Announcing The Correct Color Channel

[Correct Color]

After an expenditure of quite a bit of time and energy, here finally it is...

A Youtube channel that will be dedicated exclusively to large and grand format inkjet printing. With a take and a perspective unlike anything you've ever seen before.

Check it out.

And please offer any thoughts you might have.


Mike Adams
Correct Color

 

[bteifeld]

Thank you for making these videos- I am waiting with anticipation to learn about
your views of G7. I viewed both your current videos and found them helpful.

A suggestion, should you ever consider revising or improving the existing videos,
or making one to specifically address one improvement-

When you display a comparison between two or more conditions that show what
to some may be a subtle difference in color outcome- I would suggest you display
them side by side, with arrows or some other markers identifying the points of
comparison. Currently you switch frames, which requires viewers to remember
what they previously saw, or remember where the previous image was to
enable them to review the video and switch between them.

I would also appreciate if you could develop a video which would amplify
and explain the use of evaluation imagery such as ink channel gradient
views and Grainger Rainbows.

Also, with regard to your mention of Monaco Profiler- this sadly is a now
deprecated product that requires a 16-bit Windows environment to run.
I hope you might one day discuss a comparison and review of existing
engines, both in-rip as well as general(i1profiler, basiccolor, copra).

Once again, thank you for your time and efforts.

 

[Correct Color]

I am waiting with anticipation to learn about your views of G7.

If you don't want to wait for the video version:

Certified Correct Color Vs G7 For Large Format Printing

When you display a comparison between two or more conditions that show what
to some may be a subtle difference in color outcome- I would suggest you display
them side by side, with arrows or some other markers identifying the points of
comparison. Currently you switch frames, which requires viewers to remember
what they previously saw, or remember where the previous image was to
enable them to review the video and switch between them.

That's an excellent suggestion. I wish I'd thought to do that. This was a presentation I originally made to do live, and I'd point out the areas in question. But even then, that would have been more than worthwhile. I'd go back and alter it, but that project at this point has consumed me enough.

Also, with regard to your mention of Monaco Profiler- this sadly is a now
deprecated product that requires a 16-bit Windows environment to run.
I hope you might one day discuss a comparison and review of existing
engines, both in-rip as well as general(i1profiler, basiccolor, copra).

Believe me, I'm aware. I have to go to great lengths to keep using it -- I run it in Windows XP through Parallels on a Mac -- and I can understand why most people wouldn't go to the trouble.

Thanks for the input.


Mike

 

[bteifeld]

Thank you for bringing to my attention your comments on G7 in writing.
I did look them over- but I am confused about something.

I use the Ergosoft RIP and was considering G7 given the claims regarding
grey balance. In the process described in the Curve4 manual, I am not sure
whether your comments about how G7 forces you to leave gamut on the table
apply in the context of the Ergosoft RIP.

My reason for mentioning/asking about this is that within the Ergosoft RIP, you
do the base linearization, per-channel ink limiting, and total ink limiting to
establish the base conditions for work with Curve4. You then use Curve4 to
generate its data, which in the context of the Ergosoft RIP are incorporated
in the target density settings step at the end of the linearization/per-channel
ink limiting/total ink limiting process. For the extended gamut inks, you use the SCTV
procedure in Curve4 and similarly incorporate its results in the target
density settings.

Two questions:

Is all this in the end causing loss of gamut and no real benefits?

Are the grey balance claims of the G7 proponents with regard to profile
efficacy improvement little more than snake oil?

I realize in writing this that my tone is at risk of misperception- please assume
I am sincerely asking the questions with an intent to understand.

 

[Correct Color]

No offense taken, I assure you.

To answer:

Is all this in the end causing loss of gamut and no real benefits?

Well, if you do G7 correctly, it shouldn't necessarily decrease your gamut.

What I meant when talking about leaving gamut on the table is that achieving maximum gamut is not part of the G7 routine. You can G7 a machine, but nothing in the process requires you to set the ink limits to achieve maximum gamut. I'm not implying that you can't do it. But only if you -- or the three-day "expert" you hired -- know how. What you can do is achieve way less than optimum gamut and inking characteristics and still pass G7.

Are the grey balance claims of the G7 proponents with regard to profile
efficacy improvement little more than snake oil?

Honestly... Yes.

There's much more to it, and there'll be more in the video. But G7 is a technique that was developed to address a specific issue in offset lithography years ago when plate-making was an analogue process. It's an issue that never even existed in digital printing.

The fact is that G7, as originally envisioned and in its original release was designed to be used as a standard calibration routine on offset presses, so that they could then run standard profiles, e.g: GRAcol; SWOP, and avoid the profiling process altogether.

Once you make an ICC profile, it is the profile that tells the RIP what dots to generate, rendering G7 as just another form of linearization. Done correctly it works fine, but it has no particular benefit over any other linearization.

In fact, one of the most robust linearization routines in any RIP is the one in Ergo. You do that right and no way you're gonna beat that with G7.

And if you're asking, sellers of G7 have been making the claim of late that using G7 makes profiles "last longer" which is absurd.

Profiles last forever. Machines either drift... or they don't. The components of how a profile was made do not change that at all.



Mike

 

[bteifeld]

Thank you for your replies, and especially the affirmation of the robustness of the linearization routine in Ergosoft.

I remain looking forward to your future videos.

 

[gordo]

@ Correct Color - Could you explain your use/meaning of the term "linearization"?

 

[pauly92]

Nice sets of videos Mike,

Hopefully you keep creating and possibly make shorter video (~5-10mins) on specific topics.

Would love to hear your opinions on different ICC engines. I've tried a fair few and i have my opinions on all of them.
I've seen some require a specific set of data to produce good profiles, i've seen some produce very consistent profiles regardless of what's done.

Took some time, but i've found the tools i need and that fit my workflow and more than happy with them.

 

[Correct Color]

Gordo,

Well, you caught me using shorthand... again, as I recall.

But in the video, I define linearization as, "The process of bringing a printing device into a calibrated state to compensate for dot gain."

And on the website, I describe G7 as follows, "What G7 actually is is a linearization/calibration routine. Now they call it a “grey balance” routine, and it is in the sense that it arrives at its targeted values by reading composite grey patches as opposed to individual patches of each colorant; but it’s done in the linearization phase of the media profile-making process, and its results become the linearization file which becomes a part of the machine state of which the ICC profile is then a characterization."

And, yes, there's the whole issue of just what a "linearization" is, and just what a "calibration" is. I recall you and I have had this discussion before. And the way I describe it is to say that to be technically correct, all linearizations are calibrations, but not all calibrations are linearizations.

However, in all large format inkjet RIP's, the calibration phase is referred to as linearization, whether the end result is linear curves or not. So speaking here strictly to large format inkjet printers, I use what has become a shorthand term for calibration in this end of the industry.


Mike

 

[gordo]

That right there is a continuing major problem in this industry. This general misuse of terminology leads to misunderstanding, confusion, and wrong thinking.
BTW, that wasn't a shot at you by a new subscriber to your channel.

 

[Correct Color]

That right there is a continuing major problem in this industry. This general misuse of terminology leads to misunderstanding, confusion, and wrong thinking.
BTW, that wasn't a shot at you by a new subscriber to your channel.

I agree completely, actually.

In the videos, I mention that problem. Right now the top of my worst-offender list is the term "greyscale."

You may not have seen this, but inkjet printhead and printer manufacturers have co-opted that term to mean multi-dot, calling multi-dot printheads "greyscale" printheads. Why they needed to use a term that has a long-standing completely different meaning in this industry is beyond me.

So, okay: Calibration.


Mike

 

[PricelineNegotiator]

Subscribed. Haven't watched the videos on there yet, but I will. Good on you for making a channel.

 

[gordo]

I agree completely, actually.

In the videos, I mention that problem. Right now the top of my worst-offender list is the term "greyscale."

You may not have seen this, but inkjet printhead and printer manufacturers have co-opted that term to mean multi-dot, calling multi-dot printheads "greyscale" printheads. Why they needed to use a term that has a long-standing completely different meaning in this industry is beyond me.

So, okay: Calibration.


Mike

Like using “resolution” instead of “addressability” because it makes it sound as if the printers have a higher specification than they actually do.

 

[Correct Color]

Like using “resolution” instead of “addressability” because it makes it sound as if the printers have a higher specification than they actually do.

I will say that one's pretty far into the weeds if you ask me. In fact I've never had a client even use the term addressability. I haven't even seen "perceived resolution" in years

Day-to-day, what I see as the biggest offenders are dpi being used to describe ppi, and the term "profile" having dual uses, as either a media profile, or an ICC profile. I've been tripped by that one more than once.


Mike

 

[gordo]

I will say that one's pretty far into the weeds if you ask me. In fact I've never had a client even use the term addressability. I haven't even seen "perceived resolution" in years.

You see "resolution" used in vendor tech sheets and advertising:
e.g.:

"Printing Technology FINE: Full-Photolithography Nozzle Engineering Head Configuration 12-Channel Integrated Type Maximum Print Resolution Up to 2400 x 1200 dpi"

"dpi" is not resolution - it is addressability. But you're right, most clients probably never heard of addressability.

 

[Correct Color]

Ohhhhh...

I get what you're saying. I actually wrote out a reply and then it dawned on me.

Only thing I'd say though is that when you print a sheet, and hold it in your hands and look at it, it was printed with a specific number of dots per inch. So from that standpoint, dpi is a correct term.

And in working with the machines, and profiling the machines, it's the dots that matter. So addressability is kind of beside the point. Once the ink is on the media, it's dots, and there has to be a way to measure and quantify them, and that's dpi.


Mike

 

[gordo]

Ohhhhh...

I get what you're saying. I actually wrote out a reply and then it dawned on me.

Only thing I'd say though is that when you print a sheet, and hold it in your hands and look at it, it was printed with a specific number of dots per inch. So from that standpoint, dpi is a correct term.

And in working with the machines, and profiling the machines, it's the dots that matter. So addressability is kind of beside the point. Once the ink is on the media, it's dots, and there has to be a way to measure and quantify them, and that's dpi.


Mike

Maybe it should be "spi" - splats per inch.
"dpi" doesn't tell you how large the splats of ink dots are. So, two printers can have the same dpi but the size of the splats of ink can be very different. This might not matter so much when printing images, however when printing text it can make a big difference in the integrity/clarity of the letters.

 

[pauly92]

Maybe it should be "spi" - splats per inch.
"dpi" doesn't tell you how large the splats of ink dots are. So, two printers can have the same dpi but the size of the splats of ink can be very different. This might not matter so much when printing images, however when printing text it can make a big difference in the integrity/clarity of the letters.

You're talking about Dot-Gain.

 

[gordo]

You're talking about Dot-Gain.

No. I’m talking about picoliters since the OP is referring to inkjet printers.

 

[Correct Color]

Maybe it should be "spi" - splats per inch. "dpi" doesn't tell you how large the splats of ink dots are. So, two printers can have the same dpi but the size of the splats of ink can be very different.

Well...

That's what linearization is for.


Sorry. Couldn't resist.