Jump to content
Sign in to follow this  
jmahto

Pushing a film in dev *same* as boosting exposure in scanning/PP

Recommended Posts

Well, after experimenting and scanning with several rolls in past few months I had this thought. 

 

Is pushing a film in chemical development *same* as either boosting exposure while scanning or increasing exposure in PP?

Look at this example of Tri-X 
 

Without any adjustment in scanning. On visual inspection also negative was on the lighter side (not sufficiently dense anywhere) and histogram on scanner confirmed that it was underexposed.

 

After +1 exposure in LR. It seems to me that this exposure boost in PP has increased the contrast in the same way chemical push would have done.

Why am I thinking about this? Because, if my observation is correct then I have the flexibility to push process frame by frame in PP after normal development.

 

And yes, that is E-Type on the road

Edited by jmahto

Share this post


Link to post
Share on other sites

Kinda-sorta.

 

"Pushing" is amplifying or multiplying or "cranking up the gain of" whatever base image is there. You start with a captured brightness range of between 0 and x, and if x is still too dark, you can multiply the whole range of brightnesses in the picture by, say, 4 (the "push"), and get a brightness range of between 0 to 4x (any number times zero is still zero, of course).

 

Result - a steeper line connecting 0 and the maximum (higher "gamma") = more contrast, plus a stronger maximum and mid-range - a picture you can see.

 

Thus all of the following do roughly similar things (assuming, in the case of a digital capture, that the sensor is "ISO-invariant"):

 

- underexpose a sensor by setting the camera to a higher ISO than base, and letting the camera "push" or multiply the image values for you

- underexpose a sensor by leaving the camera ISO setting at base ISO, but using a hand-held metered exposure (or exposure-compensation dial) to feed it too little light, and then doing the amplification in LR or Photoshop or whatever.

- underexpose film, and amplify the silver in it by extended development (pushing it chemically)

- underexpose film, and give it normal development, and then scan the film, and use LR to "crank up the gain" just like any other digital image.

 

The main obvious difference will be this: chemical "pushing" tends to increase the "grain/noise" in the highlights, by creating more and more silver in the only place there is any exposure to develop. Digital pushing tends to increase the "grain/noise" in the shadows, because that is where the signal/noise ratio is worst. *

 

Pushed film detail -                                                             Pushed (high ISO - 16000) digital -

clean shadows, noisy bright tones                                    noisy shadows, smoother highlights

 

Digital pushing is rather like recording sound with a weak mic or preamp, and then trying to recover the original sound by cranking the primary amp or playback volume up to 11 - you can hear the song, but also every tiny pop and scratchy bit of static noise.

 

There may be other differences, just due to the fact that the whole "film" thing tends to be non-linear, especially at the extremes of exposure (toe and shoulder of the film's response to light = "characteristic curve" ** ), and in development (where to compensate for a whole stop of underexposure (half the needed light) you give 1.4x (root-2) the development. You may find that playing with a "curve" in LR better replicates chemical pushing than an overall exposure bump of "+1". Or some combination of curves, exposure, contrast, shadows or highlights, etc.

_________________

 

* the one thing that frustrates me about digital - after 40 years of seeing pushed film pictures with (usually) grainy highlights and smooth empty shadows, the noisy shadows and "smoother" highlights of high-ISO digital just look - wrong - to me.

 

** http://www.sprawls.org/ppmi2/FILMCON/#The%20Characteristic%20Curve

 

Share this post


Link to post
Share on other sites

Kinda-sorta.

 

"Pushing" is amplifying or multiplying or "cranking up the gain of" whatever base image is there. You start with a captured brightness range of between 0 and x, and if x is still too dark, you can multiply the whole range of brightnesses in the picture by, say, 4 (the "push"), and get a brightness range of between 0 to 4x (any number times zero is still zero, of course).

 

Result - a steeper line connecting 0 and the maximum (higher "gamma") = more contrast, plus a stronger maximum and mid-range - a picture you can see.

 

Thus all of the following do roughly similar things (assuming, in the case of a digital capture, that the sensor is "ISO-invariant"):

 

- underexpose a sensor by setting the camera to a higher ISO than base, and letting the camera "push" or multiply the image values for you

- underexpose a sensor by leaving the camera ISO setting at base ISO, but using a hand-held metered exposure (or exposure-compensation dial) to feed it too little light, and then doing the amplification in LR or Photoshop or whatever.

- underexpose film, and amplify the silver in it by extended development (pushing it chemically)

- underexpose film, and give it normal development, and then scan the film, and use LR to "crank up the gain" just like any other digital image.

 

The main obvious difference will be this: chemical "pushing" tends to increase the "grain/noise" in the highlights, by creating more and more silver in the only place there is any exposure to develop. Digital pushing tends to increase the "grain/noise" in the shadows, because that is where the signal/noise ratio is worst. *

 

pushdigi.jpg

Pushed film detail -                                                             Pushed (high ISO - 16000) digital -

clean shadows, noisy bright tones                                    noisy shadows, smoother highlights

 

Digital pushing is rather like recording sound with a weak mic or preamp, and then trying to recover the original sound by cranking the primary amp or playback volume up to 11 - you can hear the song, but also every tiny pop and scratchy bit of static noise.

 

There may be other differences, just due to the fact that the whole "film" thing tends to be non-linear, especially at the extremes of exposure (toe and shoulder of the film's response to light = "characteristic curve" ** ), and in development (where to compensate for a whole stop of underexposure (half the needed light) you give 1.4x (root-2) the development. You may find that playing with a "curve" in LR better replicates chemical pushing than an overall exposure bump of "+1". Or some combination of curves, exposure, contrast, shadows or highlights, etc.

_________________

 

* the one thing that frustrates me about digital - after 40 years of seeing pushed film pictures with (usually) grainy highlights and smooth empty shadows, the noisy shadows and "smoother" highlights of high-ISO digital just look - wrong - to me.

 

** http://www.sprawls.org/ppmi2/FILMCON/#The%20Characteristic%20Curve

 

Thanks for detailed reply. I will have to think about your following comment that is very fascinating:

...The main obvious difference will be this: chemical "pushing" tends to increase the "grain/noise" in the highlights, by creating more and more silver in the only place there is any exposure to develop. Digital pushing tends to increase the "grain/noise" in the shadows, because that is where the signal/noise ratio is worst....

Share this post


Link to post
Share on other sites

BTW - how did you meter your Jag picture?

I did sunny16 (f/16 and 1/500), Tri-X 400 lab developed. I am just wondering whether 1/500 sec is not that accurate or I might have made a mistake and set speed higher (1/1000). Later is a possibility since in the same roll there is another sunny16 shot in similar sun that looks ok. I shot from running car and might not have set exposure correctly.

Share this post


Link to post
Share on other sites

One stop would not make that big a difference. Something else is going on.

Share this post


Link to post
Share on other sites

One stop would not make that big a difference. Something else is going on.

Any guess? I have used Sunny16 many times after this and it has been fine (even in the same roll). My own guess is mistake in exposure setting (most probably) or inaccurate shutter (less probable).

Share this post


Link to post
Share on other sites

1) side-lighting: need to add ~1 stop for that (with side lighting, on average, half your subject is front-lit and half your subject is backlit, just like a half-moon)

 

2) tinted car windshield - need to add a "filter factor" for that. Anywhere between 2/3rds-stop for basic faintly green safety glass (blocks 35% of the visible light), to several stops with a serious dark tint (factory tinting blocks ~75-80% of the visible light).

 

https://oceanoptics.com/transmission-characteristics-sunglasses-tinted-windows/#foobox-5/0/Figure-5-UV-Vis-Transmission-in-Car-Window-Tinting.jpg

 

3) plus, of course, under sunny 16, Tri-X nominally needs 1/400th sec and f/16 - 1/500th is technically a tad under to begin with. I usually cheat to the half-stop between 11 and 16 ("f/13") with 400 film and 1/500th sec.

 

OTOH, if a film Leica shutter (except electronic M7) is off-spec, it is usually too slow, and gives back the needed half-stop. Typically, unless very recently serviced, a Leica M clockwork shutter at "1/1000th" is running about 1/750th.

 

But it all adds up.

Edited by adan

Share this post


Link to post
Share on other sites

When I push film (usually from 400 to 1600) I do it also for other 2 reasons, besides the extra flexibility of the higher iso:

1 - I want a certain look with more contrast (and of course I embrace the grain)

2 - the extra 2 stops gives me the possibility to use a smaller aperture and get more depth of field in worse light conditions

 

You might get the constrast in PP, but not the DOF.

 

Well, after a second thought, you could use anyway a small aperture, shoot underexposed, don't push and correct in PP (which is what you want to do) but you'll get a lot of extra digital noise/grain in the shadows if it's more than 1 stop. That's my experience, at least. 

 

PS I usually cheat with Sunny 16 as well: iso 400 -> 1/500 F11

Edited by AntonioF

Share this post


Link to post
Share on other sites

Anyway - here is your first jpg, revised with a PhotoShop "Curves" adjustment. The blockiness in the shadows is from the jpg compression of your post; you shouldn't see that, working from your original scan file.

 

At this scale, of course, any noise or grain is mostly too small to see. And the exact curve drawn can vary according to taste - you don't have to like mine.

 

Share this post


Link to post
Share on other sites

Anyway - here is your first jpg, revised with a PhotoShop "Curves" adjustment. The blockiness in the shadows is from the jpg compression of your post; you shouldn't see that, working from your original scan file.

 

At this scale, of course, any noise or grain is mostly too small to see. And the exact curve drawn can vary according to taste - you don't have to like mine.

 

revisedarkscan.jpg

Thanks Andy, That looks much better than simply using exposure slider.

 

I tried to mimic the curve in LR and results came out similar. I had to add 0.5 exposure as well. See the settings. The grain (crop) doesn't look bad either. I guess now I have a tool to recover from underexposure.

 

 

Crop (licence plate masked)

Edited by jmahto

Share this post


Link to post
Share on other sites

You could have had one for a half million...

 

https://www.hemmings.com/blog/2013/11/22/series-i-jaguar-e-type-sets-auction-record-at-new-york-sale/

 

Or a lot cheaper; it was an available color for some variations.

 

The one in the OPs picture was a +2 model, no?

 

Half million? Can we get snow tires for it?

Share this post


Link to post
Share on other sites

Thread drift - I thought one only saw "Jaggies" in digital photos.

 

Jayant, you need a higher-megapixel film.

 

https://en.wikipedia.org/wiki/Jaggies

Where did you see "jaggies" in my scan? I might be missing something.

 

When you said high megapixel film, you mean TMax100 instead of Tri-X?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×