Denistometer V/S Spectrophotometer

[nilay]

Hi,

I am not sure that this point was discussed before or not. Which device should be better to the newspaper pressroom: densitomter or spectrophotometer? At present we are scanning gray bar (not black) on the color pages and try to keep the C, M and Y density values within given specification on the presses. We use status T (absolute) on densitometer. It happens that due to different shades of paper: same density may have different L, a and b values. Also spectrophotometer may be useful to check the Delta E between proofs and printed copies and also 'b' value of the paper. It would be great to have some information on which device we should apply and why?

Thanks.

 

[gordo]

[snip]Which device should be better to the newspaper pressroom: densitomter or spectrophotometer? At present we are scanning gray bar (not black) on the color pages and try to keep the C, M and Y density values within given specification on the presses. We use status T (absolute) on densitometer. It happens that due to different shades of paper: same density may have different L, a and b values. Also spectrophotometer may be useful to check the Delta E between proofs and printed copies and also 'b' value of the paper. It would be great to have some information on which device we should apply and why?

Many, and possibly most, newspaper printing does not make use of a proof. Presswork is typically done by eye and densitometer.
This is in part because many (if not most) newspapers use ink optimization reseparation workflows which reduce the impact of solid ink density fluctuations. So, as long as the press operator gets the densities somewhere within tolerance sellable color will fall into place automagically.

For the pressroom I would say that a densitometer is the right device - after all the press operator cannot change the hue of the inks on press. SID is the main real control the press operator has. The inks are stored in huge vats which tends to moderate batch to batch ink hue variation. The paper starts shifting hue as soon as it's exposed to air and light so monitoring the b* value of the paper has limited value (for the press operator). I don't believe that coldset paper contains any OBAs so that point of variation is not a concern.

The grey balance bar, IMHO, is virtually useless in newspaper work to monitor variation in color in large part because of the impact of ink optimization reseparation (max GCR) on the live image area vs no ink optimization in the grey patch in the color bar. So, image area and color bar are very disconnected from a color point of view.

If you're not using an ink optimization reseparation workflow then you'll see greater variation as SIDs fluctuate - however SIDs are still what the press operator will adjust.

It may be worth having a spectrophotometer as part of the QC kit used by management, but IMHO, not of much use in the press room.

best, gordo

 

[Cornishpastythighs]

Just a word of advice, if you get a Densitometer for your on press measuring, get one that is easily used by the press crews and not one that is more at home in a QA Lab. I much prefer the older Xrite 530 series Densitometers for onpress work, try and stay away from the large new Xrite 939's they have no business on a press in my opinion. If you will be monitoring density for trends and process improvements get software that is press crew friendly or it wont get used.

 

[Schnitzel]

Also, a densitometer is a lot more intuitive for the press operator than a spectro. Changing the density will change a single reading up or down - simple enough. On the other hand, delta-E values won't tell you what action you need to take to correct the colors, and comparing target and measured L*a*b* coordinates can be slow and confusing.

 

[chromix]

Why not get both?

Most hand-held "densitometers" these days are spectrodensitometers. They take full spectral readings which can be converted into both density values and colorimetric values.

Your description of your needs in the press room are fairly typical these days. Density is still the reliable, known method for much of the process control required but sometimes you still want to have delta-E for a variety of reasons...

If you HAVE to pick one, I'd say densitometer. Do you HAVE to?

regards,

Steve

 

[gig0]

Many, and possibly most, newspaper printing does not make use of a proof. Presswork is typically done by eye and densitometer.
This is in part because many (if not most) newspapers use ink optimization reseparation workflows which reduce the impact of solid ink density fluctuations. So, as long as the press operator gets the densities somewhere within tolerance sellable color will fall into place automagically.

For the pressroom I would say that a densitometer is the right device - after all the press operator cannot change the hue of the inks on press. SID is the main real control the press operator has. The inks are stored in huge vats which tends to moderate batch to batch ink hue variation. The paper starts shifting hue as soon as it's exposed to air and light so monitoring the b* value of the paper has limited value (for the press operator). I don't believe that coldset paper contains any OBAs so that point of variation is not a concern.

The grey balance bar, IMHO, is virtually useless in newspaper work to monitor variation in color in large part because of the impact of ink optimization reseparation (max GCR) on the live image area vs no ink optimization in the grey patch in the color bar. So, image area and color bar are very disconnected from a color point of view.

If you're not using an ink optimization reseparation workflow then you'll see greater variation as SIDs fluctuate - however SIDs are still what the press operator will adjust.

It may be worth having a spectrophotometer as part of the QC kit used by management, but IMHO, not of much use in the press room.

best, gordo

gordo strikes again with surgical precision! Thanks, Gordo.

 

[Prepper]

The grey balance bar, IMHO, is virtually useless in newspaper work to monitor variation in color in large part because of the impact of ink optimization reseparation (max GCR) on the live image area vs no ink optimization in the grey patch in the color bar. So, image area and color bar are very disconnected from a color point of view.

Gordo, don't mean to sidetrack but I was just thinking about this recently, we're separating our images with MaxGCR but color bar doesn't get separated, just straight 50-40-40 gray. If we print to gray balance in color bar, are our images gray balanced? We're using G7 method, when have good curves for the paper and ink we're using, we get good proof to press matches. Curves are getting applied to both images and color bar.

I understand they are disconnected color wise then, but what about just process control, are we doing the right thing? Would there be any other way to do it? I guess you could take the color bar and resep it the same way or plug MaxGCR sepped numbers into the color bar instead of the 50-40-40? Would there be a benefit to that? Well, never mind, that wouldn't work, that would be 15-12-12-41.

What do the G7 experts think of that? What is the best or recommended approach in this case?

Thanks

 

[rich apollo]

First, I will disagree with Gordo, at my peril, on a couple of points.

1) I think a spectro is much more versatile. Combined with software you can do so much more with a spectro.

2) I think the reseparation idea that Gordo raises is a non-issue. The point of the colorbar is to monitor press performance. Regardless of what you've done to the imagery the gray bar is still going to indicate fluctuations in press performance - and in the case he cites, the gray bar will show variation long before the imagery.

2b) Colorbars MUST have the same plate curves applied. On that, I think, Gordo and I would agree.

To Prepper, NO! Do not reseparate your colorbars!!!! Curve yes - reseparate NO!

About the only scenario that I see a densitometer being a good fit is on gray patches. There you can get all the information you need from a single reading. On a regular colorbar, densitometers are simply too inconvenient and end up being unused.

By the way, nilay, if you subtract the paper from your measurements, you should be able to used the same values regardless of changes in the paper.

 

[gordo]

First, I will disagree with Gordo, at my peril, on a couple of points.

1) I think a spectro is much more versatile. Combined with software you can do so much more with a spectro.

2) I think the reseparation idea that Gordo raises is a non-issue. The point of the colorbar is to monitor press performance. Regardless of what you've done to the imagery the gray bar is still going to indicate fluctuations in press performance - and in the case he cites, the gray bar will show variation long before the imagery.

2b) Colorbars MUST have the same plate curves applied. On that, I think, Gordo and I would agree.

To Prepper, NO! Do not reseparate your colorbars!!!! Curve yes - reseparate NO!

About the only scenario that I see a densitometer being a good fit is on gray patches. There you can get all the information you need from a single reading. On a regular colorbar, densitometers are simply too inconvenient and end up being unused.

By the way, nilay, if you subtract the paper from your measurements, you should be able to used the same values regardless of changes in the paper.

Beware the wrath of Gordo! LOL ;-)

1) As I wrote a spectro would be good to have - but for the pressroom (i.e. what the press operators would use) a densitometer is more useful.

2b) Yup, agreed.

Yup do not reseparate your colorbars. Curve yes - reseparate NO.

RE:

About the only scenario that I see a densitometer being a good fit is on gray patches. There you can get all the information you need from a single reading. On a regular colorbar, densitometers are simply too inconvenient and end up being unused.

In newspaper work, when a 3/c grey bar is used at the foot of the page a densitometer is used to measure the 3/c bar to get SID information which then informs the press operator about any needed ink key adjustments. On a regular color bar (not typically used in newspaper work) the press operator uses the densitometer to measure the solid process patches and generally ignores the grey balance (and other) patches. The solid process patches inform the press operator about any needed ink key adjustments to bring SIDs into spec.

2) I think the reseparation idea that Gordo raises is a non-issue. The point of the colorbar is to monitor press performance. Regardless of what you've done to the imagery the gray bar is still going to indicate fluctuations in press performance - and in the case he cites, the gray bar will show variation long before the imagery.

That's the crux of the matter: "the gray bar will show variation long before the imagery". Effectively, the grey bar provides non-actionable information. It is therefore irrelevant. "The point of the colorbar is to monitor press performance." No (and here I will be anal). The point of the color bar is to provide relevant information to the press operator. The press operator monitors and interprets press performance using the information provided by the color bar (and other image elements). Press operators know that presses are inconsistent but stable when it comes to laying down ink. So, the minor fluctuations in grey balance are irrelevant if the SIDs are within tolerance. Also those fluctuations in grey balance will not show up in the live image area - especially true if ink optimization (reseparation) has been applied to the imagery. So, those fluctuations in grey balance will not be adjusted for otherwise the operator will end up continuously adjusting SIDs to chase grey balance.
Even the so-called father of grey balance - Felix Brunner of System Brunner - understood this because he advocated densitometers over colorimeters (and spectros) to monitor grey balance on press. N.B. this was in the days of UCR rather than today's GCR and Max GCR separations when grey balance targets in color bars had a closer relationship to the live image area.

best, (wrath over) gordo

 

[Deleted member 16349]

Beware the wrath of Gordo! LOL ;-)


RE:

In newspaper work, when a 3/c grey bar is used at the foot of the page a densitometer is used to measure the 3/c bar to get SID information which then informs the press operator about any needed ink key adjustments.



best, (wrath over) gordo

Just for clarity.

I hope the comment is not implying that the operator can use the 3 SID values to directly adjust the ink.

Yes, it will tell them that an adjustment is needed but there is no direct relationship with the three SID values of CMY in the screen to the actual CMY ink levels. Judgement on the operator's part is required with this SID information.

 

[danremaley]

densitometer

densitometer

Hi,

I am not sure that this point was discussed before or not. Which device should be better to the newspaper pressroom: densitomter or spectrophotometer? At present we are scanning gray bar (not black) on the color pages and try to keep the C, M and Y density values within given specification on the presses. We use status T (absolute) on densitometer. It happens that due to different shades of paper: same density may have different L, a and b values. Also spectrophotometer may be useful to check the Delta E between proofs and printed copies and also 'b' value of the paper. It would be great to have some information on which device we should apply and why?

Thanks.

Here's my teaching method- use a densitometer set with "all" filter readings. The "V" number is the "visual" density reading, generally around .60. The 3-filter readings show Y-M-C representing each filter reading. If the "V" number is high - then you need to lower the higher Y-M-C numbers, if the "V" number is low, then you need to raise the lower Y-M-C numbers. Getting all 3-readings the same represents gray. The readings may indicate high or low and they could be because of dot gain OR solid ink density. If the solid ink density is correct then it's a dot gain issue. Call with any questions 412.889.7643.
Dan

 

[danremaley]

Here's my teaching method- use a densitometer set with "all" filter readings. The "V" number is the "visual" density reading, generally around .60. The 3-filter readings show Y-M-C representing each filter reading. If the "V" number is high - then you need to lower the higher Y-M-C numbers, if the "V" number is low, then you need to raise the lower Y-M-C numbers. Getting all 3-readings the same represents gray. The readings may indicate high or low and they could be because of dot gain OR solid ink density. If the solid ink density is correct then it's a dot gain issue. Call with any questions 412.889.7643.
Dan

 

[Prepper]

That is what I promote, use the densitometer to check gray balance first, get balanced, use those SIDs to run the job. In our experience with G7, when press is gray balanced, there's a pretty good proof to press sheet match. That one single measurement will tell you everything. If you have to move SIDs too far to get balanced, you throw your solids out of Lab tolerance, the match is not so good in some areas depending of course on the inks affected and the colors in the images. The gains are probably out of tolerance, higher or lower, than they were when curves targets were ran.

I wish I could get pressman to check that first. Too often he "knows" it won't work and starts adjusting color to try and match proof before getting balanced and 1-2 hours later and part way thru the run gets a pretty good match, check it and guess what, now it's also gray balanced! Just can't get that through to him though even when it happens several times.

I know what Gordo is saying about the gray and color not having a lot to do with each other when using MaxGCR, but I also know that when the press gets close to gray balance, it matches very close to our G7 certified proof.

Another thought lately, with MaxGCR, if SIDs are being adjusted too far (out of Lab tolerance) trying to achieve gray balance, would I be better off going with a MediumGCR? Would that give pressman a larger window to adjust SIDs more with smaller moves and maybe keep them in Lab tolerance?

 

[gordo]

Just for clarity.

I hope the comment is not implying that the operator can use the 3 SID values to directly adjust the ink.

Yes, it will tell them that an adjustment is needed but there is no direct relationship with the three SID values of CMY in the screen to the actual CMY ink levels. Judgement on the operator's part is required with this SID information.

I'm just reporting what is done in practice - I'm not advicating it. The notion is that there were three associated CMY values to the grey balance patch when the press was initially set up. The press operator then uses judgement to move SIDs until those values are reached.

Gordo

 

[danremaley]

Felix Brunner says that color shifts are caused 80% of the time, because of dot gain, only 20% because of solid ink density. So what do we measure? SID . . .go figure!
The absolute best color is achieved when the solids are near correct and the sheet is in gray balance, even if it's a little light or dark.
Standard deviation on an offset press is +/- 2% midtone gain. 50% GCR will eliminate this shift, it takes 80% GCR to eliminate a 4% variation. The "secret" is to keep the gains in balance to maintain gray. . .

 

[Deleted member 16349]

Felix Brunner says that color shifts are caused 80% of the time, because of dot gain, only 20% because of solid ink density. So what do we measure? SID . . .go figure!
The absolute best color is achieved when the solids are near correct and the sheet is in gray balance, even if it's a little light or dark.
Standard deviation on an offset press is +/- 2% midtone gain. 50% GCR will eliminate this shift, it takes 80% GCR to eliminate a 4% variation. The "secret" is to keep the gains in balance to maintain gray. . .

I would argue that much of the shift in dot gain is caused by a shift in SID ink film.

The colour of the solid print may not be perceived as shifting much but the resulting dot gain due to the change in ink film that printed the dot will show the colour shift.

Therefore controlling SID is important. Not because of a shift in colour in the solid but because it affects dot gain which does shift colour.

 

[danremaley]

dot gain

dot gain

I would argue that much of the shift in dot gain is caused by a shift in SID ink film.

The colour of the solid print may not be perceived as shifting much but the resulting dot gain due to the change in ink film that printed the dot will show the colour shift.

Therefore controlling SID is important. Not because of a shift in colour in the solid but because it affects dot gain which does shift colour.

Studies have suggested that even an increase of .20 density (which would be beyond the standard range) would only increase the dot gain 2%. Dot gain is changed by roller pressure, water balance, viscosity, temperature, ph&conductivity, more so than SID

 

[danremaley]

Here's my thought's on the color bar. If you apply the GCR values (50%) that represent standard deviation variance (+/-2% gain) to the gray patch, the only "change" you will see or measure will be beyond "standard deviation". . .even if the images are 80% GCR and won't change. . the press will be on a measurable path BEFORE the color shifts.
Also research has shown that a 2% gain shift represents a Delta 2 - a 4% gain shift represents a delta 4 in neutral gray. so you will never be a "delta 2" printer because that's standard deviation!

 

[gordo]

[snip]
I wish I could get pressman to check that first. Too often he "knows" it won't work and starts adjusting color to try and match proof before getting balanced and 1-2 hours later and part way thru the run gets a pretty good match, check it and guess what, now it's also gray balanced! Just can't get that through to him though even when it happens several times.

I know what Gordo is saying about the gray and color not having a lot to do with each other when using MaxGCR, but I also know that when the press gets close to gray balance, it matches very close to our G7 certified proof.

Another thought lately, with MaxGCR, if SIDs are being adjusted too far (out of Lab tolerance) trying to achieve gray balance, would I be better off going with a MediumGCR? Would that give pressman a larger window to adjust SIDs more with smaller moves and maybe keep them in Lab tolerance?

First off, there's no such thing as a G7 proof. G7 is a method for grey balancing an output device - it is not a color standard/specification. A proof that's been set up to represent an industry standard specification will, by definition, be in grey balance.

The press operator should not be making color on press. That's not his/her job. So, adjusting SIDs to better align press to proof can be problematic - except in specific circumstances (e.g. to deal with inline issues). Adjusting SIDs to achieve grey balance in the color bar is pretty much a complete waste of time. Not only because the grey patch in the color bar is disconnected from the live image area (including neutral greys) but also because a press is mechanically quite different than a proof as far as making color is concerned.

gordo

 

[Prepper]

Ok, why then when the press gets a close match to the proof, it is also "gray balanced"? When it is color shifted away from gray, not the best matches. In that case, how do you suggest adjusting the color to match closer on a live job right then? Talking smaller tweaking moves here, not large moves, because large moves throw the solids off.

Are there presses that can be started up, ran to target densities and the color is right and doesn't have to be tweaked at all? If it is being tweaked, it is towards something, more or less color, and gray balance is something that can be checked with a single measurement and provide info on leveling out the color, to meet the NPDC curves which are applied to the plate, which the profile is geared toward I assume (maybe not?), so it seems logical to me that moving to gray balance the sheet would improve the proof to sheet match, which is my only concern here. I really don't care how it gets there, if the proof I produce is a very close simulation of what the printed sheet looks like off the press, that is a success to me.

G7 proof to me here means one that is verified "Ok" off my Epson 7900 and EFI rip to meet the Gracol conditions if that's a better term for you. Spec, standard, conditions, whatever you call it, G7 is the process we use here to print. That to me just means we're using the Gracol profile, proofing with that profile, NPDC curves, and when the press is gray balanced it matches really well and gets to color very quickly.

Probably don't use the "correct" terminology exactly but I do know the results we get are pretty good.

Thanks, and keep teaching, I've learned a lot from everyone's experiences and insights on this forum and no doubt have a lot more to learn.