picoliter droplet size question

[RoyZ]

While researching for a printer, the term picoleter droplet has been presented to me.

consider a 2-picoliter droplet printer vs a 4-picoliter droplet printer.

This sounds like the droplet size would be twice as large on the 4-pico size.
Am I correct, or what is the real difference?

Would its difference surface largely in a smaller print, say 4 x 6, rendering an image of less quality ?

All of this makes me belive that the Canon i 950 would print a higher quality image(4 x 6 size) than its more expensive counterpart the i9100.

Something doesn't add up. I must not understand all I know about this, the more I research the more confused I get. Two more hours of this and I won't be able to figure out how to get downstairs to my bedroom.
Roy

 

[Todd Felmly]

this, the more I research the more confused I get. Two more hours
of this and I won't be able to figure out how to get downstairs to
my bedroom.
Roy

LOL, tell me about it. I've been researching this for weeks. Just got home from Best Buy and had the Canon and Epson reps arguing with each other. I really like the Epson 925 especially with the card reader and roll paper accessory but the Canon is nice since it has 6 separate ink tanks and a replaceable ink head.

Decisions!

Todd

 

[Altaman]

Hi Again Roy:

Think of a picoliter as putting a bunch of pin pricks on two different pages to make an image, one page has 2 millimeter in diameter pin pricks and the other page using 4 millimeter. You are going to create the same image on each page, but with the two millimeter one you can put the dots more closely together, hence a more tighter image especially when you get closer.

Same thing happens with my HT projector...it is WVGA resolution with LCD panels. Now the bigger my screen size, the further I have to sit back to get away from "screen door" effect, which is basically the LCD panels creating a screen door image like you would see on your screen door to create the image on the screen. Now the further you are away or the smaller the screen squares the less chance you will see the screendoor effect. Same idea for picoliters, now the question is where is the happy medium...Well that is something your own personal taste has to decide...Me I can't see the need of going really any lower than two picoliter but who knows! I personally feel that at 4 picoliter there is a loss of sharpness in the image and you really don't get quite that photo-realistic look, but I have people look in aw at the prints that come off my I850 which is a 2 picoliter printer.

As a final note also to remember the printer can only print the quality of the image you give it!

Randy

While researching for a printer, the term picoleter droplet has
been presented to me.

consider a 2-picoliter droplet printer vs a 4-picoliter droplet
printer.

This sounds like the droplet size would be twice as large on the
4-pico size.
Am I correct, or what is the real difference?

Would its difference surface largely in a smaller print, say 4 x 6,
rendering an image of less quality ?

All of this makes me belive that the Canon i 950 would print a
higher quality image(4 x 6 size) than its more expensive
counterpart the i9100.

Something doesn't add up. I must not understand all I know about
this, the more I research the more confused I get. Two more hours
of this and I won't be able to figure out how to get downstairs to
my bedroom.
Roy

 

[Donald Cooper]

Inkjets create the illusion of continuous tone color by putting down lots of little drops (dpi doesn't matter for inkjets). So the smaller the drop, the better the image, all other things being equal. And drop size is measured in picoliters, though ultimately when the drop hits the media you're going to need to talk about the size of the resulting drop in microns.

But there are other things to consider -- the drops have to be round rather than oddly shaped and the printer has to position them correctly.

You're also getting into the area of diminishing returns with a 2 pico drop. Plus you can start getting some undesireable artifacts.

The size of the print doesn't matter. You'd just get a better looking 4X6 or 8X10. But the media is going to matter a great deal since that is what lets the drop keep its shape.

Generally speaking all the new photo printers do about the same quality job. You can quibble but its a waste of breath. Canons are faster but aren't any better than the Epsons (a lot faster really), and the output from the i950 doesn't look any better than the output from the i9100.

So you'll be happy with your choice.

While researching for a printer, the term picoleter droplet has
been presented to me.

consider a 2-picoliter droplet printer vs a 4-picoliter droplet
printer.

This sounds like the droplet size would be twice as large on the
4-pico size.
Am I correct, or what is the real difference?

Would its difference surface largely in a smaller print, say 4 x 6,
rendering an image of less quality ?

All of this makes me belive that the Canon i 950 would print a
higher quality image(4 x 6 size) than its more expensive
counterpart the i9100.

Something doesn't add up. I must not understand all I know about
this, the more I research the more confused I get. Two more hours
of this and I won't be able to figure out how to get downstairs to
my bedroom.
Roy

 

[SteveB]

My maths is a little rusty but I make it that a spherical droplet that flattens out on the paper will have a square root relationship between picolitres and diameter. In other words a 2pl droplet will appear to be 70.7% the size (dot diameter) of a 4pl droplet when it hits the paper assuming they both have the same depth of ink on the paper surface.

Donald Cooper wrote:
And drop size is measured in picoliters, though ultimately

when the drop hits the media you're going to need to talk about the
size of the resulting drop in microns.

 

[fotografer]

Steve!

It's not square root (factor 2 as in area), but cube root (factor 3)!!! So their actual radius or diameter ratio is even closer!

Assuming they are spherical, let's use a 5pl from printer H (guess which printer I am talking about?) and 2pl from printer C (guess?):

Let r be the radius of the 2pl droplet from printer C, let its diameter be d and volume be v;

Let R be the radius of the 5pl droplet from printer H, let its diameter be d and vulume be V.

So volume of printer C = 2pl = (4/3)(pi) r(raised to the power of 3)
And vol of printer H = 5pl = (4/3)(pi) R(raised to the power of 3)

So v:V ratio = (4/3)(pi) r (power 3) : (4/3)(pi) R (power 3)

Simplifying (cancelling constants in the equation)
therefore, v:V ratio = r (power 3) : R (power 3) = 2:5 [in picoliters]

Now r:R ratio = d ratio (their corresponding diameter, since both their diameters are twice of their radius)

and the answer is d/D = cube-root of (2/5)
this yield the d/D ratio as = 0.74

So a 2pl droplet has only 26% less diameter than a 5pl droplet. Or 2pl droplet as 74% diameter of the 5pl droplet. Not even less than half as we have assumed it to be, eh?

What about 4pl? Their calculations worked out to be 0.79. Or 2pl droplet is only 21% less in diameter to a 4pl, or 2pl droplet is 79% the diameter of the 4pl droplet!!!

Thus, on paper, the drop size of a 2pl is only 26% improvement to a 5pl, and 21% improvement to a 4pl.

On actual print test in HP labs, we visually see only 10% improvement. But that's another story...

Oh, for those interested in the 4pl droplet from printer E and 5pl droplet from printer H???...

E printer's 4pl has a droplet size less of H printer's 5pl at a tiny value of 7%. Or rather, it has an 'improvement' of a mere 7%.

And we are not even talking about how the driver LAY those droplets, which is the REAL secret to inkjet printing...

So much for picoliter talk! Just look at the printout quality and JUDGE FOR YOURSELF. Forget about numbers!

My maths is a little rusty but I make it that a spherical droplet
that flattens out on the paper will have a square root relationship
between picolitres and diameter. In other words a 2pl droplet will
appear to be 70.7% the size (dot diameter) of a 4pl droplet when it
hits the paper assuming they both have the same depth of ink on the
paper surface.

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[RoyZ]

Steve!

It's not square root (factor 2 as in area), but cube root (factor
3)!!! So their actual radius or diameter ratio is even closer!

Assuming they are spherical, let's use a 5pl from printer H (guess
which printer I am talking about?) and 2pl from printer C (guess?):

Let r be the radius of the 2pl droplet from printer C, let its
diameter be d and volume be v;

Let R be the radius of the 5pl droplet from printer H, let its
diameter be d and vulume be V.

So volume of printer C = 2pl = (4/3)(pi) r(raised to the power of 3)
And vol of printer H = 5pl = (4/3)(pi) R(raised to the power of 3)

So v:V ratio = (4/3)(pi) r (power 3) : (4/3)(pi) R (power 3)

Simplifying (cancelling constants in the equation)
therefore, v:V ratio = r (power 3) : R (power 3) = 2:5 [in picoliters]

Now r:R ratio = d ratio (their corresponding diameter, since both
their diameters are twice of their radius)

and the answer is d/D = cube-root of (2/5)
this yield the d/D ratio as = 0.74

So a 2pl droplet has only 26% less diameter than a 5pl droplet. Or
2pl droplet as 74% diameter of the 5pl droplet. Not even less
than half as we have assumed it to be, eh?

What about 4pl? Their calculations worked out to be 0.79. Or 2pl
droplet is only 21% less in diameter to a 4pl, or 2pl droplet is
79% the diameter of the 4pl droplet!!!

Thus, on paper, the drop size of a 2pl is only 26% improvement to a
5pl, and 21% improvement to a 4pl.

On actual print test in HP labs, we visually see only 10%
improvement. But that's another story...

Oh, for those interested in the 4pl droplet from printer E and 5pl
droplet from printer H???...

E printer's 4pl has a droplet size less of H printer's 5pl at a
tiny value of 7%. Or rather, it has an 'improvement' of a mere 7%.

And we are not even talking about how the driver LAY those
droplets, which is the REAL secret to inkjet printing...

So much for picoliter talk! Just look at the printout quality and
JUDGE FOR YOURSELF. Forget about numbers!

fotografer
...the great paper chase! (see
http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

Thank you SIR!
and thank everyone who posted.

fotografer you have hit the nail on the head.

It all resolves down to printout quality and what winds ones clock "JUDGE FOR YOURSELF" (why didn't I think of that). Then make the final selection based on how much print quality (if any) one has to sacrifice for less expensive consumables, time of printing, and usable vs unuseable functions .

Since I am not selling pictures, and B&W is farther down the road for me, and I would like a wide format printer, (but not at the cost of noticable image quality loss ) the field is Epson 2200 and Canon i9100, (I don't think HP makes a consumer large format inkjet)

Instead of sitting at my computer, I will have to (get off lazy posterior), and visit my local retailers to make my choice. This is what did it on my camera, so it should work on printer selection.

------Roy

 

[fotografer]

Royz,

Since I am not selling pictures, and B&W is farther down the road
for me, and I would like a wide format printer, (but not at the
cost of noticable image quality loss ) the field is Epson 2200 and
Canon i9100, (I don't think HP makes a consumer large format inkjet)

Er, actually, HP just introduced their consumer version to my ps7150, it's the deskjet 9600 series large format printer. Doing borderless up to 11x17! But it uses the same teeny tiny carts as my ps7150!

Instead of sitting at my computer, I will have to (get off lazy
posterior), and visit my local retailers to make my choice. This is
what did it on my camera, so it should work on printer selection.

Like I say, the secret actually lies in the WAY inks are laid. This dithering algorithm is a very guarded REAL secrets of manufacturers.

BTW, Vincent compared some pictures of flowers done with the s900 (same printhead design as the i9100 large format), and the six-color HP (like my ps7150) and the eight-color ps7960. See http://www.photo-i.co.uk review of the ps7960 (I think it's page 9 or something).

Here're some enlarged version of the 2pl Canon i950, Epson sp2100 and the HP ps7150. Bear in mind that over-saturation is one way sometimes manufacturers do to hide their grains.

These are 8x magnification, so in actual fact, they are VERY comparable under normal viewing conditions (to approximate 'normal' viewing conditions, click the 50% magnification at the top left hand corner to reduce the magnification to 4x):







In actual prints, I like the colors from Epson sp2100 the best, because it's the closest to the 'natural' look of the original file (look at the greens Epson is capable of).

Note that ps7150 is the same print engine as the large format HP deskjet 9600 series printer (selling at $399 retail, debut anytime this month!)...

Happy Shopping! ))

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[WillieB38717]

Hey, I see you've snuck in an i950 scan :-; That one is looking good, too.
--
WillieB

 

[RoyZ]

Wow!
fotografer, you have muddled my contrections!

The HP ps7960 is awesome! Vincents website is a -"must View" for anyone considering the purchase of a new printer.

Now I am totally disoriented (maybe two printers??)--Roy

 

[fotografer]

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[fotografer]

Royz,

Why stop at just reviewing the ps7960? Look at Vincent's equally excellent Epson sp2100 and Canon i950 review! The Epson sp2100 review is below the home page, while the Canon i950 is found in the 'review' sub-page'. The Canon s830D should be what a s900 or i9100 would look like, they have the same 4pl design, except, the s900/s9000/i9100 has wider printhead swath so they print faster.

Good Luck!

Wow!
fotografer, you have muddled my contrections!

The HP ps7960 is awesome! Vincents website is a -"must View" for
anyone considering the purchase of a new printer.

Now I am totally disoriented (maybe two printers??)--Roy

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[SteveB]

I think we're both wrong actually, as the flattened ink will have feathered edges so everything is very approximate but it makes you think eh?
Other things that make me think are:-

Does the ink partly dry on it's way to the paper?

How are the HP greens so faded on your tree leave prints? Never seen that on my 5550 prints.
Do printer driver updates address dithering patterns used?

Steve!

It's not square root (factor 2 as in area), but cube root (factor
3)!!! So their actual radius or diameter ratio is even closer!

 

[fotografer]

SteveB,

This comparison is 'not fair' in that Epson PGPP is used on all the printers. All standard set-up, except for my HP, I turned off the HP Digital Photography and I select (get this) Premium Glossy Film paper setting. I like this setting more than the usual Colorfast or HP PPPP setting on porous papers, because HP Premium Glossy Film is the only HP porous papers at the moment, and ink volume laid are the least. If I use Colorfast or HP PPPP the results would be too saturated (artificial) and yellowish.

The profiled results are much better than the ones you see here. Of course, if I print on HP PPPP media, the greens would have been richer (but not more saturated). The scans are processed together in PS. When we view each prints invidivually the HP greens are not so light, but the Canon prints are still the most saturated, not the way I like, but it should satisfy people who wants loads of punch (the black eye variety) in their color prints.

To each its BEST.

Does the ink partly dry on it's way to the paper?
How are the HP greens so faded on your tree leave prints? Never
seen that on my 5550 prints.
Do printer driver updates address dithering patterns used?

Steve!

It's not square root (factor 2 as in area), but cube root (factor
3)!!! So their actual radius or diameter ratio is even closer!

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[filip]

Hi Steve,

don't forget that these are magnified prints so in reality the difference becomes a matter of subtelty in the rendering of the different printers.

subteltly and clean output are the sp2200 trademarks. the hp prints have excellent sharpness and are also clean in real life. the i950 does not as well in color fidelity and rendering of the shadows though it is still very capable. i think one of the points that canon has still so far not addressed is the dithering algorithm which from the different samples that i have seen is still on the weak side and have not evolved much over the years. they make up for this somehow by using smaller dots (2pl).

philip

Does the ink partly dry on it's way to the paper?
How are the HP greens so faded on your tree leave prints? Never
seen that on my 5550 prints.
Do printer driver updates address dithering patterns used?

Steve!

It's not square root (factor 2 as in area), but cube root (factor
3)!!! So their actual radius or diameter ratio is even closer!

 

[fotografer]

Philip,

There is someone at Vincent's forum http://www.photo-i.co.uk , that wants to know your AMP setting for neutral greyscale. Look under the thread that talks about "B&w: 7960 v. 2100 with ip5.6 or ijc/opm", at page 9, I think...

Hi Steve,

don't forget that these are magnified prints so in reality the
difference becomes a matter of subtelty in the rendering of the
different printers.

subteltly and clean output are the sp2200 trademarks. the hp prints
have excellent sharpness and are also clean in real life. the i950
does not as well in color fidelity and rendering of the shadows
though it is still very capable. i think one of the points that
canon has still so far not addressed is the dithering algorithm
which from the different samples that i have seen is still on the
weak side and have not evolved much over the years. they make up
for this somehow by using smaller dots (2pl).

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[fotografer]

Wrong thread title, you should look into http://www.photo-i.co.uk/cgi-bin/ikonboard/ikonboard.pl?s=3f59d89264deffff ;act=ST;f=37;t=4;st=140 as the direct link...

Philip,

There is someone at Vincent's forum http://www.photo-i.co.uk , that wants
to know your AMP setting for neutral greyscale. Look under the
thread that talks about "B&w: 7960 v. 2100 with ip5.6 or ijc/opm",
at page 9, I think...

--
fotografer

...the great paper chase! (see http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )

 

[filip]

thanks fotografer. i replied to the post
philip

http://www.photo-i.co.uk/cgi-bin/ikonboard/ikonboard.pl?s=3f59d89264deffff ;act=ST;f=37;t=4;st=140 as the direct link...

Philip,

There is someone at Vincent's forum http://www.photo-i.co.uk , that wants
to know your AMP setting for neutral greyscale. Look under the
thread that talks about "B&w: 7960 v. 2100 with ip5.6 or ijc/opm",
at page 9, I think...

--
fotografer
...the great paper chase! (see
http://forums.dpreview.com/forums/read.asp?forum=1003&message=5683956 )