Re: Possible ColorSync discontinuity bug introduced in OS 10.14.6 with display "r/g/bTRC" tags?
Re: Possible ColorSync discontinuity bug introduced in OS 10.14.6 with display "r/g/bTRC" tags?
- Subject: Re: Possible ColorSync discontinuity bug introduced in OS 10.14.6 with display "r/g/bTRC" tags?
- From: Wire Moore via colorsync-users <email@hidden>
- Date: Sun, 18 Aug 2019 12:28:04 -0700
Would someone restate the problem being discussed I don't have access to
the earlier posts
Thanks
On Wednesday, August 14, 2019, William Hollingworth via colorsync-users <
email@hidden> wrote:
> On 8/14/2019 11:57 AM, Neil Barstow wrote:
>
>>
>> So, am I reading this right please?
>>
>> An LUT profile would be OK and a "MATRIX" not, in this setting.
>>
>> Is an LUT what you are calling "encoded as - - an array of points"
>> I guess a 16 bit LUT may be OK? I am just thinking about basICColor
>> display.
>>
>
> The rTRC. bTRC and gTRC tags in the display profile are "curv" types
> encoding the tone reproduction curve of each color. I believe LUT based
> display profiles must also include TRC tags, but maybe someone like Graeme
> Gill could step in and confirm.
>
> According to the ICC.1:2001-04 spec:
>
> ----------------------------------
>
> 6.5.3 curveType- The curveType contains a 4-byte count value and a
> one-dimensional table of 2-byte values.
>
> The count value specifies the number of entries in the curve table except
> as follows:
>
> - when countis 0, then a linear response (slope equal to 1.0) is assumed,
>
> - when countis 1, then the data entry is interpreted as a simple gamma
> value encoded as a u8Fixed8Number. Gamma is interpreted canonically and
> notas an inverse.
>
> Otherwise, the 16-bit unsigned integers in the range 0 to 65535 linearly
> map to curve values in the interval[0.0, 1.0].
>
> ----------------------------------
>
> The problem occurs with the case of using a single value to represent a
> simple gamma value in the profile. In this case ColorSync seems to mess up
> curve generation. It looks like most software nowadays is encoding as an
> array of points - which is probably how this one slipped by Apple.
>
> I would have thought that especially canned profiles would use the simple
> gamma rather than an array of points. I'd be interested in knowing the
> reasoning for not doing this.
>
> Thanks
>
> Will
>
>
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
> email@hidden
>
> This email sent to email@hidden
>
_______________________________________________
Do not post admin requests to the list. They will be ignored.
colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden