YOUR ACCOUNT

Login or Register to post new topics or replies
Sjeiti
sock puppet

Posts: 722
Filters: 71
  Details E-Mail
Sjeiti
sock puppet

Posts: 722
Filters: 71
Let me start off by saying that I think that this is one of the coolest snippets around.

I can't really pinpoint it but I think there is a slight error in this one (or I shouldn't use it the way I want to).
If you look at the attached image you'll see a sphere on the left that is slightly noise distorted. If I run it through the unspherify offsets it looks ok... but when I offset it another 50% horizontally there is a slight difference in lightness (right image).

The weird thing is that when I unspherify and inmediately spherify again there is nothing wrong. But I want to do some horizontal offsets in between and that's when it goes wrong.

nb: seamless tiling is turned on

  Details E-Mail
uberzev
not lyftzev

Posts: 1890
Filters: 36
I made an updated version of this filter that may address some of this issues. Guess I'll send it in since byRo seems to be MIA these days.
  Details E-Mail
uberzev
not lyftzev

Posts: 1890
Filters: 36
  Details E-Mail
Kraellin
Kraellin

Posts: 12749
Filters: 99
byRo can normally be found at http://www.retouchpro.com . however, he's currently moving and has had to do so a couple of times in the recent past. i'm sure he'll be around when things settle down.
If wishes were horses... there'd be a whole lot of horse crap to clean up!

Craig
  Details E-Mail
byRo
an Englishman in Brazil

Posts: 138
Filters: 8
Sjeiti, thanks for the nice words and...
Kraellin, thanks for the helping hand.

Yes, there is a slight imprecision in the spherify / unspherify filters - however, as I remember, the problem was in the "spherify" part where there was a need to fabricate an arctan() function and not in the "unspherify" which uses basic sines and cosines.
The arctan() function cannot be fabricated exactly, so there was some simplification in the filter.
Also, as I remember (don't have access here),uberzev's versions are more precise.

In later discussions over at RuntimeDNA a more elegant solution (using arcsin() ) was posted. Not sure if I ought to post a link from here, though.

all the best,

Rô
_________________________________
My favourite question is "Why?".
My second favourite is "Why not?"
  Details E-Mail
Sjeiti
sock puppet

Posts: 722
Filters: 71
Sorry Uberzev,
still the same problem. I want use it in Galaxy and distort it before throwing in the swirly bit.

I've attached the important part of the filter. Maybe somebody can make cheese of it smile:-p.

unsperiphyproblem.ffxml
  Details E-Mail
CFandM
ForgeSmith

Posts: 4761
Filters: 266
Sjeiti and Uberzev I did have this some problem with this Ultra-Cool filter..But on some I did manage to hide it decent. This one the seam can still be seen but not as
much..Not the same problem but not really an issue..

http://www.filterforge.com/filters/1390-v8.html
Stupid things happen to computers for stupid reasons at stupid times!
  Details E-Mail
Sjeiti
sock puppet

Posts: 722
Filters: 71
Well I sort of evaded the whole problem now by not unspherifying at all and directly creating an unspherified starting point (must be faster as well). It's not the same (the center is a bit cramped) but I can work with it.


unsperiphyproblemevaded.ffxml
  Details E-Mail

Join Our Community!

Filter Forge has a thriving, vibrant, knowledgeable user community. Feel free to join us and have fun!

33,712 Registered Users
+19 new in 30 days!

153,533 Posts
+31 new in 30 days!

15,348 Topics
+73 new in year!

Create an Account

Online Users Last minute:

30 unregistered users.