W3C home > Mailing lists > Public > whatwg@whatwg.org > May 2007

[whatwg] Canvas operators (was Re: several messages)

From: L. David Baron <dbaron@dbaron.org>
Date: Sun, 20 May 2007 15:30:15 -0700
Message-ID: <20070520223015.GA16523@ridley.dbaron.org>
On Sunday 2007-05-20 17:53 +0100, Philip Taylor wrote:
> If you care about alpha, and take cX as premultiplied colour
> components (and CX as non-premul, and aX as alpha), then you get:
> Mozilla:
>  cO = cA * min(1, (1-aB)/aA) + cB
>  aO = aA * min(1, (1-aB)/aA) + aB

Correct.  This is the cairo "saturate" operator, which I think is
the same as Xrender's saturate operator.

> Safari:
>  CO = 1 - (1-CA)*aA - (1-CB)*aB  except not quite
>  aO = aA + aB

I thought what Safari does is basically "lighter", but with the
color components inverted (where inverting is CX = 1-CX, or cX = aX
- cX), so that it comes out to:

 aO = min(1, aA + aB)
 cO = aO - min(1, aA - cA + aB - cB)

> you change alpha from 0.0 to 0.01. In all the implementations,
> 'lighter' is confusingly unlike 'darker' - names like 'plus' and
> 'multiply' make much more sense.)

Hrm.  I thought lighter was confusingly *like* darker, in that they
act exactly the same when the alphas add to less than or equal to 1.
(I think that's true for plus (a.k.a. lighter), Cocoa's darker, and
for saturate.)

-David

-- 
L. David Baron                                <URL: http://dbaron.org/ >
           Technical Lead, Layout & CSS, Mozilla Corporation
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/attachments/20070520/b9486d16/attachment.pgp>
Received on Sunday, 20 May 2007 15:30:15 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 16:58:55 UTC