Re: Printing profile test targets WITHOUT photoshop
Re: Printing profile test targets WITHOUT photoshop
- Subject: Re: Printing profile test targets WITHOUT photoshop
- From: James Horne <email@hidden>
- Date: Wed, 19 Aug 2009 21:44:22 +1000
- Thread-topic: Printing profile test targets WITHOUT photoshop
I had a little play with Klaus suggestion. I can configure the printer
driver through CUPS with an appropriate media selection for the paper I wish
to profile and colour management (or lack of) settings. I wasn't sure if it
was suggested in fun or not but part of the solution seems sensible, to me
at least. I wonder though what happens in the conversion to postscript in
tiff2ps, presumably using libtiff, to rendering of colours. I might print
my colour test chart out that way and again through my current method with
Preview and see if there is any colour shift. If there is none it would
probably confirm both methods are leaving the colour alone but if there was
a difference it wouldn't tell me which method was changing it, only that
there was a difference.
Anyway, the immediate annoyance with submitting the tiff2ps postscript
output was that it printed in portrait orientation and went off the edges of
the paper and AFAIK one can only manage a 180 degree rotation. I had a play
with Imagemagick "convert" function to achieve the same but there are
different issues with scaling. My profiling service insists the colour
charts do not get scaled, just 100% size centred on an A4 page. They don't
really suggest converting to postscript either..........but hey you only
live once!
I'll probably have a little more play with the postscript conversion, at
least until somebody wiser tells me I'm being a fool!
Cheers
J
On 19/8/09 6:19 PM, "Graeme Gill" <email@hidden> wrote:
> edmund ronald wrote:
>> I can't be bothered to look at the Generic RGB Apple Profile, and I'm
>> too dumb to figure out what is in a profile anyway - but maybe some of
>> our tech geniuses here can tell us what it is, or just make us a
>> genuine "do nothing" profile of some sort to act as a placeholder ?
>
> It doesn't work like that. The only type of real device profile
> that will "do nothing" is the same profile that will be used
> as the destination, the very (fragile) system that is now
> (it seems) not working so reliably.
>
> You can make a device link "do nothing" profile, but this isn't
> going to be a valid way of tagging an image made of device values.
>
> The correct, robust method is to tag the image with a flag
> that says "Image is in output device space, do not color manage",
> but unless someone like the ICC comes up with a standard that
> uses the ICC format for such a thing (a meta profile), it is
> still going to be printing system or operating system
> and print file format dependent.
>
> [ If it were up to me there would be two flags:
> Do not color transform.
> Do not use calibration.
> So that there was a way of tagging calibration system and
> profiling test charts in a way that guarantees that they
> are printed properly. That's what we did with Postscript
> comments on the DICENet/Cyclone, and it worked very reliably.]
>
> Graeme Gill.
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> Colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
> au
>
> 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