What's new

Does the reciprocal focal length rule not really apply for super wides?

fjrabon

Been spending a lot of time on here!
Joined
Nov 3, 2011
Messages
3,644
Reaction score
757
Location
Atlanta, GA, USA
Can others edit my Photos
Photos OK to edit
So I hand shot a few today right after sunset. I was shooting at 10mm. Many of my shots were with shutter speeds in the range of 1/10 or 1/15. Yet a lot came out soft, much softer than when I normally shoot my 35mm at 1/35, for instance. It perhaps could have been that I was shivering without realizing it because of the cold, but I didn't think I was.

I know the reciprocal focal length rule is more just a guideline, but does anybody know if the rule of thumb is even further off for super wides?
 
Most people can not handhold below 1/60 regardless of Focal length. Some people that know ho to take a good photographer stance and hold their breath can do more. but the average person cannot.
Also the farther you are from your subject the faster you need your shuuter speed to be. Think oif it this way, You have a 200mm lens and at the end of the lens your movment is 1/4" up and down. At 8 feet that is 2" of movment, at 100' that 25 inches of movement
 
I can usually shoot down to 1/4 to 1/8 second and get acceptable results at very short focal length settings on "most" frames. Some will be blurred, most will be "acceptable". If you want really super-sharp images that can stand large image blow-ups, a tripod is supremely useful at slower shutter speeds. Breath control and practice helps. I used to shoot a lot of pistol and also some air pistol--breath control and trigger control, sight picture, body mechanics fundamentals--it all comes together.

Most of the reason one can get away with slow shutter speeds with W/A lenses is due to the low image magnification...low mag hides blur by virtue of the blurred objects' small size...
 
1/10th or 1/15th is too slow a shutter speed regardless of focal length unless your subject is being frozen by flash. I would say you are at the extreme end of the 'one over' rule.

While it can be done, most photographers will experience very inconsistent results. 1/10th of a second is a tripod territory, regardless of lens used.
 
It can be done. But you should realise that 36mm = 50 degrees field of view doesn't mean 18mm = 100 degrees field of view, instead it is 90 degrees. So, the lower your focal length, the less accurate the reciprocal focal length rule will be. I have a good formula to calculate shutter speed needed handheld here -

Shutter speed = (angle of view/circle of confusion/50) second

Circle of confusion - micron
Angle of view - degrees
 
It also depends on how far the subject is. If you use wide and you are really close, a tiny movement will be a very visible motion blur. Not when you shoot subject farther away.
 
I've never found that reciprocal rule to be terribly accurate, there are many more variables involved in determining how much camera shake is too much beside just focal length. I pretty much never like to shoot below 1/40, no matter the focal length. You can make it work, but it's a close call... from 35mm up to like 100mm the rule is sort of half decent, if you take into account any crop factor, and if your subject is at a middle distance. Once the lens gets long the rule stops working again, subject distance becomes more important. Using mirror lockup, or shooting in live view in one of the silent modes, will reduce some of the in camera vibration and let you get away with a slower speed. Too many variables to make rules. Best advice is to chimp if you're not sure, and then adjust.
 

DSC_0038 by franklinrabon, on Flickr

here's an example of what I was talking about. Don't rail on it too much :) , it was purely just me practicing with my CPL, trying to get a feel for how it reacted to different light. (which darkened things, thus why my shutter speeds started getting long). I was also trying to get more used to the effect of varying apertures on my super wide in a way that I could more easily visualize. It's not awful, I don't think, but definitely not a sharpness level I'd deem acceptable for a 'real' shot.
 
In that shot, 1/10 is actually slower than the reciprocal, since you have a crop sensor. The rule would put 1/15 at the lower limit. Additionally, the D3100 is not known for outstanding ISO performance, so even at 800, noise (or noise reduction) will be affecting your ultimate sharpness. On top of that, that Sigma lens is a tradeoff design, where sharpness is weighed against other factors. It's big brother the f/3.5 does much better in that department. In the leaves of the trees I'm also seeing some digital loss of detail... I'm not sure if that's noise reduction, or jpg compression. Did you shoot these raw? jpg? manual noise reduction? auto? final jpg compression level? Sharpness is easy to fix when it's grossly off, when you're getting closer and closer though, the variables become more subtle, as there can be many weak links in the chain.
 
At 1/10sec even very light wind is going to make all those leaves highly likely to blur from their own slight motions; same for the grass as well.

And yeah as said above there are many variables around the best handheld shutter speed and they'll all vary between different people. Some people are naturally very still, whilst others are not - in addition you've your condition at the time (eh are you shaking from cold or have you just walked a good few miles and exhausted yourself - did you just have a big meal and are well rested - have you been getting enough sleep etc...).

Also the rule relies upon your posture and holding stance:


if you've got a good posture and holding style as well as good physical condition you'll do far better than if you're tired and holding the camera wrong.

Also don't be afraid of natural supports - leaning on a tree, bench or wall will greatly help .

EDIT - just a point about the end of that video and the "overhand" grip; its not valid for handholding (as explained) but it is a method used by and valid when using a long lens (300mm+) on a tripod or monopod for adding stability and you'll likely see some sports/wildlife photographers using that method (but not when handholding - heck if you can overhand grip a 300mm f2.8 you've got massive hands!)
 
Last edited by a moderator:
In that shot, 1/10 is actually slower than the reciprocal, since you have a crop sensor. The rule would put 1/15 at the lower limit. Additionally, the D3100 is not known for outstanding ISO performance, so even at 800, noise (or noise reduction) will be affecting your ultimate sharpness. On top of that, that Sigma lens is a tradeoff design, where sharpness is weighed against other factors. It's big brother the f/3.5 does much better in that department. In the leaves of the trees I'm also seeing some digital loss of detail... I'm not sure if that's noise reduction, or jpg compression. Did you shoot these raw? jpg? manual noise reduction? auto? final jpg compression level? Sharpness is easy to fix when it's grossly off, when you're getting closer and closer though, the variables become more subtle, as there can be many weak links in the chain.

the leaves are, I think, the flickr to TPF bug that a lot of people have been talking about here, ie two layers of jpeg compression from the original image. They were shot in RAW. There was some noise reduction, but not a lot.

Yeah, I realize the sigma 10mm is a little soft around the edges, but I think the main thing I noticed was a good bit more motion blur than I expected. I also didn't realize that the reciprocal focus rule was quoted for full frame sensors. I think most of the times I've seen people quote it here they were referring to crop sensors. Also, I think the noise was mostly because I was trying to get as much of the sky around the sunset from blowing out, and thus it was probably a little underexposed. They were test shots so I could see how the CPL affected skies at varying angles from the sun, so I was mostly concerned, at the time, with how the sky looked.

It was after going over them in post and seeing how much camera shake softness there was that I thought about the shutter speed being too slow for handheld there.
 
At 1/10sec even very light wind is going to make all those leaves highly likely to blur from their own slight motions; same for the grass as well.

And yeah as said above there are many variables around the best handheld shutter speed and they'll all vary between different people. Some people are naturally very still, whilst others are not - in addition you've your condition at the time (eh are you shaking from cold or have you just walked a good few miles and exhausted yourself - did you just have a big meal and are well rested - have you been getting enough sleep etc...).

Also the rule relies upon your posture and holding stance:


if you've got a good posture and holding style as well as good physical condition you'll do far better than if you're tired and holding the camera wrong.

Also don't be afraid of natural supports - leaning on a tree, bench or wall will greatly help .

EDIT - just a point about the end of that video and the "overhand" grip; its not valid for handholding (as explained) but it is a method used by and valid when using a long lens (300mm+) on a tripod or monopod for adding stability and you'll likely see some sports/wildlife photographers using that method (but not when handholding - heck if you can overhand grip a 300mm f2.8 you've got massive hands!)


Thanks for the link, I've watched a lot of Joe's videos and read a few of his books, Joe is the man. I think I'm usually pretty good, as I can usually get reciprocal focal length shots sharp even on a crop sensor. I just didn't realize that 1/10th of a second is pretty much impossible to be super sharp with regardless of focal length.

And I've tried to get 'da grip', as he talked about it in a few of his books as well. But a D3100 body just isn't deep enough to get it into that meaty section between your shoulder and collar bone. I can get it on my shoulder, and that's often how I shoot, but shoulder isn't as steady since any movement in that joint will be amplified.
 
Last edited by a moderator:

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

New Topics

Back
Top Bottom