Kodak 400TX at EI 12800
Cropped from 6x6 (120 film)
Kodak HC-110 Dilution B
140 minutes at 68 Fahrenheit

  2016 March 25  /  Updated 2016 May 4    Film   Developing


It's fun to push film.  In theory, you could turn a 400 film into an effectively-6400 film, or more.  This allows you to take photographs in very low light.  

To push film, you set the camera at a higher ISO (technically, it's EI, not ISO).  Then you develop the film as if it were that speed. 

Let's push some Kodak 400TX film.

A Quick Note

Articles like this one are possible only with the support of readers like you, when you use the links on here to purchase your gear. 

Your help is greatly appreciated and helps keep this website on-line.  Thank you to everyone who helps make this website what it is.

In This Article

Basic Idea


Developing Times, Mark I

Developing Times, Mark II

Simplified Curve


Basic Idea

Push processing means underexposing the film (relative to the box speed), then over-developing it.  For Kodak Tri-X the box speed is 400.   Same thing with Ilford HP5 Plus;  it's a 400 film, but you can push it to 3200 or even higher.

Any of the major B&W "400" films can be pushed to 12800 or perhaps even 25600, depending on lighting conditions and the developer you choose.  None of the film companies are going to recommend this.  They would prefer that you see their films at their best, which is understandable.  400 films at 400 or 800 have fantastic range and detail; at 1600, they still look great.  Above 1600, the shadow details start to go away, fast.  Depending on what you're trying to achieve, that might be OK.

If you want to shoot the film as 6400, that's four stops under-exposed, relative to a box speed of 400.  To complete the equation, you would then process the film as if it were 6400: four stops past the box speed.

If you have a camera that offers ISO/ASA setting of 6400, just set it there.  When you shoot photos, meter at that setting.  Then, develop according to a recipe for 6400 film speed. 

Back to Top

Table of Contents


Pushing film to 6400 and beyond requires realistic expectations.

First, poorly-lit scenes offer a very low signal-to-noise ratio.  There aren't that many photons falling on the sensor (film or digital) in a short time interval.  With highly-pushed films, noise (or I should say grain) can really start to overwhelm the scene.

Highly-pushed films are not picking up true speed increases.  The setting you choose on the ISO dial is not the real ISO of the film;  it's actually the E.I. (exposure index).  A 400 film pushed to 6400 is not going to have the full range of shadow detail.  It might produce a very worthwhile image, though.

35mm Kodak 400TX at probably 6400 but developed as 3200
HC-110 Dilution B @ 68F, 27 minutes (should have been 45-60 minutes).

The gritty, contrasty "noir" look goes well with night photography.  It's an artistic style that brings with it an important benefit: higher shutter speeds.  Actually, though, if you develop it right, the pictures won't be so gritty and contrasty.  Look at the photo at the top of this page;  that's 12800, and it looks like 400. 

If you're taking pictures where anyone is moving, you'll have to use shutter speeds of probably 1/30th or faster.   (Preferably faster, or they'll blur.)  At night, you need a very fast film to do that. 

Because an SLR has mirror-slap, it also does best with shutter speeds of 1/30th or higher.  (Get yourself a Leica M3 and and you can shoot handheld at 1/8th if you're steady.  A lens such as this one will allow for faster shutter speeds.)

Next up, some tentative data for pushing Kodak Tri-X.


Back to Top

Table of Contents

Developing Times, Mk. I

How long should you develop 400TX shot at EI 12800?  How about EI 25600?  I found there was no easy answer.  Kodak publications don't even recommend pushing beyond 1600. 

Out on the 'net I was able to find only one dev time for Tri-X at 6400, Dilution B, and 68 Fahrenheit.  It says 26 minutes.

It was tougher to find 400TX @ 3200 in Dilution B.  The times I was able to find were 19 to 22 minutes, and I can't even remember where that was.

I graphed out some known and sort-of-known times and obtained a best-fit equation. The result is a power equation of the form y = bx^a, where a and b are some real-number constants.  I've seen this type of curve many times before with chemical reactions.  (Does it fit this one?  I'm starting to think not.)

The dev times on this one do not increase in a linear fashion.  As we'll see, this might not really be what's happening;  that's why this was the "Mark 1" graph.

Problem is, I think some of the data points are wrong, especially for 3200 and 6400.  Next up, "Dev Times, Mark II".

Back to Top

Table of Contents

Developing Times, Mk. II

Looking back at rolls that were rated at 6400 and 12800, I noticed that most of the pictures were a whole stop under-developed or under-exposed.  Some were just light-meter failures, because there wasn't enough light.  In some of them, though, I knew the meter was working.  These were still one or two stops under.  What was happening?

A few scenes looked exactly right;  later I realized they were actually shot at least one stop lower EI than I'd originally thought.  If you were supposed to shoot f/8 and 1/125th for EI 12800, but you actually used f/8 and 1/60th, the real EI was 6400.  So, at first when I'd thought I was shooting at 12800, it was really 6400. 

Preliminary results suggest the dev time for 6400 should be at least 42 to 45 minutes with Dilution B.  Later I revised that upward to 60 minutes.

I also got thinking about the best-fit equation.  Why would the time increases decrease?  A power equation is correct for some chemical reactions, but this particular graph should either be linear, or increasing at an increasing rate.  Exponential was the right direction but didn't fit the data at all;  logarithmic was the wrong direction (increasing at a decreasing rate, like the power equation).  So how about this, instead:

This chart makes more sense.  26 minutes was giving under-developed negatives at 6400.  42 to 45 minutes yielded much better results.  Better yet, go with a full 60 minutes. 

For 12800, the develop time would be over two hours (140 minutes) according to the curve.  I tried this and it seemed to work very well.  This time the photos were really shot at 12800, because this time around I metered carefully.

Here's one reason why I think the upswept curve is reasonable.  When you start getting into dev times longer than about an hour, it gets to be a chore to keep agitating the film every five minutes or whatever you were doing.  Thus, as dev times increase, the development increasingly has to resemble full-stand development (i.e., no agitation).  Are you really going to be there every five minutes to agitate the film over an eight- or nine-hour span? 

See also "Agitation", below.

Back to Top

Table of Contents

Simplified Curve

The Mk.II chart has a third-order polynomial equation.  That's a little bit tedious;  you're probably not going to want to sit there and do a cubic equation every time.  It just so happens that the curve is very linear through 6400, and not-really-that-far-from linear at 12800.  So, here's a much simpler graph and a much easier equation:

Be sure you have enough total HC-110 in the solution that it doesn't deplete halfway through the process.  For a roll of 135-36 you'd want there to be about 6 ml of concentrate in the solution.  True push-processing should not require more total developer.  If you're using enough Dilution B to cover one roll in a Paterson tank, you'd have about 11 ml of HC-110 there anyway;  so it's kind of a moot point.

Give these times a try and see if they work out;  I haven't tried 25600 yet.  Don't forget to use de-gassed distilled water for your developer and fixer.  Temperature control is also crucial.  Try to keep the developer at 68 F throughout the whole dev time.  Higher temps will elevate fog and grain.

Back to Top

Table of Contents


To keep the photos from being too gritty and contrasty, the dev times are long and use low agitation.  Start the developing with 30 seconds of gentle agitation, followed by ten seconds every five minutes.  With the Paterson tank, that's about four slow back-and-forth rotation cycles. 

For 12800 it was 140 minutes of develop time, with the ten seconds of slow rotary agitation every five minutes.  As you can see at the top of the page and here, it worked well.

Could you develop the film faster, using a higher temperature or with more agitation?  Yes, but I think it would increase grain and decrease the tonal range of the negatives.  Dilution B can act as a compensating developer if you don't agitate too much.  This is what you want for extreme push-processing. 

With long develop times (2 hours or more) you might be able to do the agitation every ten minutes instead of five.  For 25600, the projected dev time is over nine hours, so there you could probably agitate every hour or perhaps even every couple hours.  Another idea would be to maintain the five-minute agitation intervals for the first hour, then go with perhaps ten or twenty minute intervals, then more.  Whatever you do, keep track of the results so that you can duplicate the successful ones.

Back to Top

Table of Contents


Pushing Kodak Tri-X to 6400 and beyond is more than just possible;  it can work very well.  (See this gallery for more examples.)  We've looked at dev times for Kodak HC-110 Dilution B, one of the most useful developers for pushing film.  One reason I like this developer is that it's easy to make high dilutions, and the concentrate has very good shelf life. 

I'm finding that push-processing above 1600 requires longer dev times than you'll probably see listed elsewhere.  It took me a lot of work to formulate these times and get these results.  I had to do this on my own, because if I listened to "everyone", they would have said don't bother, you're wasting your time, the shadows will block up and fall out of your computer monitor and land on your toes, etc., etc. 

Here are some times so far;  I might revise these, but they are making sense based on the results I've been getting:

3200.........32 minutes
6400.........60 minutes (1 hour)
12800........120 to 140 minutes (2 hours to 2:20)
25600........560 minutes (9 hours and 20 minutes)

For 3200 and 6400, agitate slowly for ten seconds every five minutes.  For 12800, try ten seconds every ten minutes, although I went with five minutes and had great success.  For 25600, you'll have to decide.  I would try mild agitation every hour, though, if possible.

From what I've seen, Ilford HP5+ might have a one-stop advantage over Tri-X in HC-110 B.  That means the develop times for pushed Ilford HP5 are shorter.  I first noticed a hint of this in the published times for EI 1600;  the HP5+ time is shorter by several minutes.  I'm going to look again at the HP5+ dev curve and see if it might work with a 3rd order polynomial.

I've tried to provide specialized information that you might not even find in a book.  Hopefully this article was helpful to you.  Please help me out by shopping for your stuff through the links on here.  (If you're outside the USA and you can't get Amazon stuff shipped to you, I put some Ebay links on here.)  It's the only way I can keep this site on-line.  Your help is greatly appreciated.

Thanks for visiting my website!

Contact me:

3 p o.t o . 1 2 0 s t u d i o.. c o m

This won't directly copy and paste.  Please manually type it into your mail program.
No spaces between letters.

Home Page

Site Map

What's New!


Copyright 2010-2016

Back to Top of Page