Final Frame – End of the Roll

From my reading about photography and photographers over the years I’ve accumulated a feeling that many a shoot has been saved by the telling image made on the last frame (not to be confused with a ‘film end’).  It’s probably a false impression, but it remains.  Googling something like ‘last frame’ or ‘end of the roll’ provides no evidence one way or the other, instead you get taken into the realms of film buffery and carpet retailing. And I did discover: Sami Matarante (aka Sami Jones), who has compared first and last frames (www.thesamijones.com/blog); and an Instagram account dedicated to the first frame of a 35mm roll, f1rstoftheroll.

Perhaps there are examples of photograph gold in the last frame, but I’ve probably turned a couple of examples into a generality without any evidence to back it up.  It isn’t without a certain logic, however.  As you use a roll of film on a single subject or event, you can build confidence, get to know the subject and refine what you are trying to achieve, so the last 10 frames might be better than the first.

I decided to look for evidence in my recent walking project (now called Footfall), in which I made a final cut of 85 images from nine 35mm rolls.  Pictures were selected from all but four of the 36 frame numbers, giving an average of 2.5 pictures per frame.  Four images each came from frames 6, 8, 12, 17 and 36.  Amalgamating the numbers in four groups gives the following picture: frames 1-9, 20 images; frames 10-18, 17; frames 19-27, 19; and frames 28-36, 26.  Taking a doubly selective tack, a selection from a selection, my favourite four images are from frames 3, 20, 26 and 34.  There’s not enough here to support the hypotheses of film end bias.  The bias is in my mind.

Footfall 30-31 (Frame 34)
This entry was posted in Uncategorized and tagged . Bookmark the permalink.

Leave a Reply

Your email address will not be published.

Please do this simple sum to prove you are real! *

This site uses Akismet to reduce spam. Learn how your comment data is processed.