Spring RTS Engine

Soooo.... they'll take the "lit angles" technique for making normalmaps... convert the results to a grayscale heightmap, and expect you to pay them money for the privilege of doing something you can do for free with xNormal, or for a very reasonable fee with CrazyBump?

Help me out here, but that's what it looks like from here.

Way more awesome: Videotrace...

http://www.youtube.com/watch?v=vda2RAEuW_g

Master-Athmos wrote:Way more awesome: Videotrace...

http://www.youtube.com/watch?v=vda2RAEuW_g

That does look pretty cool, would be a lot cooler if it could figure it out on it's own. I am continually disappointed in the state of AI when it comes to image processing.

If you'd want it to create a correct result on its own you'd need to type in some parameters for the specific situation and I guess choosing the right behaviour in terms of when it should recognize something as a seam takes just as much if not even more time than setting up some seams by yourself. Imo the system already is extremely competent in interpreting the input - especially when predicting the position of something not directly visible anymore while having no information of the camera distance or size / depth relations...

oohhhh shinyyyyyy.

Argh wrote:Soooo.... they'll take the "lit angles" technique for making normalmaps... convert the results to a grayscale heightmap, and expect you to pay them money for the privilege of doing something you can do for free with xNormal, or for a very reasonable fee with CrazyBump?

Help me out here, but that's what it looks like from here.

CrazyBump > Everything
(Including your mum)

expect you to pay them money for the privilege of doing something you can do for free