[argyllcms] Re: DTP20 (Pulse) patch size error

  • From: "Hal V. Engel" <hvengel@xxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Mon, 15 Aug 2011 09:54:23 -0700

On Monday, August 15, 2011 02:44:18 AM Graeme Gill wrote:
> Philip Reed wrote:
> > Argyll doesn't ask to read the TID strip.  This is what I get:
> If you do an un-tethered set of readings (ie. using the TID),
> does it work any better ?
> 
> > The row patches from Argyll are 6.4 by 10 mm while the patches from
> > MonacoPROFILER are 6.7 by 12 mm.  The TID patches are 6mm by 10mm for
> > Argyll and 6mm by 12 mm for Monaco.
> 
> I'd be really surprised if a width difference of 2 mm is going to make
> any difference, given how much margin it has over the length.
> 
> Are the Argyll patches really 6.4 mm ? (ie. is the average over
> the length of all the patches in the strip ?).
> I wonder if there is a small scaling error in the printing,
> and it is enough to take the patch below its minimum.
> 
> > Is the instrument sensing bleed from the rows above and below that aren't
> > being masked using the Argyll chart?  About a mm of the row above and
> > below isn't covered.
> 
> Unlikely, since the aperture must be less than 6 mm if it is to read
> the patch cleanly.
> 
> Looking at the test chart spec., it is possible to change the DTP20
> patch length to some other values, although I haven't currently allowed
> for it. I'll see about adding this for the next release.
> 
> Graeme Gill.

My experience with the DTP20 and Argyll is that the charts had to be printed 
in a certain very specific way to get the readings to work.  The issues that 
the OP describes are exactly what I was seeing before I figured out how to 
correctly print the target.  

I don't remember exactly what I had to do to get this working but I think it 
had something to do with scaling and/or borders.  SInce the OP is on a Windows 
machine and I was using Linux what I had to do may not be directly applicable.  
But I had to spend some time figuring out exactly how to print the target to 
get this working.  But I did get it working after printing perhaps 4 or 5 test 
targets.

Hal

Other related posts: