Push, pull and dev choice

[Roger Hicks]

A lot of people talk about 'pushing' and 'pulling' films as if any deviation from the box speed were a 'push' or 'pull'.

It isn't. ISO speeds are determined sensitometrically (not in camera) and the film is developed in a stated developer to a standard contrast (gamma approximately 0.62). The film speed is based on the exposure needed to give a density of 0,10 above film base plus fog.

Develop to a higher gamma and you will lift the speed point: this is 'pushing'.

Develop to a lower gamma and you will depress the speed point: this is 'pulling'.

Develop to the same gamma in a different developer that gives a different speed and you are neither pushing nor pulling: you are changing ISO speed through developer choice. Fine-grain devs can knock 2/3 stop or more off the true ISO speed, so that an ISO 400 film drops to ISO 250 or less. Speed-increasing devs, often loosely though inaccurately called 'push' devs, can add 2/3 stop or a little more, raising an ISO 400 film to ISO 650 or better.

Of course you can also push or pull in these other developers: HP5 at EI 1000 in Microphen is about a half-stop push (small increase in contrast) while HP5 at EI 1000 in Perceptol would be more like a 1-1/2 stop push (bigger increase in contrast).

Merely rating the film differently is something else again. If you rate HP5 at 200 but develop to the standard contrast, you are not pulling: you are simply overexposing. This will reduce sharpness and increase grain size but you may prefer the tonality.

This is before you start considering equipment variations such as slow shutters and variations in lens flare, or variations in metering technique such as incident, spot readibng shadows, etc.

Why post this? Partly to clarify matters for people whjo don't understand the terms and partly as bait for my web-site, www.rogerandfrances.com -- take a look at the Photo School.

Cheers,

Roger

 

[sunsworth]

I think one of the exceptions to this is Delta 3200 which just says '3200' on the box and not 'ISO 3200'

I used to shoot and develop Delta 3200 at 1600, but I prefer the look of Neopan 1600 these days.

Steve

 

[Poptart]

Will this be on the test?

 

[Roger Hicks]

Dear Sunsworth,

Absolutely, and thanks for pointing it out.

I take your point on Neopan at 1600 (just over a 1-stop push)but I think Ilford is better when you start to push it to 3200 and beyond; after all, the Ilford material is nearly twice as fast in ISO terms.

Cheers,

Roger

 

[sunsworth]

Roger, I understand what you're saying, but I prefered the look of Neopan 1600 at 1600 to that of Delta 3200 at 3200 or 1600.

Last night I was looking at some prints that someone I know made in a Jazz club. Neopan 1600 pushed to 6400 and developed in Microphen, and they looked stunning. Far better than I expected.

Steve

 

[Roger Hicks]

Dear Steve,

Once again, I wouldn't argue. You prefer one; I prefer another. That's how they both stay in business. All I was after pointing out was the terminology.

Cheers,

Roger

 

[sunsworth]

No problem Roger. It would be a very boring world indeed if we all agreed on everything.

Steve

 

[BJ Bignell]

Roger, thanks for the info. Sensitometry is far beyond what most people are willing to deal with and actually use, at least for most amateur or recreational photographers. However, it's good to understand the basics, and at least have an idea of what's happening on your film.

I want to try to understand this correctly; correct me if I'm wrong anywhere:

When pushing, you usually underexpose the film. To compensate, you overdevelop (or at least, develop more) to bring details up off of the toe, and more towards the middle of the curve. Because of this higher development, however, the gamma will increase (overall slope of the curve increases). So, if in-camera you shoot HP5+ @ 800 and then develop to a higher contrast, it's a push.

But, if you use a dev that gives a higher speed, and develop the same film to normal contrast in that developer, it's not a push? You're simply taking advantage of the characteristics of a different developer, and getting more image with less exposure? However, this speed-increasing developer could also be used to push, but by developing to higher contrast?

Did any of that make sense? Am I just talking in circles? I feel dizzy...

 

[varjag]

But, if you use a dev that gives a higher speed, and develop the same film to normal contrast in that developer, it's not a push? You're simply taking advantage of the characteristics of a different developer, and getting more image with less exposure? However, this speed-increasing developer could also be used to push, but by developing to higher contrast?

A speed enchancing developer increases the actual speed and retains the shadow details at that effective speed. A push does not produce any more detail in shadows; it "amplifies" what the developer could bring out for given exposure.

Naturally, the developer+film combos providing higher effective speeds to start with are better choice for pushing.

 

[BJ Bignell]

Alright, so we're amplifying what shadow detail can be given by the current developer? We make the details in the shadows more obvious, but we don't necessarily get more details from the push?

 

[Roger Hicks]

Dear BJ,

That's pretty much it, except that you're not moving it OFF the toe; you're simply increasing the slope of the toe, so the further along the toe a given point is, the more density it has. This is why films that push well have long toes.

With a speed increasing developer, as Eugene says, the whole toe is moved to some extent upwards; wih a push, it starts off at the same point but the contrast increases faster.

The question is the point at which the overall contrast becomes unacceptable, and this varies with the subject (low-contrast subjects can be pushed more), total flare (lens/body flare reduces contrast but also changes tonality) and personal preference.

The great thing about the ISO standard is precisely that it's standardized; the advantage of personal testing, based on ISO speeds (because they're the best we have, far above the maunderings of self-proclaimed experts) is that we can see what we like best and how to get there.

Cheers,

Roger

 

[Todd.Hanz]

So what am I doing when I "pull" Tri-X to 200 and develop as ISO 200? Is that considered an actual pull?

Todd

 

[BJ Bignell]

If you're overexposing it and then developing it to lower contrast less than normal - for your chosen developer - then it would be a pull. If you were getting 'normal' contrast (based on whatever your normal was) by using a different developer, then that wouldn't technically be a pull.

 

[GeneW]

A new factor in all this is scanners. I find negatives that are thinner scan better, but preserve shadow detail well (if it's there). I've been exposing Tri-X at 800 and increasing development around 10% over my iso400 times and getting good scanner negs that would probably be a little thin for enlarging. Not scientific -- just an observation... with all the usual cautions regarding varia such as meters, water quality, agitation, etc.

Gene

 

[Roger Hicks]

BJ's got it apart from one techical quibble -- it's not a pull in strict ISO terms unless you are developing to below ISO contrast.

In practical terms, he's spot on: if you are over-exposing and under-developing, that's a 'pull', just as under-exposing and over-developing is a 'push'.

The point is that few people can be bothered to calculate the actual gamma to which they develop (OK, G-bar or C.I.), and it doesn't really matter all that much. I can do it, and I have done it, but only when I had to write about it.

Simple over-exposure, with development to normal contrast, often gives results that people prefer, and it may be that is what you are doing: without running sensitometric tests on your negatives, no-one can say.

Another possibility is that your metering technique (in bright sun, anything other than spot metering the shadows) effectively under-exposes the film, so you are actually rating the film at or close to its ISO speed.

Or you may be using a developer that gives a true ISO 200 (ISO speed point and ISO contrast)

None of this is 'wrong' in terms of getting the negative that gives the result you like most. What is important, though, is having a good idea of where the changes may lie, and avoiding terminology which may confuse other people. I'm not saying for a moment that you do it, but a lot of people do use 'pull' when they simply mean 'overexpose' and that was what I was trying to overcome.

Cheers,

Roger

 

[Roger Hicks]

Dear Gene,

Very scientific, in the strictest sense of putting forward an idea; making predictions based on that idea; and then testing the predictions to see if you are right.

I don't know but I strongly suspect that you are correct that underexposed negatives scan better, though I would be less confident about over-development: I'm not arguing, I genuinely don't know because I have never tried it with a range of films, scanners, etc.

What you are doing, of course, is underexposing and (slightly) overdeveloping, which constitutes a true push. But I have heard others say they get the best results for scanning with normal exposure and underdevelopment, or over-exposure and still greater under-development, i.e. thinness is more important than contrast. Your way sounds easiest and probably best.

Cheers,

Roger

 

[GeneW]

I don't know but I strongly suspect that you are correct that underexposed negatives scan better, though I would be less confident about over-development: I'm not arguing, I genuinely don't know because I have never tried it with a range of films, scanners, etc.

What you are doing, of course, is underexposing and (slightly) overdeveloping, which constitutes a true push. But I have heard others say they get the best results for scanning with normal exposure and underdevelopment, or over-exposure and still greater under-development, i.e. thinness is more important than contrast. Your way sounds easiest and probably best.

Roger, that sounds solid to me. I think thinness, without being too contrasty, is the key to good scanning. Thinness seems to reduce grain somewhat too, in scanners. Maybe a little less aliasing?

Gene

 

[Roger Hicks]

Dear Gene,

Less exposure always reduces grain for a given development regime, so I don't know if its aliasing reduction or straight grain reduction. But reduced development also reduces grain. I don't know enough about aliasing to know if there are 'sweet spots', i.e. grain of a particular size scans better than smaller OR bigger. If there is -- and I suspect it would vary with scanner design -- maybe you've hit one of those points.

Have you tried either of the other two regimes -- say EI 650 or maybe 500 and normal development, or EI 320 and reduced development (by 20 oer cent, say)? Or have you (as most of us have) better things to do than to change a winning game?

Cheers,

Roger