ArgyllCMS V1.6.3 has been released.

[Graeme Gill]

This is a bug fix release.

See <http://www.argyllcms.com>.

Graeme Gill.

[m***@public.gmane.org]

Hi,
With these commands:
targen -v -d4 -p1.88 -s200 -e4 -f0 Printer_calibration
printtarg -v -ii1 -R123 -pA4 Printer_calibration
in version 1.6.2 printtarg output:
....
After optimisation, worst delta E = 66.372105
Worst case direction distinction delta E = 61.873128
In version 1.6.3 i have (the same commands):
....
After optimisation, worst delta E = 21.000236
Worst case direction distinction delta E = 21.914342
Is such a difference normal?
Michał

[Ivan Tsyba]

I try to measure printer chart that created by ArgyllCMS 1.4.0 using this
version and always get "Strip read failed due to misread (Not enough
patches)". 1.4.0 chartread measure this chart without problem. I'm using
Colormunki spectro.

My command: chartread -v -H RGB_210x297_529_1_page(s).ti2
This ti2 file attached to this message

Something changed between 1.4.0 and 1.6.3 and this cause my problem?

Post by m***@public.gmane.org
Hi,
targen -v -d4 -p1.88 -s200 -e4 -f0 Printer_calibration
printtarg -v -ii1 -R123 -pA4 Printer_calibration
....
After optimisation, worst delta E = 66.372105
Worst case direction distinction delta E = 61.873128
....
After optimisation, worst delta E = 21.000236
Worst case direction distinction delta E = 21.914342
Is such a difference normal?
Michał

[Xavi]

Hello.



I also have a ColorMunki but I have not installed the 1.6.3 version but I installed the 1.6.0 and if I read the strips fast I receive some errors.



I think that Graeme Hill changed the maximum read speed of the ColorMunki, to avoid electrical interferences.



Have you tried to read the strip slowly?



Xavi.





From: argyllcms-bounce-***@public.gmane.org [mailto:argyllcms-bounce-***@public.gmane.org] On Behalf Of Ivan Tsyba
Sent: Friday, January 31, 2014 10:43 PM
To: argyllcms-***@public.gmane.org
Subject: [argyllcms] Re: ArgyllCMS V1.6.3 has been released.



I try to measure printer chart that created by ArgyllCMS 1.4.0 using this version and always get "Strip read failed due to misread (Not enough patches)". 1.4.0 chartread measure this chart without problem. I'm using Colormunki spectro.

My command: chartread -v -H RGB_210x297_529_1_page(s).ti2

This ti2 file attached to this message



Something changed between 1.4.0 and 1.6.3 and this cause my problem?



2014-01-31 mbielowicz-h7QdYz1kt/***@public.gmane.org <mbielowicz-h7QdYz1kt/***@public.gmane.org>:

Hi,

With these commands:
targen -v -d4 -p1.88 -s200 -e4 -f0 Printer_calibration
printtarg -v -ii1 -R123 -pA4 Printer_calibration

in version 1.6.2 printtarg output:
....
After optimisation, worst delta E = 66.372105
Worst case direction distinction delta E = 61.873128

In version 1.6.3 i have (the same commands):
....
After optimisation, worst delta E = 21.000236
Worst case direction distinction delta E = 21.914342

Is such a difference normal?

Michał

[Ivan Tsyba]

Post by Xavi
Have you tried to read the strip slowly?

Yes, with _very_ slow reading this problem not occurs. I need move
instrument three time slower than before.
-T option of chartread is deal with this? Or for better readings I need
move device much slower than before?

[Graeme Gill]

Post by Ivan Tsyba
instrument three time slower than before.

That would be about right - the scan rate has been reduced
by about a factor of 3 to improve the black patch accuracy.
This is the same sample rate that the X-Rite driver
uses.

Graeme Gill.

[Graeme Gill]

Post by Ivan Tsyba
I try to measure printer chart that created by ArgyllCMS 1.4.0 using this
version and always get "Strip read failed due to misread (Not enough
patches)". 1.4.0 chartread measure this chart without problem. I'm using
Colormunki spectro.
My command: chartread -v -H RGB_210x297_529_1_page(s).ti2
This ti2 file attached to this message
Something changed between 1.4.0 and 1.6.3 and this cause my problem?

Hi,
1.4.0 is 6 releases ago, so yes, things may well have changed.

I think the reduction is scan speed by a factor of 3 to 55Hz made in V1.6.1
may have made it less able to be read at high speed, but greatly improved
it's dark patch reading accuracy. There simply aren't enough samples within
the patch itself to clearly distinguish the edges of the patch when
read too fast. If this is the explanation, then there are only a couple
of possible solutions - the simplest is to reduce your reading pace.
The other would be to increase the length of the patches, so that
there are relatively more samples within the patch than over the patch
transitions. I suspect that the overall chart read time would end
up about the same though, at the expense of more paper used and
strips to read.

Graeme Gill.

[Graeme Gill]

Post by m***@public.gmane.org
targen -v -d4 -p1.88 -s200 -e4 -f0 Printer_calibration
printtarg -v -ii1 -R123 -pA4 Printer_calibration
....
After optimisation, worst delta E = 66.372105
Worst case direction distinction delta E = 61.873128
....
After optimisation, worst delta E = 21.000236
Worst case direction distinction delta E = 21.914342
Is such a difference normal?

Hi,
I don't seem to be able to reproduce this. I get
the same numbers as 1.6.3 for releases back to 1.5.0 and before,
given the same .ti1 file - ie. nothing seems to have changed
with printtarg.

Note that the effect of targen -p has been made more uniform (it wasn't
being applied very consistently), so you will get slightly different .ti1
files out of 1.6.2 and 1.6.3, so maybe this is what is changing
the contrast numbers.

Graeme Gill.

[m***@public.gmane.org]

Hi,

Graeme you're right, this effect occurs only when targen -p is used.

cheers
Michał

Post by Graeme Gill

Post by m***@public.gmane.org
targen -v -d4 -p1.88 -s200 -e4 -f0 Printer_calibration
printtarg -v -ii1 -R123 -pA4 Printer_calibration
....
After optimisation, worst delta E = 66.372105
Worst case direction distinction delta E = 61.873128
....
After optimisation, worst delta E = 21.000236
Worst case direction distinction delta E = 21.914342
Is such a difference normal?

Hi,
I don't seem to be able to reproduce this. I get
the same numbers as 1.6.3 for releases back to 1.5.0 and before,
given the same .ti1 file - ie. nothing seems to have changed
with printtarg.
Note that the effect of targen -p has been made more uniform (it wasn't
being applied very consistently), so you will get slightly different .ti1
files out of 1.6.2 and 1.6.3, so maybe this is what is changing
the contrast numbers.
Graeme Gill.