Re: DPX reader vs. Northlight scans with infra-red pass

On Fri, 29 Aug 2008, Lewis Saunders wrote:
>
> What would work well for me personally would be an option to write the alpha 
> using this format.  I can imagine sometimes wanting to copy the alpha 
> information from a scanned sequence into a rendered sequence, for which 
> that'd be handy.  For reading, I don't see any advantage to keeping it as 
> 1bit if that creates problems transporting it to other file formats - for 
> example if converting a sequence to TIFFs it seems reasonable for the alpha 
> to be bumped up to 16bits to match the colour channels.

Since the data is not actually alpha, the results would be odd if it 
is saved to other formats.  Particularly saving to TIFF could be a 
disaster since the alpha is pre-multiplied with the pixels.

> One thing I'm thinking of is to use gm to change metadata in sequences - at 
> the minute I can't seem to get this to work, using "gm mogrify -define 
> dpx:mp.frame.rate=25 0899992.dpx" for example.  Happily I can use "gm

Try putting the -define after the file name.  Maybe that will work 
since there is an image to attach it to.

> identify -format '%[dpx:*]' 0899992.dpx" to quickly get the metadata - very 
> pleased to see that it does this without reading the image, unlike Shake 
> which I was previously using to extract the information.  An order of 
> magnitude faster!
>
> Again thanks for all your hard work,

I am glad to hear that it is useful for you.  That is my reward.

Bob
======================================
Bob Friesenhahn
[email protected], http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/