W3C home > Mailing lists > Public > www-svg@w3.org > July 2012

Re: Compositing math in SVG

From: Rik Cabanier <cabanier@gmail.com>
Date: Sat, 28 Jul 2012 16:06:36 -0700
Message-ID: <CAGN7qDALwm8Pew3B9n558VjxmcghKzR9BK2TjPKaCHiMx4=PZA@mail.gmail.com>
To: Calculemus <calculemus1988@gmail.com>
Cc: Dirk Schulze <dschulze@adobe.com>, "steve@fenestra.com" <steve@fenestra.com>, "www-svg@w3.org" <www-svg@w3.org>
On Sat, Jul 28, 2012 at 2:44 PM, Calculemus <calculemus1988@gmail.com>wrote:

> Wooow, thanks Rik! :)
>
> I have some questions:
>
> 1. Why don't people mention this when they write about blend
> modes? :) I read even the Adobe PDF reference and it was also
> talking about the compositing formula the way I did it.
> Where can I read more about this?
>

PDF explicitly doesn't have color spaces with alpha so it didn't have to
cover this case. I think it was excluded because it causes exactly the
confusion you are facing.

In the case of a soft image (which what you have) you can find the
description in the PDF reference manual (
http://wwwimages.adobe.com/www.adobe.com/content/dam/Adobe/en/devnet/pdf/pdfs/PDF32000_2008.pdf)
in chapter 11.6.5.3


>
> 2. I am confused about the second case. Does not alpha
> always come from the images?


No, for most cases alpha will come from the 'color' or 'opacity' value.


> It is kinda strange to think
> of a blend mode with some alpha, while both images have
> alpha 1.
>
> It seems what I want always is the first case, since alpha
> will always come from the images. Not sure when would
> I need the second case though?
>

If you look at the layer palette in Photoshop, you will see an entry for
'alpha'.
If you want to implement that or you want to apply alpha to a group of
images, you need the second case.


>
> I was so excited when I saw this, I wrote this quickly, I am
> going to the shop to get some coffee and chocolate and
> I am getting back to try if this change will fix my problems :D
>

I am getting myself a beer :-)
Received on Saturday, 28 July 2012 23:07:05 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:51 GMT