Re: Humans (and cameras and scanners) do not have a color.. simple answer
Re: Humans (and cameras and scanners) do not have a color.. simple answer
- Subject: Re: Humans (and cameras and scanners) do not have a color.. simple answer
- From: Wire ~ via colorsync-users <email@hidden>
- Date: Fri, 10 Jan 2020 16:35:52 -0800
Marc, yes!
That was the system: the scanner profiling was from the same pass as the
printed swatches were measured, then the calculator could be agnostic to
the scanner, under certain reasonable assumptions such as the scanner has
sufficient color resolution and linearity (cough) to distinguish the gamut
of patches (cough, harrumph!) and everything gets referred to the matching
IT8 data file.
And the brilliance of this setup—it's genius!—was that it trusted that ICC
color tech actually works. It ate its own digital dog food, so to speak.
Thank you Marc
On Fri, Jan 10, 2020 at 4:24 PM Marc Levine <email@hidden> wrote:
> Hi list,
> Here’s the deal with MonacoEZcolor (MEC): the product Wire is asking
> about.
>
> MEC does use an uncalibrated scanner. However, there’s a trick involved.
> For that product, you would output your printer profile chart with color
> management off. The printed chart has a ghosted “scanner it8 image”, to
> which you would tape your photographic IT8 (came with the software). You
> would then scan the whole the whole thing with all the smarts turned off
> and feed that into the software.
>
> The software would:
> 1) collect the scanned RGBeez from the taped-on photographic IT8
> 2) look up the reference lab values for the taped-on IT8
> 3) using those 2 sets of data, build an input profile for the scanner.
> Note: you could save that profile as the scanner profile, effectively
> making the scanned “calibrated”
> 4) collect the RGBeez from the imaged of the printed output target (same
> scan)
> 5) convert that second set of RGBeez to Lab using the scanner profile (now
> you have Lab... just like it you measured it)
> 6) use the “simulated lab” and the output target chart def (a bunch of
> RGBeez or CMYKays, depending on your print driver) to build your printer
> profile.
>
> It was pretty trick and often worked really well (sometimes too well!) it
> wasn’t perfect due to the constraints of the lightsource of the scanner. In
> those days, inkjet inks were not as mature as they are today. Sometimes
> they were a little wonky in their spectral characteristics... which made
> them look really different when illuminated by the scanner lamp compared to
> a bonafide spectro.
>
> I never thought of this back then.... but I guess you could have even made
> a transmissive print profile provide using transparent film it8 and
> scanning the whole thing transmissively. I must have a copy of that
> software around here somewhere!
>
> ; )
>
> A little long winded... but hopefully clear.
>
> -Marc
>
_______________________________________________
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