Largest File = Best Shot

[Colin500]

Hi,


instead of taking a single shot of any subject, I often take several, in order to choose the one with the best focus/sharpness later at home. And I'm sure that most everybody here does the same thing.


In theory, when dealing with compressed images, with everything else being the same, the more high-frequency detail there is in the signal, here the image, the larger the file is.


In my experience this works so well that I often just keep the largest file from a sequence of identical shots, without even bothering with pixel-peeping and 200% magnifications etc., since the large file size indicates more detail, which comes from best focus and sharpness.


Of course this method won't always work, and can be put off by some overly-detailed background, but still, I was wondering whether anybody (or everybody) else was using this method to quickly find the best image in a sequence...?


Ciao, Colin

[canon-ball]

i've never thought of going about it that way... gonna have to check that out and get back to ya

[Justin T]

At one time, when I was revamping my work flow, I gave this method a try by using a blind test by selecting the sharpest image, marking it then comparing it to the file sizes. I personal found that it did work IF everything between the group of photographs was the same. I found that if ISO,shutter, speed, aperture, framing, light levels, or subject distance varied that it would not work so well. Another place that I found this to brake down was with macro, selective focus, or photographs where not everything is in focus. Also, it only worked on jpeg compressed images. Loss less compression(some RAW formats) images I found no apparent correlation between file size and best sharpness. In fact I found most/some of the time it was the opposite.


In the end, I did not adapt this method because it was not reliable for me. However, your mileage may very.

[mikehillman89]

I guess this could work theoretically but it would be very hit or miss in my opinion. I still prefer to dump them and spend quality time doing some quality control at a later date if I have to.

[Sheiky]

Hmm tricky...I wouldn't take my chances on this technique. With portraits I would never use it and if your composition has a higher dynamic range it wouldn't always mean the more signal the better (for instance silhouette shots or landscapes, perhaps you select a photo wityh a lot of detail in the foreground while actually you want detail in the background etc etc). Furthermore if settings are the same, it could be done, but I like more safety when sorting my photos. To be honest it doesn't take to much time downloading them all to a pc and loading them into Lightroom or whatever program you use and deleting the bad ones.


And if settings aren't the same same, forget about this technique. Higher ISO also means bigger files and a higher ISO isn't always preferable.


Anyway I like taking a little more time (only a little) and be safe and in control, but if it works for you be my guest [H]


Jan

[neuroanatomist]

Makes perfect sense. I do that with my mail, too. I keep and open the largest/heaviest one, and the smaller/lighter ones I just throw away without looking at them. Someday I'll figure out why those pesky bill-collectors keep calling me.


[:P]


Your largest and 'best' portrait, with the best focus, most sharpness, and most detail, might just be the one where the subject blinked. Tack-sharp eyelashes on closed eyes. Hmmmm.....


Sorry for the hyperbole, but personally, when I want to find the best image from a sequence, for me that process involves actually looking at the images.


[:)]


--John

[Colin500]

Hm, it seems that I might have evoked a bit of a wrong impression. It's not that I'm using an "ls -l" to choose my keepers...


...but when I'm in Aperture looking at e.g. a sequence of close-ups of a flower taken in rapid succession, where just a huff of wind can throw off the focus, I'll use the file size as an indicator on which one to keep.


Cheers, Colin

[Sheiky]

That would make sense Colin, but then your technique would only be used in such specific situations. And the way I understood it: you use the technique for more general situations as well. But for such specific situations it doesn't really matter what technique you use I think. It could work out perfectly, but for the majority of sorting it could go terribly wrong...