split filter contrast adjustment calculations

nwilkins · Oct 18, 2016

Hi everyone, just wondering if there is a rough rule of thumb that exists if you are split-filter printing and you have a work print which needs a little more contrast both in the blacks and in the highlights. In other words, the 00 time needs to be reduced and the 5 time needs to be increased.

Is there any way to calculate the degree of change for both exposure times which would still give roughly the same overall print brightness? Or do I just have to go back to the beginning and calculate new times for both filters again using test strips?

Maybe thinking in terms of stops would work (ie reduce 00 by a quarter stop and increase 5 by a quarter stop) but in instances where one time is long and one is short it seems like using stops wouldn't make sense. ie 24s (0) and 4s (5) will be a much lighter print than 48s (0) and 2s (5).

ic-racer · Oct 18, 2016

I made a chart/graph with my own emperic data. A lot of work! I wound up getting variable filtration and use that now.

nwilkins · Oct 18, 2016

Wow how extensive is the chart? Does it reveal any patterns a person could use to calculate increases/reductions? Would reducing and increasing by the same percentage work in most instances, at least when the times for both are pretty similar?

nwilkins · Oct 18, 2016

also I'm wondering if "percentage of total time" is better to think about than stops of light when printing?

they don't quite give the same results. increasing or decreasing a 20 second time by half a stop gives 30 or 15. Whereas increasing or decreasing a 20 second time by 25% gives 25 or 15.

Frank53 · Oct 18, 2016

nwilkins said:
Hi everyone, just wondering if there is a rough rule of thumb that exists if you are split-filter printing and you have a work print which needs a little more contrast both in the blacks and in the highlights. In other words, the 00 time needs to be reduced and the 5 time needs to be increased.

Is there any way to calculate the degree of change for both exposure times which would still give roughly the same overall print brightness? Or do I just have to go back to the beginning and calculate new times for both filters again using test strips?

Maybe thinking in terms of stops would work (ie reduce 00 by a quarter stop and increase 5 by a quarter stop) but in instances where one time is long and one is short it seems like using stops wouldn't make sense. ie 24s (0) and 4s (5) will be a much lighter print than 48s (0) and 2s (5).

Maybe I do not understand it correctly, but, if you already have a work print and you already figured out how much it should roughly be, why don't you just dodge 0 during initial exposure and than burn 5 afterwards? That's the way I print with a Heiland split grade system.
Regards,
Frank

nwilkins · Oct 18, 2016

Frank53 said:
Maybe I do not understand it correctly, but, if you already have a work print and you already figured out how much it should roughly be, why don't you just dodge 0 during initial exposure and than burn 5 afterwards? That's the way I print with a Heiland split grade system.
Regards,
Frank

Hi Frank, I am wondering about a work print which is too low contrast, so essentially I am asking if there is a way to calculate adjustments in the base times for both 00 and 5 without starting again from scratch.

Frank53 · Oct 18, 2016

If you don't have a possibility to measure contrast, it's a matter of trial and error, but if contrast is too low, I would start with adding say 20% to 5 time and only touch 0 if the highlights are really too dark. Otherwise you have too many variables.
But you could still do this by burning-in.
Maybe you should look at the way you determined the exposure times for the work print. If contrast in the workprint is too low, the relation between the two exposures is not correct, in this case too little 5.
How much extra 5 exposure, depends on the print in front of you.
I'm afraid you have to spend some extra paper on that.
Regards,
Frank

Jim Noel · Oct 18, 2016

Make a full image test sheet with the #5 filter going one direction, and the #0 at 90 degrees. After drying the test sheet, chose the best pairing.

Bob Carnie · Oct 18, 2016

I use identical aperture and time settings for the Low Filter and the High Filter.

I will make a test with both hit one after the other,

Figuring out the starting point for Low and the time you use is pretty easy.

Once I have a good Low Filter I now look at the print with an equal hit of the 5 filter.

If I need overall more contrast I will hit the 5 filter a second time, sometimes it will be multiple times. After time and multiple images you get a 6th sense about a certain negative range, original lighting range and also how the image needs to be presented.

I have found over thousands of split prints that this method really works for me as I can concentrate on the image and the only adjustments required is keeping track on how many times on hits the 5 filter exposure..

Obvious question is how do I deal with contrasts that require e a tweak more or less.. I have been trained to count in my head all exposures and I have mentioned many times here I like a 10 second blast so I just put my hand under the lens for a percentage of the time
by counting down. I will never ever change the aperture or timer now other than finding the overall timer setting to work. this allows me to concentrate completely on my split filter and dodge and burn sequences.

ic-racer · Oct 18, 2016

nwilkins said:
Wow how extensive is the chart? Does it reveal any patterns a person could use to calculate increases/reductions? Would reducing and increasing by the same percentage work in most instances, at least when the times for both are pretty similar?

This graph was constructed from data obtained using a cold-cathode light source, Ilford MG paper, and Rosco filters. I don't use that system any more, I use a single base exposure, combining graded green and blue together.

ic-racer · Oct 18, 2016

The radial lines are ISO(R) contrast numbers. The colored lines are just colored to keep one's eye from wandering as you follow the curve.
For example if you have 20 seconds of blue and 5 of green, it is about ISO(R) 0.7 (or 70 or Grade 4). If you want to change contrast and not alter overall print density, then you can use 6 sec blue and 25 seconds of green and get about ISO(R) 1.35 (or 135 or Grade 1.5).

Brook Hill · Oct 18, 2016

If you use the Dave Butcher method of determining the exposure at grade 2 1/2 then use
half the time at grade 0 and half at grade 5 you can alter the contrast by upping the grade 0 or reducing the grade 5 values. Altering a grade by 1 grade would alter the contrast by half a grade.

Tony

DREW WILEY · Oct 18, 2016

Depends on the respective densities of the specific filters you're using, in relation to the actual color temperature of your enlarger light, which trends
toward warm. So the net effect is that the low-contrast filter (green or yellow) is going to pass significantly more light, and therefore need less time than the high-contrast filter with greater density (blue or magenta). I ignore all that grade nonsense and just do it. If I want to speak about grades,
I use real graded paper. Otherwise, that vocabulary is redundant, especially when split printing.

RalphLambrecht · Oct 19, 2016

nwilkins said:
also I'm wondering if "percentage of total time" is better to think about than stops of light when printing?

they don't quite give the same results. increasing or decreasing a 20 second time by half a stop gives 30 or 15. Whereas increasing or decreasing a 20 second time by 25% gives 25 or 15.

please note that you have an error in your stop ca;curation;going from 30 to 15s is a decrease of a full stop not half a stop

silveror0 · Oct 19, 2016

nwilkins said:
Hi everyone, just wondering if there is a rough rule of thumb that exists if you are split-filter printing and you have a work print which needs a little more contrast both in the blacks and in the highlights. In other words, the 00 time needs to be reduced and the 5 time needs to be increased.

Is there any way to calculate the degree of change for both exposure times which would still give roughly the same overall print brightness? Or do I just have to go back to the beginning and calculate new times for both filters again using test strips?

Maybe thinking in terms of stops would work (ie reduce 00 by a quarter stop and increase 5 by a quarter stop) but in instances where one time is long and one is short it seems like using stops wouldn't make sense. ie 24s (0) and 4s (5) will be a much lighter print than 48s (0) and 2s (5).

I think you could benefit from this chart, provided in Ralph's excellent book Way Beyond Monochrome Ed.2:
http://www.waybeyondmonochrome.com/WBM2/TOC_files/TimingExposureEd2.pdf

Michael Wesik · Oct 20, 2016

A couple thoughts...

I've never found a "rule of thumb" when it comes to split filter printing and I wouldn't limit your distribution of filtration to just Grade 00 and 5. There are tons of permutations to work with and no one methodology works better than another because the results are subjective.

I'm not sure I exactly understand what you're asking in terms maintaining the overall print brightness? Manipulating your exposure one way or another will invariably change the tonal distribution of your print...so if you reduce 00 and increase 5 you'll have a brighter print with more contrast. If you're looking for more contrast in your blacks and highlights, I would try a base exposure of Grade 0 or 1 and layer Grade 5 on to that. But again, you're tonal distribution will change.

I'm also unclear what you mean by starting from scratch? The way I figure it, you keep playing with the split until you find something that you're happy with and over time the process becomes intuitive.

split filter contrast adjustment calculations

Attachments

Privacy & Transparency

Privacy & Transparency