Hard surface modelling and prototyping at the same time

dzibarik

I found that doing hard surface modelling and prototyping at the same time in a standard modelling (not sculpting) app is not doable for me. When I started learning modelling I thought that there will be a time when I'll be able to model without relying on reference images that much. Well, it seems that it's not happening ever. When I model stuff I'm concerned with right edgeflow first and foremost. It's almost purely technical thing and there is no room left for prototyping in my head when I do this.

So it seems that if I want to model sci-fi gun for example I need to make a pretty detailed concept in Photoshop or any sculpting app and only after that start making shapes.

Is it just me not being able to this or some of you prototype and model at the same time?

BARDLER

Well two major things I want to touch on here to try to help you out.

"When I started learning modelling I thought that there will be a time when I'll be able to model without relying on reference images that much."
Never ever under any circumstance do art without reference. You make it sound like using reference is a bad thing, or has some negative connotation to it. Reference will ground your designs and models to the real world and make them look functional and believable. Before I start a new modeling project I gather tons and tons of reference, everything from little material samples, to blueprints, to lighting reference, to color pallets, and everything in-between. Usually by the end of my project my reference folder has 200 plus images in it that I pulled information from.

I got some amazing advice from Paul Pepera a while back on here when I asked him how he makes his sci-fi doors look so damn functional. His portfolio: http://www.peperaart.com/
"Look at real world reference. I always have photographic reference up on my monitor when I model. If I am working on a spaceship door I look at submarine doors, ship hatches, bank vaults, manholes, aircraft landing gear, etc. Derive your functionality from real world functionality. Depending how serious you want to get with incorporating function in your design it is best to study the subjects that are closest to the subject matter of your work. Try to understand why certain things are built the way they are - what functional problem are they trying to solve. Also, knowing when to restrain from details is highly important. Many people tack on useless details in a desperate effort to inject functionality into their work.

P"

Second thing is, if you are prototyping ideas then just use basic shapes/primitives bashed together. You don't need to do a highpoly model of an idea, just a sketch to see if it will work or not.

Hope that helps.

dzibarik

I'm using tons of references too but sometimes when I want to do something that doesn't exist and I have references I get inspiration from I still run into "THIS SHAPE SUCKS" issue when I model it to a pretty detailed level. Basic blockout doesn't help sometimes. Well, I guess I was right, I must get a pretty clear image of what I'm going to do and model only after I'm sure that it's final.

Tobbo

Try drawing it first. Doing this forces you to actually get something concrete down.

You can even take screenshots of the detailed level that you have now and create a new layer in Photoshop and draw on top of that! Use this as a tool to see if the shape, color, silhouette, etc. is going to work or not with your scene before you get too deep in the modeling phase.

If you're new to drawing I'd recommend checking out this site. www.ctrlpaint.com/

Give it a try!

fattkid

I do concepting and prototyping a fair amount in 3D. I would say the way to do it is to work simply - just focus on simple geometry and your proportions, shapes, relationships, and structure of the parts. Details and beveled edges aren't necessary in this stage.

As the design evolves, I may add some smaller elements like lights or vents or whatever, but even these are basic mass out geometries that are just floating on the main model, or somewhat simplified or representational/placeholders, that I can move and adjust easily.

I find that I can fully realize a design - as to whether or not it's cool looking, functional, designed well (as well as I can anyway) etc., pretty well in the mass out stage. There is a bit of translating the mass out design to the finalized design in my mind's eye, which I will incorporate one I start refining the model. Perhaps that ability to visualize the end result a bit is a matter of experience.

Ultimately, if the model isn't "working" at the end of the mass out, no amount of detail is going to fix it. You can get pretty far into a strong overall design with a well done mass out.

Dataday

Sounds like the OP needs a dose of Mesh Fusion (add echo)

https://www.youtube.com/watch?v=kjzfhbW2hNY

Otherwise a great alternative is 3d coat's voxel boolean workflow.
Though that would probably count as being in a "sculpting package"

https://www.youtube.com/watch?v=qw2qQFh64UI

dzibarik

Dataday wrote: »

Otherwise a great alternative is 3d coat's voxel boolean workflow.
Though that would probably count as being in a "sculpting package"

https://www.youtube.com/watch?v=qw2qQFh64UI

yes, I'm learning 3D Coat now.