making large negs on an epson printer ?

A place to talk about photography, the meaning of life and anything that doesn't quite fit elsewhere
Post Reply
Dean Willcox
Posts: 28
Joined: Tue Apr 26, 2011 10:19 am Etc/GMT-1+01:00

making large negs on an epson printer ?

Post by Dean Willcox » Thu Dec 01, 2011 11:49 am Etc/GMT-1+01:00

having sold my 10 x 8 and finding myself with plenty of time on my hands I fancy having a go at platinum printing . I mentioned this whilst on a shoot and one of the guys said that you can produce negs for printing using your epson printer and a clear film ? with the advantage of adjusting contrasting via photoshop . to me it makes sense but is it viable ?
regards to all

patawauke
Posts: 56
Joined: Wed Sep 16, 2009 4:34 pm Etc/GMT-1+01:00

Re: making large negs on an epson printer ?

Post by patawauke » Thu Dec 01, 2011 12:38 pm Etc/GMT-1+01:00

Dean, search up on making digital negs, there's a whole science (or maybe craft) out there, loads of books have appeared on the subject, look up in amazon etc. It seems to have a growing following, and when I have time I certainly intend to pursue it.
John

banana_legs
Posts: 46
Joined: Tue May 25, 2010 6:45 pm Etc/GMT-1+01:00
Location: North Wilts

Re: making large negs on an epson printer ?

Post by banana_legs » Thu Dec 01, 2011 11:14 pm Etc/GMT-1+01:00

Dean,

Yes it is very possible and if you are happy to live with only up to 10x8, it can be done quite cheaply too on an A4 printer, but larger printers give you more flexibility. Be prepared though for lots of 'tweaking' to get the negatives looking correct to your eye. If you do have a printer capable of greater than A4, I would still suggest starting with 10x8 or smaller as the transparency material is much much cheaper in A4 size.

You can generate digital negatives on any inkjet printer really; it is just that for Epson printers, there is a software tool called 'QTR' which gives you full control over how your printer lays down ink. It helps to have either a densitometer, or a scanner and a reflective step-wedge in order to work out print densities (I use a Stouffer R2110 as a reference). It also would be nice to have a calibrated screen, however it is possible to live without it.

The basic process is to generate a step-wedge in a package such as photoshop and then print that onto transparency. You then contact print the wedge and can now use a densitometer to relate how say 10% black in photoshop actually translates to density in your print. With a bit of graph drawing, you can now work out a correction curve so that the pixel values on the screen correspond to specific density values in the print.

The first mistake people make is that they want to 'linearise' their print. What they actually mean is that say a step wedge in photoshop which has 11 levels, each changing by 10% (i.e. 0% 10% 20% .. 100% black etc) corresponds to log(density) on the print: a 50% pixel level on a print that ranges in log10(Density) from 0.1D to 2.1D therefore will correspond to (2.1 - 0.1)/2 +0.1 = 1.1D (where 0.3D is a 1 stop change). If the print was truly linear, 50% in photoshop would be only 1 stop darker than the 0% patch. The second mistake is that 'linear' is not actually ideal as although the response of the human eye is approximately logarithmic, it is not the best approximation in the world. A better approximation is to use the equations for how the eye perceives changes in density which will then map not to 'linear' log10(D) using the transform D to Y, where if D is your log10(density) measurements, Y=(D-min(D))/(max(D)-min(D)); but rather to the better approximation of (10^(Y/3)-1)/1.1544. As I mentioned, there is some graph drawing involved! You also have to be careful as you are mapping through printing negatives to positives etc. so need your curves to bend in the right direction.

I mentioned the calibrated screen; a prime mistake is to not check that the step-wedge in photoshop actually does give you a nice range of tones! If you have a calibrated screen, the steps in the wedge should look good, however it does depend on how you do the calibration i.e. if you try to 'linearise' the intensity of your screen, you are again assuming a direct logarithmic response of the eye. I tend to adjust my screen and save it as a monitor profile so that the step-wedge looks good with nice even steps, then work out a basic curve and tweak it until the steps in the printed wedge also look nice and even and then hey presto, if a picture looks ok on the screen, it should also look an equivalent ok in the print. That does assume that humidity, timings, coating density, phases of the moon, time of year etc. all conspire to give the same printing conditions as you printed the reference step wedge in!

As you can see, this is only my method of making digi-negs and as John mentioned, there are books written on the subject. On a practical note, hunt around local shops and on the web to find a small box of cheap inkjet negative material to begin with. The quality of the prints may be a bit grainy but at least the very many mistakes you make in finding out how the curves work/ don't work won't cost you too much. Once you are happy with how to calculate a curve, then go for the more expensive transparency material. I would warn against starting to try for a cyanotype curve to keep costs down as it prints with so few steps, good curves are not easy to make. I find VDB or Argyrotype are good for experimenting with curves as they are quite quick to dry/print/process and are quite low cost per print. Only when you think you can work out a good curve is it worth going for the more expensive metals.

Most of all, have fun!

Best regards,

Evan
More mad ramblings at http://blog.concretebanana.co.uk

Dean Willcox
Posts: 28
Joined: Tue Apr 26, 2011 10:19 am Etc/GMT-1+01:00

Re: making large negs on an epson printer ?

Post by Dean Willcox » Fri Dec 02, 2011 12:27 am Etc/GMT-1+01:00

Thanks for the info makes sense - some work to go ( not sure how my moniter is calibrated ) some work to do but welcome any tips ?

Post Reply