[argyllcms] Re: Homebrew ColorChecker

  • From: Pascal de Bruijn <pmjdebruijn@xxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sat, 13 Feb 2010 20:36:54 +0100

On Sat, Feb 13, 2010 at 8:21 PM, Ben Goren <ben@xxxxxxxxxxxxxxxx> wrote:
> On 2010 Feb 13, at 11:55 AM, Pascal de Bruijn wrote:
>
>> And another welcome advantage of  using something that is near a
>> ColorChecker is that lots of utilities can recognize the chart.
>
> Well...yes and no.
>
> Those  utilities  assume  that  the  ColorChecker  has  a  certain
> spectral  characteristic, which  your  chart  won't have. That  is
> to  say,  you've created  a  metameric  match under  D50  lighting
> conditions,  but it's  a  near-certain guarantee  that your  chart
> won't match a ColorChecker under fluorescent lighting.
>
> It would  be educational  for you,  I think,  to compare  plots of
> the  spectra  of  your  chart  with the  spectra  from  an  actual
> ColorChecker.

I know about spectra and metameric matches... I'm _guessing_ a strobe
is practically similar-ish to D50...

And I only talked about chart recognition. To get any results at all
an application would have to use my custom measured reference files...

The inkjet prints aren't close enough in any case to be used with the
original reference files...

Regards,
Pascal de Bruijn

Other related posts: