characters & buildings for the GDD and texture it to make it look "pretty." I might even go the additional step to create a trailer. Therefore, poly count etc is almost tossed out the window. Even if rendering times are long on my primary comp (P4 2.0, ATi 9250 AGP, 640MB RAM), I can always find ways down the line when I actually have to render (such as using my Intel MacBook Pro, or my friends comp.)

Thanks

Great,
I'm glad I helped a bit and thanks for giving us a bit more of an outline on what your trying to acheive. If you look at the models section on my site you can see a few models i did for an RTS game.

http://www.marionettestudios.com/animation_models.htm

There is not really a game it was just a concept idea, although that is one reason i joined this site. I hope one day that i'm not stopping at concepts. It that more what you were after? But textured?

IMO i think you should stick with Max. I have found it to be the best. Let me change that a bit. You should stick with whatever you feel most comfortable with! But if you choose Max...you'll go far :) Hehe.

As for human modelling. Its one of the toughest things to model in 3d. Not only are humans the most complex shapes you can imagine but your audience, people, humans!!! Know them so damn well. Its annoying! I wouldn't worry about your front and side views being a bit off. As long as there almost(closer the better though!) you'll be ok. They should only be a guide. If there a little off resize them in photoshop to try and get the best match possible. Have a look at www.3d.sk for some reference pictures (!!WARNING THIS SITE CONTAINS NUDE IMAGES!!)
Also take a look at this tutorial,

http://67.15.36.49/ffa/tutorials/max/joanofarc/joanmenu.asp

This is hosted on 3dtotal but the above url will take you straight to it. Its one of the best tutorials for 3d I have ever seen. Well worth it!


Also on the subject of polycount. I admit you have lots less to worry about if your pre rendering but I would still try to keep them as low as possible. Even in a prerendered animation you can see the benifts. Render times will be effected, especially when you use more advanced lighting techniques like Global Illumination and soft shadows. Plus you will be thankfull when trying to work with your seen in the viewport (this is all done in real-time don't forget). Its tough working with Highres models. Plus is just good practise!

Good luck with your work!!
Thanks

Dan