[Openicc] Google Summer of Code 2009: Oyranos CUPS backend

Leonard Rosenthol leonardr at pdfsages.com
Mon May 17 05:48:31 PDT 2010


One thing we are working on at Adobe, that fits into this work, is a
standard way to have printer manufacturers provide an actual ICC profile for
their device that can be used as a real world "Output Intent".   Without
such a profile, and with the increase usage around host-based CM, we are
finding too many color shifting issues on various devices that expect colors
to be "left alone".  In addition, we believe that information about
rendering intent preference are also useful.

So before you go hacking around the PPD format, you probably want to involve
the various printer manufacturers and folks such as us, who are producing
the PPDs that are distributed with the various OS platforms and devices, and
whose applications also need to work with them.

Leonard Rosenthol
Adobe Systems


On Sun, May 16, 2010 at 6:03 AM, Kai-Uwe Behrmann <ku.b at gmx.de> wrote:

> Hello,
>
> the CUPS backend needs to tell Oyranos, which PPD colour related entries a
> PPD has. These entries are used to select a meaningful user configured ICC
> profile.
>
> Could you imagine a PPD entry, which solely lists all colour related
> entries? Something like:
> *colourKeyWords:
> "ColorModel;StpColorPrecision;MediaType;StpQuality;Resolution;StpInkSet;StpInkType;StpBrightness;StpFineBrightness;StpContrast;StpFineContrast;StpSaturation;StpFineSaturationStpImageType"
> Note this example is incomplete.
>
> A disadvantage might by that it has to be included into every PPD. For a
> driver with a large number of supperted printers and hence many PPDs like
> the Gutenprint one, it would mean to repeat that stuff. The advantage is a
> included PPD entry would be very simple to parse along the CUPS delivered
> PPD.
>
> Would this path be acceptable for foomatic, Gutenprint and hplib and other
> vendors?
>
>
>
> kind regards
> Kai-Uwe Behrmann
> --
> developing for colour management www.behrmann.name + www.oyranos.org
>
>
> Am 19.08.09, 11:59 +0200 schrieb Kai-Uwe Behrmann:
>
>  Am 19.08.09, 11:16 +0200 schrieb Till Kamppeter:
>>
>>> Kai-Uwe Behrmann wrote:
>>>
>>>> What I do not know is, how to distinct the colour related PPD options
>>>> from
>>>> the non relevant to make selection dependent on say "gamma" but not on
>>>> "paper size". Would you know a way how to select ppd_attr_t attributes
>>>> according to their influence on colour?
>>>>
>>>
>>> Principally one cannot read from a PPD option whether it influences color
>>> or
>>> not.
>>>
>>
>> This will it make harder to get PPD to ICC profile mapping outside of the
>> CUPS cupsICCProfile scheme. The need of root access to do proper colour
>> management is yet clearly a limit.
>>
>> On the other side Oyranos needs to know if a PPD option or attribute is
>> relevant for colour or not to do CM configuration outside of the
>> cupsICCProfile scheme.
>>
>> E.g. a user printing to remote printer or on a system without root rights
>> she can not locally assign a ICC profile to a given PPD in the Oyranos
>> DB and keep flexibility without complicated tweaking. But exactly that was
>> the goal we where aiming at in this years GSoC projects.
>>
>>  ICC profiles can be associated to choices of up to three options, by
>>> default
>>> to ColorModel, MediaType, and Resolution:
>>>
>>> *cupsICCProfile ColorModel.MediaType.Resolution/Description: "filename"
>>> *cupsICCProfile RGB.GlossyPhoto.1200dpi/High-Quality Photo on Glossy
>>> Photo
>>> Paper: "Acme-PhotoStar-RGB-GlossyPhoto-1200.icc"
>>>
>>> Instead of MediaType and Resolution two other options can be used with
>>> definitions like:
>>>
>>> *cupsICCQualifier2: Quality
>>> *cupsICCQualifier3: ImageType
>>>
>>> See http://www.cups.org/documentation.php/doc-1.4/spec-ppd.html
>>>
>>
>> Yes thats the next stage to support in order get the backend functional
>> with cupsICCProfile.
>>
> _______________________________________________
> openicc mailing list
> openicc at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/openicc
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/openicc/attachments/20100517/54977ea6/attachment.htm>


More information about the openicc mailing list