Page MenuHome

ASC-CDL Color Balance Node does not allow for full range of values
Closed, ResolvedPublic

Description

ASC-CDL Color Balance Node does not allow for full range of values as per spec.

Domain of Slope:

0 ≤ slope < ∞

Domain of Offset:
-∞ < offset < ∞

Domain of Power:

0 < power < ∞



Event Timeline

Hi I did some test and was able to use these domains. (it is just not clear in the user interface of the color balance node) perhaps you could try when:
- color balance click on slope-color. they a color popup will appear. here you can enter any digit.

or am I missing something?

Hi, there isn\'t enough information here to reproduce this bug.

As mentioned in the bug reporting guidelines, please include a
test file, clear step-by-step instructions on how to reproduce
this, and full system information (such as OS, graphics card,
driver versions).

Matt: you've added the code, with a reference to http://www.asctech.org/ (which is member only site). Having this information public, like in our wiki would put in active maintenance by others too :)

Hopefully you can reply to what Andrew likes to know.

Yeah, as jeroen said, if anything it allows too great a range - by editing the colour form the colour picker you can type in whatever you like. Perhaps slope and power can have their lower limits set at zero, but there aren't any upper limits there currently.

Ton: as for the docs, I'm not sure if it's allowed to be distributed (which if so is bloody ridiculous) - it came from that password protected site which someone (troy s?) gave me the login for in the past. Maybe Andrew knows if we can make these docs/test files public.

I received the latest ASC CDL specification easilly from their 'email-address' as stated on Wikipedia. (I received the response within an hour without any questions asked) The license is unclear, but perhaps we could ask them on the possibilities.

Matt: what to do then? Your suggestion "Perhaps slope and power can have their lower limits set at zero" would solve the report? It's your code, would appreciate if you also commit then :)

Well, in lieu of any response from from Andrew I can just do that for now and close this report. Will get to it in the next day or so.

cheers

closing as per previous message

Matt Ebb (broken) changed the task status from Unknown Status to Resolved.Jun 9 2011, 1:10 PM