modeling reality into sci-fi

[Neural]

Hi. I'm new here, but not so new to CG.

I'd like some opinions in regards to a specific problem I've run across a number of times. For me, it's the kind of issue that stops progress on a model and winds up with it in the folder of things I've never finished (also known as C: )

When modeling something like a mech, a robot, a human with robot parts, or other related items, how do you judge where "realistic mechanical design" ends and imagination begins?

Being very detail oriented, I have this tendency to want things to leave no question as to their function and how they work, but this often times creates mental blocks as I try and figure out how something that doesn't exist should work. The models become incredibly complex when you take into consideration what you have to design in order to make a humanoid hand that is mechanically sound.

Are there any sort of rules or standards on where to let the viewers imagination take over?
Do I fill a joint with wires and rods and boxes and hope that it looks like it might be functional?

Sorry for the lengthy post, but this is one of those things that has been a problem for me for a long time. I'm really good at taking something that exists (like a clock on the wall) and making it in CG, but creating details out of the blue seems to be a real trip up for me. :/

[MikeANash]

Hey Neural,

This is not a direct answer to your questions, but instead of a way of thinking that might help you out in the areas of trouble for you.

This is my thoughts on the subject:

It's all about the Art of Bullshitting your way through the design and a bit of research into modern technology, also your own ideas of what could be possible in the future.

I find you can get stuck on certain designs since it's pretty much something that's never been done before. I think it as a evolutionary step for yourself, you evolve your own way of thinking about these things.

Usually you do get stuck on certain areas of design since were not engineers, a lot of designs we do are suppose to look cool while being functional. So when you do get stuck on a certain area of design you need to start brainstorming new designs. Start researching mechanical designs, something will eventualy come to you, trust me hehe.
I find majority of functional designs look boring and ugly, you need to find a happy medium there. I usually go for a 50/50 share, half bullsh+t and other based on real designs.
Just do whatever you think will work on certain design areas you get stuck on, otherwise you will never be able to move on with the design.

I find most sci-fi designs are rip off from magna and usually are just your typical simplistic design, people choose this style since they have lack of a creativity or ideas for something new. Also filling joints with wires is a lack of creativity or effort.
Usually I try and take ideas from modern day car engines for jointed areas, If you look inside of a engine bay there are supporting structures, sensors etc all kinds of crap you need to think of when designing something.

5505

I try and stay away from shape inserts like this, usually their been done to death, there is always a new way to approach design details.

5502

Example of modern designs:

If you look at a piece of military hardware you can see it's built strong, it has been molded with a lot of smaller support details. If you use these type of designs as an example you will create something more believable since its based on modern technology. So your not just taking style from other artists and redoing the same old crap you see everywhere, your actually applying design standards of already established functional designs. It's your job to take modern day design standards and put your own high tech spin on it.

5503
5506

Excuse my poor grammar btw

[Kel Solaar]

Great post Mike! Inspiring I would say :)

KS

[ulf]

cool thread, thanks a lot for sharing your experience mike!

[BrettSinclair]

Great reply. Thanks Mike:beerchug:

[Neural]

Thank you, Mike.

I like the idea of mimicking the ways that a car engine looks, and the example images are good reference points.

The rest is probably the same as always. Time and practice.

I do have to ask.. What about simplified models? I guess that's an area I get tripped up in also. I've seen some great looking little animations where the robots just have sphere joints and sticks for legs.

I can see the answer to that in what you wrote though, I think.

Maybe I'm getting hyper-sensitive about it as well since it's usually other artists who are critiquing what I do, and their eyes latch onto things like hydraulic cylinders that appear to have no function and such.

Now I know why everyone does organic forms. They are much easier. lol

[MikeANash]

Thanks guys :)

No problem Neural, hopefully others can also benefit from this thread, I rarely hear anyone talk about the Art of doing Hard-surface work, you always hear about human anatomy instead hehe.

Simple robots work because it's showing a working of the joints moving with a hydraulic, most people have a understanding of that, so it just clicks for you.
Maybe you can use Pneumatic Mucles to help solve joint issues, they mimick real muscles so it's easier to design around them.

http://www.wired.com/images_blogs/gadgetlab/Festo_Airics-arm%201.jpg

Just try not to place random crap all over everything, things like bolts, wires etc are over used. If you can't give a reason for why its there don't do it haha. I try and think of a reason why I would place such a detail or construct something in a certain way.