HP5+ Experiments:Is Box Speed Best?

I ventured out today with Mollie after a hectic week of drinking, eating, and celebrating with friends and family over the holidays. It was all very enjoyable but I think both of us needed to get out in the woods for a while. For her part Mollie had a great time in Holme Fen. It rained so hard this morning when we got up she declined our daily morning walk and so by noon was ready for a romp. She quickly got up to her old tricks of jumping, spyhopping, exploring and chasing about the bracken and fallen logs.  

Blog Post Cover Image

For me this was more about some experimentation which I will get to shortly. The day was not really made for photography and the light incredibly variable with almost clear skies when we arrived to overcast to some clearing and intermittent clouds during the 3 hours or so we were out. 

I wanted to have a good demonstration of what box speed on a film means and how this may not always be the best way to shoot film. This came up as a question on Reddit and I thought a more studied look would illustrate it best. I also wanted to illustrate what happens when film is pushed as there is too much assumptions this is some kind of secret or miracle. 

At the heart of this is a useful rule-of-thumb that you should expose film at half box speed and develop at 70% the recommended time of the manufacturer (80% for t-grain films).  The objective is to give good exposure for the shadows and develop so the highlights are not blown. 

Manufacturers rate film speed for an average user and there is also some spec creep in the race to show faster film speeds to be more competitive. People who do rigorous sensitometry discover that highlights and especially shadow detail can get lost and the negative may be difficult to print in the darkroom. (I did this exercise for FP4+ here...). There is is a whole other area of expansion and contraction during development to compensate for high or low contrast scenes. I will not cover that here. 

Some of the theme is to illustrate a flatter well exposed negative is easier to manipulate in the darkroom or digitally. It goes with Ansel Adams' idea that the negative is the score and the print is the performance. 

The Experiment

The experiment I devised was simple. Shoot HP5+ at half box speed and develop at 70% of Ilford's recommended time. This is what I have opted to do with HP5+ and have seen good results. The next step is to make the same image at box speed at the manufacturer's development time. Finally a third version of the image pushed one stop to ISO 800 and developed accordingly.  

These images were taken on my Intrepid 4x5 camera on sheet film so I can vary development for each version.

Developer

I use HC-110 developer and for 4x5 I use dilution H (1+63) which is half of dilution B. My reduced (-30%) development times mean I wanted to boost the time time from dilution B. Normal time for dilution B according to Ilford is 5 minutes at 20C. 70% of this makes for short times and makes the development more subject to variations in the process. Dilution H times are twice dilution B development times so 10 minutes or 7 minutes at 70%.  

I develop in Cibachrome drums. I use a motor unit to agitate which I modified to reverse direction so I in essence have constant agitation.

Taking the Pictures

I found a simple subject and set it up with the camera. The scene was not high contrast but should show the principles. (The sun failed me at this time of the day.)   I spot metered and found the shadows to be in the range of about Ev 8 and highlights at Ev 12. In each case I placed the shadow (Ev 8) in zone III and this meant the highlights were in zone VII. The shadows were under the shelf fungus for example and the highlights were easiest found on the birch bark. 

The light was consistent across the images so the exposures are straight-forward in decreasing shutter speed with each increase EI/ISO metering. 

• Image 1 ISO 200 f22 1/4 second
• Image 2 ISO 400 f22 1/8th second
• Image 3 ISO 800 f22 1/15th second

Development

Development times were based on Ilford times which are HC-110 dilution B 5 min for ISO 400 and 7 1/2 min for ISO 800 at 20C. Dilution H is twice the dilution B times. As above my ISO 200 was 70% of these numbers.  My developer temperature was 21C so I used Massive Dev Chart's temperature compensation function to calculate actual times. 

• Image 1 ISO 200 HC-110 dilution H at 70% recommended time (8:00 at 21C)
• Image 2 ISO 400 HC-110 dilution H at 100% (10:08 21C)
• Image 3 ISO 800 HC-110 dilution H (13:49  21C) 

Scanning

The scans were done in one session using an Epson V850 Pro scanner. I used Vuescan software and opted for the 'RAW' tiff file format that gives a flat linear scan. The scans were 3200 dpi at 48 bits per pxiel. I enabled multi-exposure which gives two passes one normal and one to give more exposure time to improve scanning of high density areas. I enabled 16 samples which samples each pixel 16 times and averages them. This helps reduce some noise in high density areas. 

The result is that each scan should be comparable as the scanner is not compensating for each different film.

After scanning I converted to grayscale in Photoshop elements. I then set the back point to just below the point the first pixels show clamping to zero. I did not set white point so the scans do not become distorted in terms of dynamic range. 

Results

• Image 1 RAW0001 ISO 200
• Image 2 RAW0002 ISO 400
• Image3 RAW0003 ISO 800

The list above gives scan file names to the different EI/ISO versions. 

The first figure is the three histograms in numerical order left to right. This gives a thumbnail view of the effects of exposure and development. 

ISO 200, ISO 400, ISO 800

The first thing to note is the increased slopes on the left of the histograms as the film is pushed faster. This represents a loss of shadow detail. All the shadows in the right two histograms are pushed into a narrow band. You can also see a greater loss of highlight in the ISO 800 case. This might indicate more development time would help. 

Histograms with equal values in shadows and highlights. 

 Finally we can see some examples of crops for each case. The first shows the shelf fungus and shows where the zone III shadows lose detail has the films gets faster. This gives an apparent increase in contrast which is indeed what people use pushed film to achieve. 

ISO 200, ISO 400, ISO 800

The birch comparison shows highlights as well as shadows in the bark. 

ISO 200, ISO 400, ISO 800

Again loss of shadow detail.

If one wishes now to make an image like the ISO 800 version then a little curve work on the ISO 200 version makes a reasonable semblance. Below I present a manipulated ISO 200 version with the 400 and 800 versions untouched. 

ISO 200 (manipulated), ISO 400, ISO 800

Here we get a higher contrast version. The difference is the negative keeps more image detail and we can keep or hide as much as we wish. 

This image was fairly low contrast as the scene encompassed about 4 stops of range. If we were forced to push shadows to zone II and highlights to zone VII then the differences would have been more apparent. 

The three full images are below scaled to 600dpi. You can find the full sized scans here... 

ISO 200

ISO 400

ISO 800