Variable Droplet Size?

Started Aug 21, 2002 | Discussions
WP Senior Member • Posts: 2,114
Variable Droplet Size?

For the life of me I don't understand this claim. Epson, HP, and Lexmark claim this capability while Canon goes complete opposite and maintains they have an accurate consistent droplet size at all times. I have reviewed inkjet technology in depth and do not understand the claim to support variable droplet size. Seems to me more like the inability to CONTROL droplet size vice have it variable. Looking at prints under magnification, the variable size in droplets is random across the page and even within a small area of the same color.

The chamber that the ink in drawn into before firing is a standard size. Then either heat or an electrical charge fires the ink onto the page. A standard size chamber INDICATES that the only way you can get a variable droplet size (for example this drop is 3 picoliters, this drop is 10 picoliters) is if there is very fine control to the amount of ink released into the chamber. With the number of drops being placed on a page, this seems to me to be a little far-fetched. To control each drop in this manner, when we are talking about liquid at this size, HMMM.

Can someone explain how this works to me (I know a lot about engineering and mechanics)?

I will be honest. While variable droplet size APPEARS to be a good idea, it seems to me to be more a lack of control of droplet size hyped into a marketing idea.
--
Have a great day
http://www.pbase.com/wp12001

fotografer Veteran Member • Posts: 5,159
Re: Variable Droplet Size?

WP,

While I don't mean to give the mechanical and engineering aspects of this so-called technology, I shall venture to explain this with my knowledge of HP laser-jet printers.

HP debut their laserjets eons ago, once with 300dpi with RET (Resolution Enhancement Technology), and it explained then that at edges of curves, smaller pixel is generated, while the larger, solid area in text fonts are using larger pixel for faster printing.

The essence here is speed. Of course, with the smallest dots HP could produce then with their laserjet, they could have easily tout their laserjets to be 600dpi capable - but they don't. At that time, they could not have the resource (computing, that is) to drive the printer at reasonable speed with 600 dpi.

Fast foward to the not so distant past, the Photoret II and III. The smallest droplet size (or rather the volume, thus measured in picolitre) are roughly 12 pl and 5 pl respectively, such are employed on the edge for 'smooth' rendering in photos (that's what the prefix Photo to the RET).

The nozzle sizes are fixed, no doubt, it's the energy each nozzle size is given and the speed in which the inks are forced through them that gives them larger size required - and yes, it helps in the speed of printing in this case as well.

Thermal technology of HP and Canon printers are similar, and it would be impossible for Canon to give only or exclusively 2 pl size (if indeed this is measurable, though I don't doubt Canon's claims here) for their printing, even with large printheads in their s900x series printers. They must regulate the temperature and energy charged at each nozzle for speed and delivery of the inks concerned.

In other words, variable dot printing is a norm since the birth of inkjet technology (or at least the photo-enabled inkjets), not something to be touted at all - and even Canon cannot possibly only use 2 pl size for their printjobs. If not, they will wear out their printhead (and not achieve the speed they have now) in no time.

-- hide signature --

Fotografer

OP WP Senior Member • Posts: 2,114
Re: Variable Droplet Size?

Thanks...Seems to me when we are talking sizes of about 25 microns? or less, there seems to be less control of the size of the drops. The measurable aspect of this seems to be the key. The difference between 3 and 5 picoliters may not be measurable to the eye, but at 10-12 picoliters you can definitely see the dots. That's why when I see a claim "Variable Droplet Size from 3 - 10 Picoliter" I suspect the validity of the claim to have that much control over the drops that it can actually be accomplished. I would much more believe a claim of 3-5 picoliter consistent droplet size than a claim that the computer/printer can process each droplet individually to determine the best size to use with that great a range.

-- hide signature --
fotografer Veteran Member • Posts: 5,159
Re: Variable Droplet Size?

Thanks...Seems to me when we are talking sizes of about 25 microns?
or less, there seems to be less control of the size of the drops.
The measurable aspect of this seems to be the key. The difference
between 3 and 5 picoliters may not be measurable to the eye, but at
10-12 picoliters you can definitely see the dots. That's why when
I see a claim "Variable Droplet Size from 3 - 10 Picoliter" I
suspect the validity of the claim to have that much control over
the drops that it can actually be accomplished. I would much more
believe a claim of 3-5 picoliter consistent droplet size than a
claim that the computer/printer can process each droplet
individually to determine the best size to use with that great a
range.

I think 3 - 10 pl seems only a little ambitious, but it's not impossible, but will considerably slow down the RIP.

May I ask which manufacturer gave this range? I hope it's not the hopeless HP sales reps again.

-- hide signature --

Fotografer

Donald Cooper Veteran Member • Posts: 4,927
Re: Variable Droplet Size?

You can of course choose not to believe the claims, but note that if they couldn't control the droplet size they couldn't control placement, and if they couldn't control placement then they couldn't create the dither tiles they need to approximate continuous tone.

You can actually see this for yourself if you look at a blowup with extreme magnification on media which minimizes bleed. You will see the drops are perfectly round and all exactly the same size. A web site called dotty-spotty has scans of this. You might check for the cite, I don't have it off-hand.

As mentioned, variable drop size is used for speed. For instance, the Epson 2200 does not use variable drops at its highest quality setting.

I'm not sure of the size in microns. My rule of thumb is that a picoliter is about 5 microns, so 2 picoliters is about 10 microns and 5 picoliters is 25-28 microns. I'd be happy to be corrected on this though.

WP wrote:

Thanks...Seems to me when we are talking sizes of about 25 microns?
or less, there seems to be less control of the size of the drops.
The measurable aspect of this seems to be the key. The difference
between 3 and 5 picoliters may not be measurable to the eye, but at
10-12 picoliters you can definitely see the dots. That's why when
I see a claim "Variable Droplet Size from 3 - 10 Picoliter" I
suspect the validity of the claim to have that much control over
the drops that it can actually be accomplished. I would much more
believe a claim of 3-5 picoliter consistent droplet size than a
claim that the computer/printer can process each droplet
individually to determine the best size to use with that great a
range.

OP WP Senior Member • Posts: 2,114
Re: Variable Droplet Size?

http://www.silverace.com/dottyspotty/

Found it! Thanks. That helps much more! A little out of date however. But gives some good info on the technology. But even at that magnification I see no variable droplet size.

Donald Cooper wrote:
You can of course choose not to believe the claims, but note that
if they couldn't control the droplet size they couldn't control
placement, and if they couldn't control placement then they
couldn't create the dither tiles they need to approximate
continuous tone.

You can actually see this for yourself if you look at a blowup with
extreme magnification on media which minimizes bleed. You will see
the drops are perfectly round and all exactly the same size. A web
site called dotty-spotty has scans of this. You might check for the
cite, I don't have it off-hand.

As mentioned, variable drop size is used for speed. For instance,
the Epson 2200 does not use variable drops at its highest quality
setting.

I'm not sure of the size in microns. My rule of thumb is that a
picoliter is about 5 microns, so 2 picoliters is about 10 microns
and 5 picoliters is 25-28 microns. I'd be happy to be corrected on
this though.

WP wrote:

Thanks...Seems to me when we are talking sizes of about 25 microns?
or less, there seems to be less control of the size of the drops.
The measurable aspect of this seems to be the key. The difference
between 3 and 5 picoliters may not be measurable to the eye, but at
10-12 picoliters you can definitely see the dots. That's why when
I see a claim "Variable Droplet Size from 3 - 10 Picoliter" I
suspect the validity of the claim to have that much control over
the drops that it can actually be accomplished. I would much more
believe a claim of 3-5 picoliter consistent droplet size than a
claim that the computer/printer can process each droplet
individually to determine the best size to use with that great a
range.

-- hide signature --
OP WP Senior Member • Posts: 2,114
Re: Variable Droplet Size?

3-10 picoliters is right from the Lexmark website for their 4800x1200 DPI printers. HP uses the "Optimized" 4800x1200 DPI thing and when I asked about it to the rep, he said that means it varies the droplet size. I saw one brochure that said 4 picoliters on the new HP's and others that dont say anything about droplet size other than "Optimized".

-- hide signature --

Have a great day
http://www.pbase.com/wp12001

fotografer wrote:

Thanks...Seems to me when we are talking sizes of about 25 microns?
or less, there seems to be less control of the size of the drops.
The measurable aspect of this seems to be the key. The difference
between 3 and 5 picoliters may not be measurable to the eye, but at
10-12 picoliters you can definitely see the dots. That's why when
I see a claim "Variable Droplet Size from 3 - 10 Picoliter" I
suspect the validity of the claim to have that much control over
the drops that it can actually be accomplished. I would much more
believe a claim of 3-5 picoliter consistent droplet size than a
claim that the computer/printer can process each droplet
individually to determine the best size to use with that great a
range.

I think 3 - 10 pl seems only a little ambitious, but it's not
impossible, but will considerably slow down the RIP.

May I ask which manufacturer gave this range? I hope it's not the
hopeless HP sales reps again.

fotografer Veteran Member • Posts: 5,159
Re: Variable Droplet Size?

Hi WP,

3-10 picoliters is right from the Lexmark website for their
4800x1200 DPI printers. HP uses the "Optimized" 4800x1200 DPI
thing and when I asked about it to the rep, he said that means it
varies the droplet size. I saw one brochure that said 4 picoliters
on the new HP's and others that dont say anything about droplet
size other than "Optimized".

Wow. I never know Lexmark is that ambitious.

When you see the optimized 4800x1200 dpi tag, look closer to what carts they are using. If they only use the no.78 carts, or only no.57 carts, then whatever their 'optimized' 4800 dpi you are going to get visible dots, the no.78 carts more visible than the no.57 alone (i.e. without photocart no. 58).

A little hint: Both Canon and HP use thermal technology for their inkjets, and they both use 'variable' dot dithering of setting at either 'high' and 'low'.

Perhaps Lexmark, when they quoted their wide-range drop size between 3 to 10 pl, they mean by what HP and Canon has been doing all along: on edges, 'low' signal (small droplet), on solid color areas, 'high' signal. This TWO setting is variable on different paper settings.

Thus, for HP, on plain paper, using no. 78 inks, the 'low' is probably at around 7 to 10 pl, while the 'high' is perhaps even as high as 16 pl.

But when it comes to photopaper, their 'low' would be 5pl, while their high, maybe around 10 pl (I am not sure exactly what number, but you get my point?).

However, due to other real-life variables not controllable, there will be the 'in-between' sizes, especially obvious with high speed, plain paper option. So the range would more or less cover between 7 pl to 16 pl for the no. 78 pens when printing on plain paper.

Perhaps that's what Lexmark means.

Don't trust the sales reps explanation, whether it be from HP or Epson or whatnot...

-- hide signature --

Fotografer

Donald Cooper Veteran Member • Posts: 4,927
Re: Variable Droplet Size?

If they can control the size you shouldn't see any difference in size at this magnification. With variable drop, you use the large drops when you get a large swatch of color -- like a sky. You'd use the smaller drops in areas of relatively rapid color transition. I wouldn't think at this magnification you could get enough area -- you'd be looking at one area or the other. The only way to see it would be one magnificaiton from one area compared to a magnification from the other. Then the dots would be of different sizes.

My point was so that you can see exactly how precise the dots are.

WP wrote:

http://www.silverace.com/dottyspotty/

Found it! Thanks. That helps much more! A little out of date
however. But gives some good info on the technology. But even at
that magnification I see no variable droplet size.

Joe Barnhart Veteran Member • Posts: 3,393
Canon claims 4pl, not 2pl

I just did a search and could find nothing in which Canon claimed 2pl drop sizes. What they DO claim is that they have 4pl drop size, and the size is constant -- not variable. They say they can fire 74 million droplets per second with the S900/9000 print head.

Search Google using "advanced microfine droplet" and look for Canon press releases and marketing materials.

californiaPhoto Regular Member • Posts: 124
Re: Variable Droplet Size?

HI,

So in conclusion, do the variable droplet produce potentially higher quality with less visiblile grain that the non-variable as in the epson 2200?
Basically, 7600 vs 2200 printers.

thanks.

fotografer wrote:

WP,

While I don't mean to give the mechanical and engineering aspects
of this so-called technology, I shall venture to explain this with
my knowledge of HP laser-jet printers.

HP debut their laserjets eons ago, once with 300dpi with RET
(Resolution Enhancement Technology), and it explained then that at
edges of curves, smaller pixel is generated, while the larger,
solid area in text fonts are using larger pixel for faster printing.

The essence here is speed. Of course, with the smallest dots HP
could produce then with their laserjet, they could have easily tout
their laserjets to be 600dpi capable - but they don't. At that
time, they could not have the resource (computing, that is) to
drive the printer at reasonable speed with 600 dpi.

Fast foward to the not so distant past, the Photoret II and III.
The smallest droplet size (or rather the volume, thus measured in
picolitre) are roughly 12 pl and 5 pl respectively, such are
employed on the edge for 'smooth' rendering in photos (that's what
the prefix Photo to the RET).

The nozzle sizes are fixed, no doubt, it's the energy each nozzle
size is given and the speed in which the inks are forced through
them that gives them larger size required - and yes, it helps in
the speed of printing in this case as well.

Thermal technology of HP and Canon printers are similar, and it
would be impossible for Canon to give only or exclusively 2 pl
size (if indeed this is measurable, though I don't doubt Canon's
claims here) for their printing, even with large printheads in
their s900x series printers. They must regulate the temperature and
energy charged at each nozzle for speed and delivery of the inks
concerned.

In other words, variable dot printing is a norm since the birth
of inkjet technology (or at least the photo-enabled inkjets), not
something to be touted at all - and even Canon cannot possibly only
use 2 pl size for their printjobs. If not, they will wear out their
printhead (and not achieve the speed they have now) in no time.

fotografer Veteran Member • Posts: 5,159
Re: Variable Droplet Size?

Hi,

So in conclusion, do the variable droplet produce potentially
higher quality with less visiblile grain that the non-variable as
in the epson 2200?
Basically, 7600 vs 2200 printers.

There cannot be a non-variable printer, I doubt even Epson 2200 uses a range of dot grains to prints, just a narrower range, that's all.

As I said earlier, variable dot grain gives better speed, and is theoretically more efficient in producing 'euqivalent' quality output with small range dot grains prints, such as 2200.

I doubt there is a detectable difference between 7600 and 2200, even though the former uses variable and the latter supposedly non-variable print methodology.

-- hide signature --

Fotografer

Donald Cooper Veteran Member • Posts: 4,927
Where is this misinformation coming from ?

Guys, the 2200 is variable dot. I have no idea where you're getting your info from, but the 2200 uses variable dot AT ALL SETTINGS other than 2880.

Just look at this link from Epson UK. Go to the section on Advanced Epson technology and look at the section on Variable-Sized Drolet. The hard part was finding the info that it didn't use variable dot at 2880.

http://www.epson.co.uk/product/printers/photo/styphoto2100/index.htm

And in answer to this question, you have it backward -- same sized drops would potentially produce higher quality assuming they are small. Smaller is better. Larger is faster.

californiaPhoto wrote:
HI,
So in conclusion, do the variable droplet produce potentially
higher quality with less visiblile grain that the non-variable as
in the epson 2200?
Basically, 7600 vs 2200 printers.

thanks.

fotografer wrote:

WP,

While I don't mean to give the mechanical and engineering aspects
of this so-called technology, I shall venture to explain this with
my knowledge of HP laser-jet printers.

HP debut their laserjets eons ago, once with 300dpi with RET
(Resolution Enhancement Technology), and it explained then that at
edges of curves, smaller pixel is generated, while the larger,
solid area in text fonts are using larger pixel for faster printing.

The essence here is speed. Of course, with the smallest dots HP
could produce then with their laserjet, they could have easily tout
their laserjets to be 600dpi capable - but they don't. At that
time, they could not have the resource (computing, that is) to
drive the printer at reasonable speed with 600 dpi.

Fast foward to the not so distant past, the Photoret II and III.
The smallest droplet size (or rather the volume, thus measured in
picolitre) are roughly 12 pl and 5 pl respectively, such are
employed on the edge for 'smooth' rendering in photos (that's what
the prefix Photo to the RET).

The nozzle sizes are fixed, no doubt, it's the energy each nozzle
size is given and the speed in which the inks are forced through
them that gives them larger size required - and yes, it helps in
the speed of printing in this case as well.

Thermal technology of HP and Canon printers are similar, and it
would be impossible for Canon to give only or exclusively 2 pl
size (if indeed this is measurable, though I don't doubt Canon's
claims here) for their printing, even with large printheads in
their s900x series printers. They must regulate the temperature and
energy charged at each nozzle for speed and delivery of the inks
concerned.

In other words, variable dot printing is a norm since the birth
of inkjet technology (or at least the photo-enabled inkjets), not
something to be touted at all - and even Canon cannot possibly only
use 2 pl size for their printjobs. If not, they will wear out their
printhead (and not achieve the speed they have now) in no time.

californiaPhoto Regular Member • Posts: 124
Re: Where is this misinformation coming from ?

Good points WP, it seems the technology would decide where to put larger picos vs smaller which seems 'random' to some degree. (obviously edges would produce smaller dot's, hopefully). Donald, i called Epson staff and asked them the difference btw the 2200 and 7600 and they said only the absence of variable droplets on the 2200. That is why I came here and asked if quality would be descernable on these two different mechanisms. Personally, my belief is that variable droplets would only constitute faster prints speeds like fotographer said. I would also suspect that smaller picos throughout an image entirely would compromise the 'randomness' of larger droples and produce more continuous print. This all of which is not even discernable to the naked eye really. I purchased the 2200 at CDW for $621 and will arrive by next week. Epson has only served good to me and not once had a printer clog or stop working. I bought from them years ago and continue due to their great support. my 5 cents.

http://www.borellimedia.com/

Donald Cooper wrote:
Guys, the 2200 is variable dot. I have no idea where you're getting
your info from, but the 2200 uses variable dot AT ALL SETTINGS
other than 2880.

Just look at this link from Epson UK. Go to the section on Advanced
Epson technology and look at the section on Variable-Sized Drolet.
The hard part was finding the info that it didn't use variable dot
at 2880.

http://www.epson.co.uk/product/printers/photo/styphoto2100/index.htm

And in answer to this question, you have it backward -- same sized
drops would potentially produce higher quality assuming they are
small. Smaller is better. Larger is faster.

californiaPhoto wrote:

HI,
So in conclusion, do the variable droplet produce potentially
higher quality with less visiblile grain that the non-variable as
in the epson 2200?
Basically, 7600 vs 2200 printers.

thanks.

californiaPhoto Regular Member • Posts: 124
Re: Where is this misinformation coming from ?

I read the headline of that link and it says 2100 epson. I discarded the article for legitiacy. I also asked epson on that model number and they don't know what that is. I told them it's referred all over the web as 2100 as well as 2200. Can anyone tell me why? thanks.

Donald Cooper wrote:

Guys, the 2200 is variable dot. I have no idea where you're getting
your info from, but the 2200 uses variable dot AT ALL SETTINGS
other than 2880.

Just look at this link from Epson UK. Go to the section on Advanced
Epson technology and look at the section on Variable-Sized Drolet.
The hard part was finding the info that it didn't use variable dot
at 2880.

Donald Cooper Veteran Member • Posts: 4,927
Re: Where is this misinformation coming from ?

This is turning out to be a pretty interesting thread. Hmmmm...... so the misinformation is coming from Epson tech support. Great. The difference between the 2200 and the 7600? How about the fact that the 7600 can produce really wide (24") prints and uses nice-sized cartridges (110 ml). Now those are differences! Of course they were just referring to the printing mechanism, and yes, the heads in the 9600, 7600 and 2200 are exactly the same.

While it is true that the Epson 2200/2100 (they're basically the same printer, different model numbers in Europe vs. NA) doesn't use variable dots, that's only true at the highest resolution. At resolutions below 2880 it uses variable dots. I can't remember off hand where I've read that, but the materials were from Epson. I was very interested in this because I thought Epson had changed the dither at 2880 to a stochastic, which you can't do with variable dot, and I was curious when they kept telling me the 2200/2100 used variable dot since it didn't seem to correspond to what I was looking at. Again, note that this is not inconsistent with what the tech person told you -- it is true that the 7600 uses variable dots and the 2200 does not -- it's just that the statement isn't true for all resolutions. But it is a difference .......

BTW, if you really want to use variable dots, and you're not sure you are with the 2200, you can just use the driver for the 7600. Given the time and expense needed to do a good driver, all these drivers will drive any of the printers. So if history is a guide, you could use the 7600 driver with the 2200 and be just fine, just like you can use a 1290 driver with the 880. I've assumed the reason that they stayed with the variable dot for the 7600 and 9600 at all resolutions is that given the area these guys need to cover Epson couldn't afford the speed hit they'd take by abandoning it. But who knows?

And yes it's a speed issue. Using variable dot increases speed, not quallity. As a general matter, you'll get the best quality with very small and very consistently sized dots and a stochastic dither. (But I haven't been able to confirm the dither).

californiaPhoto wrote:

Good points WP, it seems the technology would decide where to put
larger picos vs smaller which seems 'random' to some degree.
(obviously edges would produce smaller dot's, hopefully). Donald, i
called Epson staff and asked them the difference btw the 2200 and
7600 and they said only the absence of variable droplets on the
2200. That is why I came here and asked if quality would be
descernable on these two different mechanisms. Personally, my
belief is that variable droplets would only constitute faster
prints speeds like fotographer said. I would also suspect that
smaller picos throughout an image entirely would compromise the
'randomness' of larger droples and produce more continuous print.
This all of which is not even discernable to the naked eye really.
I purchased the 2200 at CDW for $621 and will arrive by next week.
Epson has only served good to me and not once had a printer clog or
stop working. I bought from them years ago and continue due to
their great support. my 5 cents.

http://www.borellimedia.com/

AnaDigi Contributing Member • Posts: 658
Re: Where is this misinformation coming from ?

californiaPhoto wrote:
I read the headline of that link and it says 2100 epson. I
discarded the article for legitiacy. I also asked epson on that
model number and they don't know what that is. I told them it's
referred all over the web as 2100 as well as 2200. Can anyone tell
me why? thanks.

Donald Cooper wrote:

Guys, the 2200 is variable dot. I have no idea where you're getting
your info from, but the 2200 uses variable dot AT ALL SETTINGS
other than 2880.

Just look at this link from Epson UK. Go to the section on Advanced
Epson technology and look at the section on Variable-Sized Drolet.
The hard part was finding the info that it didn't use variable dot
at 2880.

Keyboard shortcuts:
FForum MMy threads