Announcing The Correct Color Channel

pauly92

Well-known member
No. I’m talking about picoliters since the OP is referring to inkjet printers.
inkjet still has dotgain. Like you said 2 printers, same picoliters, but the "splats per inch" could be different. only thing that'll change that is dotgain.

I have 2 sample prints on my desk from 2 different UV printers Printer A & B are printed with the same image & resolution/DPI. same Picoliters. one looks more grainy vs the other looks more sharp.

fundamentally, that's how the canon colorado works with gloss and matte mode. ink takes longer to cure, ink droplets spread out a little more and flattens out. ever so slightly larger dots, but creates a gloss finish. VS instant curing, dots dont get time to flatten and smooth out producing a matte finish.
 

gordo

Well-known member
inkjet still has dotgain. Like you said 2 printers, same picoliters, but the "splats per inch" could be different. only thing that'll change that is dotgain.

I have 2 sample prints on my desk from 2 different UV printers Printer A & B are printed with the same image & resolution/DPI. same Picoliters. one looks more grainy vs the other looks more sharp.

fundamentally, that's how the canon colorado works with gloss and matte mode. ink takes longer to cure, ink droplets spread out a little more and flattens out. ever so slightly larger dots, but creates a gloss finish. VS instant curing, dots dont get time to flatten and smooth out producing a matte finish.

I didn't say that 2 printers, same picoliters, but the "splats per inch" could be different.

I said that two printers can have the same dpi but the size of the splats of ink can be very different.
 

gordo

Well-known member
OK, for clarity...
DPI - when used to describe the resolution of an inkjet printer. E.g. "This inkjet proofer prints at 2880 x 1440 dpi." does not mean that the resolution is finer, or that the droplets of ink are finer in one direction. Instead it simply means that the paper is moved more slowly in one direction - i.e. the addressability is changed - while the physical size of the droplet of ink, and hence its resolution remains the same.
In any case, the actual size of the mark the droplet of ink makes on the paper is unknown. For a 600 or 1440 "dpi" ink jet printer the "dot" it images most certainly is not 1/600ths or 1/1440th of an inch in size.

The addessability grid in the RIP of a printer is a basic X/Y locator map just like any wayfinding map.

Addressability.jpg


Each square, defined by its X/Y coordinate, is a target for each droplet of ink.
The individual squares in the grid may be set at, for example, 1200 per inch (dpi) but do not tell you the size of the dot/splat of ink on the substrate.

For example, this is the addressability grids of two different 1200 dpi printers .
But the dots/splats of ink on the left are 1.5 picoliters while the dots/splats on the right are 4 picoliters.

Splats.jpg


Although both are 1200 dpi the one on the left has a higher resolution than the one on the right.

As a result, with some inkjet printers, reference is sometimes made to "picoliters" in addition to dpi when the resolution of the device is described in the specifications. A picoliter is a unit of fluid volume. A lower minimum ink volume tends to yield a smaller minimum droplet size of ink allowing more dots of ink to be imaged discretely in the same area thereby yielding higher actual resolution. While picoliter is a better indicator of the relative size of the splat of ink on the paper it is still a unit of volume and not area. So it suggests a difference in resolution but doesn't actually specify it.

Dot gain is a different matter altogether.
 

Correct Color

Well-known member
I hope you don't mean that resolution is a function of linearization.
Well, no... That was a joke.

Actually, I go into this very subject at 14:20 in Part Two of Correct Color Management: The Movie.

However...

To start:

In any case, the actual size of the mark the droplet of ink makes on the paper is unknown. For a 600 or 1440 "dpi" ink jet printer the "dot" it images most certainly is not 1/600ths or 1/1440th of an inch in size.
Sure it is. Dots per inch means dots per inch, and that's precisely what it translates out to, since only at 100% coverage is a printhead only firing at each spot.

E.g. "This inkjet proofer prints at 2880 x 1440 dpi." does not mean that the resolution is finer, or that the droplets of ink are finer in one direction. Instead it simply means that the paper is moved more slowly in one direction...
Well, no. You are right that it doesn't mean the dots are any finer in the 2880 direction-- if by finer you mean smaller. What it does mean though is that in one direction the dots are generated at 2880 dpi, and in the other at 1440. The media transport speed has nothing to do with it.

As a result, with some inkjet printers, reference is sometimes made to "picoliters" in addition to dpi when the resolution of the device is described in the specifications.
There are some, notably the HP latex machines, who aren't real forthcoming with their dot size, but that's usually because their dots are a little on the largish side and they don't feel like sharing that with the class. I'm pretty sure though their dot size is around 10 pl, which, by the way, doesn't compare well with the 4.2 pl size of the dots on the Epson S80600, currently my favorite printer in that class. (My guess is the latex ink just won't fit through smaller nozzles, and they don't want to advertise that fact. Almost every other printer I'm aware of though, the drop size is in the spec sheet.)

But no matter, anyone who knows anything about inkjet printers knows that inkjet printer drop sizes are measured in picoliters.

Correct Color clients certainly do, and understand the significance.

A picoliter is a unit of fluid volume.
A picoliter is one trillionth of a liter.

Trillionth. That's a pretty small drop.

And currently, the smallest rated drop size out there on any commercially available printer is 3.5 picoliter on the Epson aqueous machines. There used to be some 1.5 picoliter printheads out there, but my understanding is they were very good at running dye-based ink, but not so good at running pigment-based ink... and dye-based ink does not last, while pigment-based ink does.

What you are right about is that drop size is equally as important as resolution in determining print quality of an inkjet printer. (I'll refer you again to 14:20 in CCMTM Part Two.)

3.5 picoliter dots though are still incredibly small; so small in fact that it's hard to even get a picture of them. Even scanning at 600 ppi, you can't get a close-up look at them, because as you zoom in, you find that they are significantly smaller than the pixels, and you get pixelation before you get a good actual look at the dots.

Now of course there are machines out there with significantly bigger dots. There are still some old Scitex 1500XL's chugging away out there with 80 pl dots. And you're right, 1440 of those dots would not fit in an inch. That's why their max resolution is in the 290x290 range.

I'd also note a couple more things:

I'm not a printhead engineer, but my understanding of the firing graph for every printhead in every resolution is that the center of each square is the firing point, and the goal is to hit the center of the square.

Also I'd note that very few printers have just one resolution. The Epson S80600 has -- for instance -- 360x720, 720x720, 1440x720, and 1440x1440. Also some have these have multi-dot or fixed dot mode. It is, or course, the same printheads and the same nozzles, and they're all the same size, it's a question of how they're fired. But for every resolution there's an individual graph.

I'd also quibble with the definition of a "splat." Again, I'm not an inkjet engineer, and again, the printers I work with have dots so small that I'm unable to get that close a look at them, but I'd bet folding money that they do not resemble the splotches in your illustration.

What I know for a fact after over 15 years of making a living profiling inkjet printers is that there is simply no other printing technology that is so consistently stable.

I can't see dots like that achieving that stability.


Mike
 

gordo

Well-known member
I'm not a printhead engineer, but my understanding of the firing graph for every printhead in every resolution is that the center of each square is the firing point, and the goal is to hit the center of the square.

Yes, the goal is to hit the center of the square - as defined by the addressability grid.

Also I'd note that very few printers have just one resolution. The Epson S80600 has -- for instance -- 360x720, 720x720, 1440x720, and 1440x1440. Also some have these have multi-dot or fixed dot mode. It is, or course, the same printheads and the same nozzles, and they're all the same size, it's a question of how they're fired. But for every resolution there's an individual graph.

You're mixing resolution with addressability. A printer can use an asymmetrical addressability grid. The smaller number (e.g. 360 in a 360x720 dpi device) is the actual addressability grid (360x360). If, for example, you slow the rate of substrate through the printer by a factor 1/2 then you will be placing twice as many droplets of in per inch in the direction of the substrate through the device. Bingo, and simple, you now have a 360x720 dpi device. It's not because of how the printheads are fired.

I'd also quibble with the definition of a "splat." Again, I'm not an inkjet engineer, and again, the printers I work with have dots so small that I'm unable to get that close a look at them, but I'd bet folding money that they do not resemble the splotches in your illustration.

I use the term splat because that's the tongue in cheek term that the engineers I worked with used.
The droplet of ink is somewhat akin to throwing a water filled balloon (ink droplet) to someone's face The substrate):

 

Correct Color

Well-known member
You're mixing resolution with addressability. A printer can use an asymmetrical addressability grid. The smaller number (e.g. 360 in a 360x720 dpi device) is the actual addressability grid (360x360). If, for example, you slow the rate of substrate through the printer by a factor 1/2 then you will be placing twice as many droplets of in per inch in the direction of the substrate through the device. Bingo, and simple, you now have a 360x720 dpi device. It's not because of how the printheads are fired.
Well, you got me curious.

I can't say for 100% certain that this is incorrect. I really tend to doubt it because if it was true, then it would follow that the doubled-dot resolutions would print at half speed. I've never seen that.

Also if I'm following you, that would mean that the larger number of dots must always be the vertical print direction. But in several drivers that is not the case.

Here are just a couple examples, a D-Gen Telios 100 Fabric printer and an Epson P8000. Both have instances of a doubled resolution on the horizontal print path.

DGen.png
P8000.png


I will say though that to me, that's somewhat irrelevant even if true. If that's how it does it, that's how it does it. The printed resolution is still the printed resolution.

Like I said though, I am curious about this now. I'm not a printhead engineer but I've got a buddy who is. I'm going to run all this by him and see what he says.


Mike
 
Last edited:

De-Inking

Avanti
Sustainable Printing Goes Far Beyond Using FSC Certified or Recycled Paper
This informative paper on deinking: demand, principles, problems and solutions also explains why printing technologies are not all equally compatible with paper recycling systems; and why just a small fraction of printed material in the paper can cause difficulties.
Link To White Paper

   
Top