[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] Gamma correction and texImage2D/texSubImage2D



 On 07/09/2010 01:31, Thatcher Ulrich wrote:
> I'll say it again, because this is sticking in my craw -- linear 8-bit
> RGB is a mistake.  I put together a page showing the type of artifacts
> that will result. http://tulrich.com/webgl/rgb/linear_vs_srgb.html
>
Specifying the canvas to internally be sRGB is mathematically incorrect.
I agree very strongly with Steve Baker that we should not enshrine in
the standard something that is mathematically incorrect. I don't think
anybody said the spec. should require linear 8-bit. Only linear.
Implementations should be free to use more bits. 12 per channel is what
is actually necessary to avoid banding and is increasingly possible in
hardware.

Is banding more or less evil that the artefacts that come with a
non-linear  space? Artefacts include: in lighting, loss of shader
details. in blending incorrect colors and in anti-aliasing insufficient
smoothing. These artefacts become even more pronounced with high-dynamic
range imaging.

As a practical matter the spec. will probably have to allow
implementations to do everything in a non-linear space but requiring it
be so is absolutely the wrong thing to do.

Regards

    -Mark
begin:vcard
fn:Mark Callow
n:Callow;Mark
org:HI Corporation;Graphics Lab, Research & Development
adr:Higashiyama 1-4-4, Meguro-ku;;Meguro Higashiyama Bldg 5F;Tokyo;;153-0043;Japan
email;internet:callow_mark@hicorp.co.jp
title:Chief Architect
tel;work:+81 3 3710 9367 x228
tel;fax:+81 3 5773 8660
x-mozilla-html:TRUE
url:http://www.hicorp.co.jp,  http://www.mascotcapsule.com
version:2.1
end:vcard