[Openicc] What is exactly needed: Embedded Profile in CUPS raster !!

[Jan-Peter Homann]

Hello Mike an all,
There is definetely a very import use case for "embedding arbitrary ICC 
device profiles into a CUPS raster stream":

We need a mechanism, that the colormanagement in "..toraster" is 
synchronized with the driver setup.

Such workflow must include the use case, that the user can download a 
printer profile and can easily configure it in his printing environment.

For this use case, the printer profile must have metadata included to 
allow a synchronization with driver settings after download. The most 
secure mechanism in my eyes is to embed the driver settings directly 
into the profile and define a mechanism for handing over the settings to 
the driver, when the profile is used.

As the printer profile is applied in "..toraster", we need a mechanism 
that the driver knows which profile has been used in "..toraster" and 
which driver setting should be used for printing.
If we embedd the printer profile from "..toraster" conversion into the 
CUPS raster file, the driver simply can extract the driver settings from 
the profile and use it for printing.

I currently don not see an alternative approach to solve the use case of 
downloading and configuring a printer profile.


Best regards
Jan-Peter


Am 01.06.11 18:03, schrieb Michael Sweet:
> On Jun 1, 2011, at 7:46 AM, Jan-Peter Homann wrote:
>> ...
>> My question:
>> How fast is it possible to extend the CUPS specifcation to allow 
>> embedding of an ICC in CUPS-raster format like in TIFF, JPEG or PNG ?
>
> Jan-Peter,
>
> It is not possible, and we will not extend CUPS raster to support 
> embedding of ICC profiles. CUPS Raster is a streaming raster format to 
> get pixels to the driver or printer - it is a "last mile" format. 
> There is no use case for embedding arbitrary ICC device profiles into 
> a CUPS raster stream since the printer and driver are not going to do 
> anything with the information, they will just render the device color 
> data provided by the upstream *toraster filter.
>
>
>>
>> Best regards
>> Jan-Peter
>>
>>
>> Am 01.06.11 15:40, schrieb Kai-Uwe Behrmann:
>>> On 01.06.2011 15:22, Michael Vrhel wrote:
>>>> Hi Jan-Peter,
>>>>
>>>> The embedding of the data's profile into the raster format is
>>>> something that should be in place by the next release of ghostscript
>>>> in August (9.03) .  I have an open enhancement request for this for
>>>> the TIFF, JPEG and PNG output devices of ghostscript.    A similar
>>>> thing could be done for the CUPs device, which I believe is the device
>>>> used for gstoraster.   However, I am not familiar with where or how
>>>> the data is packed into the CUPs raster.
>>> > From Till's comments, embedding a ICC profile into CUPS raster seems
>>> currently not possible. Even though I agree with Jan-Peter, that would
>>> have been a more clever solution.
>>>
>>>> My plan is to add in a new device parameter that can be set on the
>>>> command line to indicate if the profile should or should not be
>>>> embedded.   I need to decide what the default behavior should be (i.e.
>>>> to embed or not) .
>>> Tagging/embedding is a good thing. Otherwise the datas colour space is
>>> not defined, which will often become data loss. So generally speaking
>>> embedding should be the default.
>>>
>>> kind regards
>>> Kai-Uwe
>>>
>>
>>
>> -- 
>> ----------  Please note the new adress --------------
>>
>> homann colormanagement --------- fon +49 30 611 075 18
>> Jan-Peter Homann ------------ mobile +49 171 54 70 358
>> Cotheniusstr. 3 -------- http://www.colormanagement.de
>> 10407 Berlin -------- mailto:homann at colormanagement.de
>>
>>
>
> ________________________________________________________________________
> Michael Sweet, Senior Printing System Engineer, PWG Chair
>


-- 
----------  Please note the new adress --------------

homann colormanagement --------- fon +49 30 611 075 18
Jan-Peter Homann ------------ mobile +49 171 54 70 358
Cotheniusstr. 3 -------- http://www.colormanagement.de
10407 Berlin -------- mailto:homann at colormanagement.de


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/openicc/attachments/20110601/310838ab/attachment.htm>