W3C home > Mailing lists > Public > www-svg@w3.org > September 2001

Re: Linear RBG and feComposite "arithmetic"

From: Thomas E Deweese <thomas.deweese@kodak.com>
Date: Mon, 10 Sep 2001 13:02:51 -0400
Message-ID: <15260.62011.624013.186400@gargle.gargle.HOWL>
To: "J. David Eisenberg" <catcode@catcode.com>
Cc: Thomas E Deweese <thomas.deweese@kodak.com>, Jim Mork <jmork@jasc.com>, "'www-svg@w3.org'" <www-svg@w3.org>
>>>>> "JDE" == J David Eisenberg <catcode@catcode.com> writes:

JDE> On Mon, 10 Sep 2001, Thomas E Deweese wrote:
>> >>>>> "TD" == Thomas E Deweese <thomas.deweese@kodak.com> writes:
>> 
>> >>>>> "JM" == Jim Mork <jmork@jasc.com> writes:
>> 
>> Actually this was the problem.  Without the enable-background
>> present, you end up 'compositing' the magenta rect with transparent
>> black (0, 0, 0, 0). The result of the combine is:
>>
>> [...]
>>
>> This is then composited with srcOver on top of the cyan rectangle
>> in sRGB, the result is:
>> 
>> (0.5, 0.5, 1.0, 1.0)
>> 
>> Giving the result you saw.

JDE> Is this the same problem that is exhibited in the following
JDE> (which shows a problem with Batik, but not in Adobe SVG viewer
JDE> version 2) and is a repost of a previous problem:

     Well, I wouldn't say the following shows a problem with Batik,
unless you consider implementing the specification a problem. :)

     I would also classify them as different problems anyway, his
basic problem is that he _expected_ the arithmetic combine to be done
in linear sRGB and he was in fact getting a _SrcOver_ in sRGB.  It
just happened that the SrcOver in sRGB was giving the same results as
the doing the arithmetic composite in sRGB would have given, thus
greatly confusing the situation.

JDE> This one had me confused for several hours.

     Sure dealing properly with color spaces can be tricky, however if
you take the time to properly understand what you are dealing with,
many 'hacks' become unneeded (like playing with the gamma of images
to get it to 'look right').

JDE> The solution is to write the filter so that it, too, works in the
JDE> correct color space:

     This is one solution and probably the correct one for your needs
since you are interested in inverting the perceptual code values.

     If you were trying to do more sophisticated image processing that
was more 'real world' based as opposed to pure digital effects you
would probably find the linear sRGB space more to your liking.
Received on Monday, 10 September 2001 13:02:55 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:20 GMT