Epson r800 neg colour and ChartThrob

[gwatson]

Hi all

I'm in need of some advice. I have just got an r800 for dig negs and plan to proceed as follows:

• Establish standard print time
• Establish rgb values for pt/pd (NA2)
• Use ChartThrob to get a starting-point curve

Is there anyone out there that uses an r800 for pt/pd that can stear me in the right direction as far as colour values go. Basically what colour values should I experiment with? I've read that greens are good with most. Is this a good starting point?

Also, once I have a colour that works, can I apply this to the ChartThrob chart to establish a curve?

Bit new to all of this, so if anyone has any advice, I'll promise to remember you in my will.

Many thanks

Geoff

 

[sanking]

Hi all

I'm in need of some advice. I have just got an r800 for dig negs and plan to proceed as follows:

• Establish standard print time
• Establish rgb values for pt/pd (NA2)
• Use ChartThrob to get a starting-point curve

Is there anyone out there that uses an r800 for pt/pd that can stear me in the right direction as far as colour values go. Basically what colour values should I experiment with? I've read that greens are good with most. Is this a good starting point?

Also, once I have a colour that works, can I apply this to the ChartThrob chart to establish a curve?

Bit new to all of this, so if anyone has any advice, I'll promise to remember you in my will.

Many thanks

Geoff

Geoff,

I have never jused the R800, but I do know the R1800 and I am told that the pigmented ink set is the same.

If so, you won't get enough UV blocking density with the green, at least not enough for straight palladium. Your best bet IMO is to desaturate the RGB file and print in color. This will give you a neutral black looking negative with enough density for palladium.

Sandy

 

[gwatson]

Thanks for your reply, Sandy. I guess the bright side is that it is one thing less to worry about.

However, when I spec'd it, the r800, r1800, and the r2400 all use the K3 ink set and I thought this ink set was quite sound for dig negs. Is here a difference in the way the printer lays the ink, or am I missing something fundamental?

Many thanks

Geoff

 

[sanking]

Thanks for your reply, Sandy. I guess the bright side is that it is one thing less to worry about.

However, when I spec'd it, the r800, r1800, and the r2400 all use the K3 ink set and I thought this ink set was quite sound for dig negs. Is here a difference in the way the printer lays the ink, or am I missing something fundamental?

Many thanks

Geoff

Geogg,

I don't really know the answer to that question. Last year I purchased an R1800 believing the same thing, i.e. that since it used the K3 pigmented ink set it would give the same UV blocking density as the R2400. But that is not the case.

As I mentioned, I have no actual experience with the R800.

However, if you desateurate the RGB file and print it in color you will get a negative with almost perfect density range for straight palladium, so you proceed directly to the next stage of establishing your curve, and then printing. And the R1800 prints very smoothly this way.

Sandy

 

[gwatson]

Hi Sandy

'Tis strange, but life would be so dull (and perhaps a little cheaper) if everything went as expected.

Very glad to read that the density is good for palladium. Your initial post left me wondering how on earth I tweek the density.

I only got the thing out the box tonight, but will do some testing over the next couple of days. Just to confirm, unless you tell me otherwise, when you say desaturate, that means I use the desaturate menu item on the rgb file and print as a colour image (bit of a duffer when it comes to all this dig stuff).

Anyway, thanks again for your time, Sandy.

Geoff

 

[sanking]

I only got the thing out the box tonight, but will do some testing over the next couple of days. Just to confirm, unless you tell me otherwise, when you say desaturate, that means I use the desaturate menu item on the rgb file and print as a colour image (bit of a duffer when it comes to all this dig stuff).

Geoff

Yes, in Photoshop make sure your file is in RGB mode, then go to Adjustments>Desaturate. Then print the file with Color selected.

Sandy

 

[Katharine Thayer]

Desaturate v greyscale?

Yes, in Photoshop make sure your file is in RGB mode, then go to Adjustments>Desaturate. Then print the file with Color selected.

Sandy

I'm curious; why not convert the file to greyscale mode and print the greyscale file in color? Converting to greyscale saves the space of two channels, if you're working with a big file, and the pixel information is the same either way, I believe. Does the color mode make a difference in how the color inks are laid down?
Katharine Thayer

 

[sanking]

I'm curious; why not convert the file to greyscale mode and print the greyscale file in color? Converting to greyscale saves the space of two channels, if you're working with a big file, and the pixel information is the same either way, I believe. Does the color mode make a difference in how the color inks are laid down?
Katharine Thayer

It definitely makes some difference, though I don't know exactly how or why. However, measuring the transmission densities will show a difference when comparing an RGB file dasaturated and printed in color to a greyscle file printed in color. At least, whenever I made the comparison that is what I found.

Also, you will see a difference even in the schedule of how you apply the curve, depending on whether you do this before or after convering to RGB mode.

Years ago I started to print this way and have continue to do some for consistency. It might be that you could get perfectly acceptable results prnting the greyscale file in color as you suggest.

Sandy

 

[gwatson]

Hi all

Many thanks for the comments and advice so far.

I set the printer up as advised: destaurate and print in colour. The step wedges printed smoothly with little to no grain, but showed a very narrow tonal range from max black to white over about 4 steps. My standard print time to achieve max black through clear film is 6 mins and this produced 4 black steps, about 4 midtones, and 4 steps at paper white. This produced a very aggresive curve which I think is trying to do too much.

So, I wonder if I established my standard print time using too much NA2. My drop count was 5:5:1 (@ 5%). Presumably, if I reduce the contrast of the solution, I will get a longer scale, but I don't know what the minimum amount of NA2 I need to use to prevent fog. This is only one drop at 20% for an 8 x 10. How much lower can I go? Can anyone help?

Cheers

Geoff

 

[donbga]

Hi all

How much lower can I go? Can anyone help?

Geoff

Try making a test without NA2 to see if you do get fog if you want a longer scale or use a 1S misture if you do encounter fog.

Don Bryant

 

[wiz]

Geogg,

I don't really know the answer to that question. Last year I purchased an R1800 believing the same thing, i.e. that since it used the K3 pigmented ink set it would give the same UV blocking density as the R2400. But that is not the case.

The 2400, when printing grayscale, uses a tight dither of three shades of carbon black, which gives it excellent UV blocking capability. And good tracking between UV blocking and visible green and blue blocking. The 1800 only has one full strength black, it tries to make grays by using a mix of red, blue (yes, it has both of those), cyan, magenta, and yellow. The end result isn't that great for a print, and is horrible for a negative.

It is possible to control an 1800 with Quadtone RIP and build curves based on the UV blocking ability of each color. The negatives will look strange as heck, but they will work.

If you really want to make an 1800 sing for negatives, get MIS spongeless carts for the cyan and magenta positions, a chip resetter, a 4oz bottle of MIS "clear base", two empty 4 oz Nalgene bottles, and two Epson photo black carts. Get a 10ml syringe, and a bottom fill adapter.

Drain the photo black carts into a bottle, then transfer the ink into the other bottle using the 10ml syringe (or a decent graduated cylinder) so you know to within 1/2 ml how much ink you have. Should be around 28ml. Then dilute it to 32% with the clear base.

Thoroughly clean the first bottle. Draw 20ml of the new 32% solution, transfer to the other bottle, and add 43ml of clear base, to make a 10% solution. You've now got about 60ml each of 32% and 10%. Fill the magenta cart with 32%, the cyan cart with 10%.

Replace the cyan and magenta carts with the new ones, and print a CMYK purge pattern until the cyan and magenta turn gray. You've wasted about $2 in ink on the purge, a small price to pay.

Build a curve for that with QTR at the highest resolution. You've now got the same three blacks as the 2400, but drops half the size of those on the 2400, so you're going to clobber the 2400.

If you want to dedicate the 1800 to negatives, pull out all the Epson inks, get a 4oz bottle of MIS 1800 photo black, a 16oz bottle of 1800 glop (glop is the dilutant for MIS photo black, "clear base" is the dilutant for Epson photo black) and new carts for all 8 positions. Now load the machine with 100%, 56%, 32%, 18%, 10%, 5.6%, 3.2%, and 1.8%, and build a QTR curve. That is as good as a negative from an inkjet gets. Build a PhotoShop curve as was outlined by Dan Burkholder. No PDN, no Chart Throb.

As I mentioned, I have no actual experience with the R800.

It is, to all intents and purposes, identical to the 1800.

However, if you desateurate the RGB file and print it in color you will get a negative with almost perfect density range for straight palladium, so you proceed directly to the next stage of establishing your curve, and then printing. And the R1800 prints very smoothly this way.

You've not begun to tap into what an 1800 can do.

 

[wiz]

It definitely makes some difference, though I don't know exactly how or why. However, measuring the transmission densities will show a difference when comparing an RGB file dasaturated and printed in color to a greyscle file printed in color. At least, whenever I made the comparison that is what I found.

It has to do with the particular grayscale and color working spaces you've set on your PhotoShop.

When you desaturate, you're still in the color working space. Both Adobe RGB and sRGB have approximately the same L curve, almost gamma 2.2.

When you convert to grayscale, you convert to the working monochrome space, which is probably dot gain 20%, not very much like gamma 2.2. If your PhotoShop is set to a working grayscale space of gray gamma 2.2, the change is harmless.

 

[Katharine Thayer]

When you desaturate, you're still in the color working space. Both Adobe RGB and sRGB have approximately the same L curve, almost gamma 2.2.

When you convert to grayscale, you convert to the working monochrome space, which is probably dot gain 20%, not very much like gamma 2.2. If your PhotoShop is set to a working grayscale space of gray gamma 2.2, the change is harmless.[/QUOTE]

That's interesting, thanks; you cleared up a mystery for me. Being a person who always has to see something for myself, I was checking out the statement in the photoshop Bible that says that desaturated RGB is the same as greyscale, by isolating individual pixels in a file and looking at the the pixel information, including HSB values, under the different modes, and finding the results somewhat confusing. But after reading your message and switching to gray gamma 2.2, suddenly the numbers started coming out right. So I'd amend my earlier statement to read that if you set your grey workspace to gray gamma 2.2, then you can probably use desaturated RGB and greyscale interchangeably.
Katharine

 

[sanking]

You've not begun to tap into what an 1800 can do.

I am sure you are right about that, but I only experimented with the R1800 for a few days. The UV blocking of the colored inks was poor, and the inks also dried much slower than on my 2200, leaving pizza wheel marks.

However, I have seen palladium work from digital negatives printed on the R1800, simply desaturating the RGB file and printing in color, and the tones were very smooth and detail outstanding. In fact, the results were on a par with printing pt/pd directly from LF and ULF negatives. So I woud have to say that if the goal is printing with alternative processes, and you can live with a maximum DR of about 2.05, the R1800 will do a real nice job right out of the box.


Sandy King

 

[sanking]

Try making a test without NA2 to see if you do get fog if you want a longer scale or use a 1S misture if you do encounter fog.

Don Bryant

I am with Don in that you need to reduce or eliminate the NA2 from your mix. If you are desaturate the RGB file as i suggested, and then print in color, you should have a negative with a DR of about 1.85-2.00. This DR is very long scale, almoste optimum for straight palladium, and should require little or no NA2 additive.

Also, you may just want to print a Stoffer continuous tone step wedge for a while until you are sure of the exposure scale of your pt/pd mix and paper. The difficult part of acquiring precision in printing with digital negatives is not the negative, but control of the process itself.


Sandy King

 

[gwatson]

Thanks again for the comments.

To update, I started again this time using only 1% NA2 and the scale was greatly improved. I recalculated the curve and this one is much more sensible. Printing using the desat image and colour inks is producing amazingly smooth tones, comparable to film. I applied the curve to the test image, (I have not been able to print this yet due to a delayed shipment of OHP), and measured and compared the values in the problem areas on both images. The horrible flat mid/high tones in the original image now show tonal separation.

This is all very exciting. Despite my initial doubts, it is now looking possible to produce a print from a dig neg that will be almost indistinguishable than one from film. I'm sure the curve will need tweaking, and I need to get rid of the pizza wheels, but I'm closer.

Thanks again for your advice.

Geoff

 

[gwatson]

Hi all. I'm now achieving half-decent results. However, I have a few more questions to bother you with:

1. I seem to be having a problem with dust on the OHP material. The negatives are peppered with very small clear specks leading to black specks in the highlights of the print. The OHP (Permajet) has so much static that blowing the dust off just doesn't cut it and I don't really want to handle the print surface. How do other people handle this?

2. Should I be scanning the neg in 24 bit rgb, or 16 bit grayscale and converting to rgb after? I could scan at 48 bit but the file sizes are silly. Since I resize the image to print at 360 dpi, would it make sense to scan in 48 bit but at a lower resolution?

Cheers

Geoff

 

[sanking]

Hi all. I'm now achieving half-decent results. However, I have a few more questions to bother you with:

1. I seem to be having a problem with dust on the OHP material. The negatives are peppered with very small clear specks leading to black specks in the highlights of the print. The OHP (Permajet) has so much static that blowing the dust off just doesn't cut it and I don't really want to handle the print surface. How do other people handle this?

2. Should I be scanning the neg in 24 bit rgb, or 16 bit grayscale and converting to rgb after? I could scan at 48 bit but the file sizes are silly. Since I resize the image to print at 360 dpi, would it make sense to scan in 48 bit but at a lower resolution?

Cheers

Geoff

I can not help you with the dust problem. This is not something I have experienced before.

As for the scanning question, I assume that the final product is a digital negative for printing pt/pd? What kind of original are you scanning (size, color or B&W, negative or positive)? If you are scanning a B&W negative my advice is to scan in 16 bit grayscale, unless it is a 35mm or medium format pyro stained negative. If it is a color negative or transparency, 35mm or medium format, you probably will get better results by scanning in RGB and then throwing away the red and blue channels (or adjusting the tonal values on each channel separately). The green channel usually has less noise/grain, and is sharper. You might also benefit by scanning 35mm and medium format pyro stained negatives in RGB and then using just the green channel (or adjusting each record individually).

Sandy King