Filter Forge - Rusty Chainlink by Mike Blackney

Active Topics Search

Vladimir Golovin
Administrator

Filter Forge, Inc.

Posts: 3106
Filters: 54

Rusty Chainlink by Mike Blackney
http://www.filterforge.com/filters/871.html

Posted: July 13, 2006 9:09 am

Details E-Mail

Vladimir Golovin
Administrator

Filter Forge, Inc.

Posts: 3106
Filters: 54

Mike, this is amazing. I can't believe this is a procedural texture.

Posted: July 13, 2006 9:11 am

Details E-Mail

Vladimir Golovin
Administrator

Filter Forge, Inc.

Posts: 3106
Filters: 54

Noticed a minor problem with the normal map -- look at the regions inside the "cells", they should be flat:

http://www.filterforge.com/filters/871-normal.jpg

Posted: July 13, 2006 9:25 am

Details E-Mail

baxx

Posts: 19
Filters: 15

I havent seen these nodes yet, but ,mike,you are the best

Posted: July 13, 2006 3:42 pm

Details E-Mail

Mike Blackney

Posts: 352
Filters: 54

Thank you both so much smile:)

If there's any qualifier, I could never imagine making textures like this with another procedural texture generator. (Or if I did, it would take twenty times as long. smile;)

)

Quote
Noticed a minor problem with the normal map -- look at the regions inside the "cells", they should be flat:

That's good to know - I was unsure whether the diffuse alpha map was enough for the empty sections of the texture. (It was enough for the display in FF and Photoshop, which was enough for me at the time smile;)

)

I've made the adjustments so now all the maps (specular and metallic as well) have the alpha - and now these maps all look flat. Is that what I should have done, or should I just flatten the color for these maps? I'll wait to hear your word on this before I resubmit.

Posted: July 13, 2006 5:56 pm

Details E-Mail

Kraellin

Kraellin

Posts: 12533
Filters: 97

mike, very impressive! i'm going to have to pull a 'Ro' and start looking at the guts of your filters to get some tips on what's going on.

craig

If wishes were horses... there'd be a whole lot of horse crap to clean up!

Craig

Posted: July 14, 2006 12:09 am

Details E-Mail

Lucato

FF addicted

Posts: 494
Filters: 39

Congratulations Mike. Really cool. Amazing job.

Quote
Mike Blackney wrote: I could never imagine making textures like this with another procedural texture generator. (Or if I did, it would take twenty times as long. )

I don't know how long did It take for you, but I can tell you I took some days when I created one chain link texture with another procedural texture generator. I can bet FF is a little easier. Well, at least for me. smile:)

Congratulations again and keep the amazing work.

Lucato

Posted: July 14, 2006 5:53 am

Details E-Mail

Whitedove

Miss Congeniality

Posts: 225
Filters: 12

Mike,
I don't know how you did that, but can you now make a filter that will take chain link fencing OUT of a picture? I went to the zoo last Wednesday and the animal photos came out awesome, but some of them are marred by the fences.
Very frustrating.
I think I'll use this filter to fence in my dogs. smile:D

Donna

Posted: July 17, 2006 10:55 am

Details E-Mail

James

James

Posts: 649
Filters: 25

Great work smile:D

Posted: July 17, 2006 1:12 pm

Details E-Mail

CFandM

ForgeSmith

Posts: 3561
Filters: 204

Mike have you been in my backyard? This is my fence. Great Work on this.

Stupid things happen to computers for stupid reasons at stupid times!

Posted: July 20, 2006 9:27 pm

Details E-Mail

Northernshadow

Posts: 208
Filters: 19

Mike, this is VERY cool! along with your barbwire filter these will be useful, (even though like Donna I tend to remove fences.. smile;)

Sherry

Posted: July 21, 2006 8:45 am

Details E-Mail

onyXMaster

Filter Forge, Inc.

Posts: 350

Minor performance advice:
You don't need to use Set Alpha for all the Result component inputs, the Diffuse input will suffice -- resulting alpha is taken from Diffuse input only.

As a sidenote, setting alpha to anything below zero for grayscale (slider) inputs (like Height, Reflectivity for Result) leads to scaling of grayscale data obtained by averaging R, G and B values by the A value (alpha premultiplication), so setting the alpha to zero where there's no chainlink effectively leads to zero height, reflectivity, etc. (not saying it's wrong, just unnecessary since adding components hurts performance).

Posted: July 27, 2006 9:13 am

Details E-Mail

Mike Blackney

Posts: 352
Filters: 54

If that's the case, why would Vladimir have said that the normal map should be flat in these areas? (Since the diffuse alpha was sufficient in the original version, I imagine that the normal maps can be whatever I originally pleased to leave them as.)

Posted: July 29, 2006 12:49 am

Details E-Mail

Darth Axel 007

Darth Axel 007