Maxforums - Study on overcoming Polycount vs Details problem via Normals Mapping

Study on overcoming Polycount vs Details problem via Normals Mapping

ravihara

I did some research and study on overcoming the details constraint when there is a low polygon budget. Here are the result clips:
http://www.youtube.com/playlist?list=PL2F2390AA3E68FC43&feature=view_all

I need feedback on the current progress above and suggestions, tips and things that I need to be aware of when following a this kind of pipeline.

Thanks!

read 274 times
5/28/2012 12:22:52 PM (last edit: 5/28/2012 12:22:52 PM)

K-tonne

questionable proportions in the model, pretty poor rigging and bad mocap all add up to make this look like a guy in a poorly fitting rubber suit
i have no idea if the normal mapping is working because you never show us the low poly without it or a wireframe
at first glance it looks ok- but look closer and things fall apart

read 254 times
5/28/2012 4:09:25 PM (last edit: 5/28/2012 4:09:25 PM)

Dave

I agree with K-tonne. You're showing us a few mini mo-capped animation scenes. How is that supposed to be a study on normal mapping? You don't show any wireframes, no different lighting set ups, no 360 rotations, no normal maps!

You're showing an end result... if you want feedback, you need to show how it's been built up, how it compares, and how it works in different scenarios.

read 249 times
5/28/2012 4:18:26 PM (last edit: 5/28/2012 4:18:26 PM)

herfst1

Agree. Watched halfway through the first vid and gave up.

read 245 times
5/28/2012 4:20:08 PM (last edit: 5/28/2012 4:20:08 PM)

ravihara

Thanks everyone for the replies. I agree with you all. Sorry for any time waste. Here is my Low poly model :

It was not rigged at all. And mocap clips are free stuff picked from here and there from the internet. Actually I wanted to study how the normals mapping will do when the model is doing bends and stuff. But I agree with your all. For now, it is just a waste of time. I will come up with a better presentation soon. Thanks for your valuable time.

read 228 times
5/28/2012 5:29:16 PM (last edit: 5/28/2012 5:41:36 PM)

advance-software

> how the normals mapping will do when the model is doing bends and stuff

ensure you normal map in tangent space for this to work properly.

http://wiki.polycount.com/NormalMap

read 224 times
5/28/2012 5:35:10 PM (last edit: 5/28/2012 5:36:03 PM)