Focal length= Shutter speed to avoid camera shake. Crop and full frame cams..

[Nikon_Josh]

Now.. I am posting this in the beginners forum because I tihnk it could be relevant to alot of beginners. Here is the question I have for you...I often hear people saying for example, that on a Crop camera you need to take into consideration the 'crop factor' when calculating shutter speeds to avoid camera shake.I feel this may be false information?

So anyway to give an example, some would tell you that with a 35 1.8G lens you would need a shutter speed of at least 1/60 to avoid camera shake, as the 35mm 1.8 is equivalent to a 52mm lens on a DX Nikon camera body.

As we know, the focal length will always stay the same regardless of the crop factor. A 300mm lens stays a 300mm lens despite the crop factor which makes it have an angle of view of a 450mm lens. So surely if you shoot a 300mm lens on a DX body you will only need to keep shutter speeds as fast as 1/300 (not 1/450) of a second to avoid image blur?

If someone clever can clarify this for me, I would appreciate it. I often see people giving what I feel is misinformation on this subject.

 

[480sparky]

300mm is 300mm, regardless of whether it's on a full-frame or crop-frame camera. Any camera shake will be identical.

 

[Nikon_Josh]

300mm is 300mm, regardless of whether it's on a full-frame or crop-frame camera. Any camera shake will be identical.

Thank you! I thought as much, I just wanted confirmation on my initial thoughts on the matter. You see so many people giving the advice that people should use shutter speeds that match the crop factor, which clearly is mis information.

 

[o hey tyler]

For the same reason the Nikon D800 came with a primer guide to handling such a high resolution sensor, you have to take into consideration the magnification of the image at which you're viewing.

The D800 has a very high resolution sensor, so proper camera holding technique and shutter speed acquisition matter very much to reduce even the finest camera shake blur. I notice blur on my 5D2 that I wouldn't have noticed on my 5D1.

So when you're using a 300mm lens and you have an effective FoV of a 450mm lens, you're going to want to account for that because the end result will be a relatively high resolution photo (between 12-18mp like most current DSLRs) that will show camera shake due to a slow shutter speed of 1/300. I don't doubt the ability of someone to get a good handheld shot with a 300mm lens on an APS-C body with 1/300s, but they'll need to have a steady hand.

IMO it's always good to have a buffer anyway, unless you NEED to be at the minimum focal length/shutter speed rule.

 

[Nikon_Josh]

For the same reason the Nikon D800 came with a primer guide to handling such a high resolution sensor, you have to take into consideration the magnification of the image at which you're viewing.

The D800 has a very high resolution sensor, so proper camera holding technique and shutter speed acquisition matter very much to reduce even the finest camera shake blur. I notice blur on my 5D2 that I wouldn't have noticed on my 5D1.

So when you're using a 300mm lens and you have an effective FoV of a 450mm lens, you're going to want to account for that because the end result will be a relatively high resolution photo (between 12-18mp like most current DSLRs) that will show camera shake due to a slow shutter speed of 1/300. I don't doubt the ability of someone to get a good handheld shot with a 300mm lens on an APS-C body with 1/300s, but they'll need to have a steady hand.

IMO it's always good to have a buffer anyway, unless you NEED to be at the minimum focal length/shutter speed rule.

There was me thinking, the question was already answered. But you have added another dimension here to to the subject. Interesting! :thumbup:

 

[analog.universe]

The amount of linear movement in the projected image at the sensor plane will be identical no matter the format, however, APS-C sensors typically have greater pixel density, and therefore that linear movement translates into a greater blur in the final image when measured in pixels.

Honestly the whole rule to begin with is pretty quick and dirty... think of it as a really rough guideline and not something that guarantees you results. The weight and speed of the mirror mechanism, weight of the camera+lens combo, distance to the subject, effective DOF, how much caffeine is in your blood, etc... will all affect the practical shutter speed limit. For me it's been one of those things you learn on a lens by lens basis. For critically sharp I'll shoot 35mm at 1/125, because I can see a difference from 1/60. I've still gotten images that we're plenty acceptable out of that lens at 1/40, all depends on the situation.

So, my suggestion is to just shoot lots of frames with all your lenses and learn how each one responds when you get into sketchy-shutter-speed territory. Some do better than others, and as long as you pay attention, you'll figure out what they all prefer eventually.

 

[SCraig]

That rule of thumb is just that: A rule of thumb. It is not cast in stone and there are other variables that play a part as well.

I have a shot of an owl that I took hand-held with a 300mm lens at 1/30 second with VR and it is sharp enough to see tiny specs in his eyes, but the distance was only something like 8 feet. I also have shots that I've taken at 300mm at over 1/1000 that are lousy with blur from me shaking. Is it windy? Are you cold? Did you drink too much coffee this morning? There are lots of things that are going to have an effect on whether that "Rule" works today or not.

 

[Big Mike]

I've been on both sides of this debate...and I've yet to see a really definitive argument or test result.

The argument for including the crop factor in this equation, is that even though the focal length is the same with crop or full frame, the image from the crop camera gets enlarged more (from the smaller sensor size, to an 8x10 print, for example). So as you enlarge more, you amplify the visibility of blurriness from camera shake.

Another thing to factor in, is that (for the most part) we are more critical of sharpness now than back when everyone shot film. Back then, most people evaluated their photos by looking at the negatives, or small prints. Now, practically everyone can load their photos onto a computer and zoom in to 100% or more.
And because, this rule of thumb is subjective...and really refers to 'acceptable levels' of blur from camera shake...if we are more critical of it, then we must increase the rule to get sharper images.

A better rule of thumb would be to use a shutter speed (reciprocal) that is equal to 2x the focal length.

 

[o hey tyler]

The amount of linear movement in the projected image at the sensor plane will be identical no matter the format, however, APS-C sensors typically have greater pixel density, and therefore that linear movement translates into a greater blur in the final image when measured in pixels.

Basically what I was saying, except more AU was much more eloquent.

 

[Nikon_Josh]

The amount of linear movement in the projected image at the sensor plane will be identical no matter the format, however, APS-C sensors typically have greater pixel density, and therefore that linear movement translates into a greater blur in the final image when measured in pixels.

Honestly the whole rule to begin with is pretty quick and dirty... think of it as a really rough guideline and not something that guarantees you results. The weight and speed of the mirror mechanism, weight of the camera+lens combo, distance to the subject, effective DOF, how much caffeine is in your blood, etc... will all affect the practical shutter speed limit. For me it's been one of those things you learn on a lens by lens basis. For critically sharp I'll shoot 35mm at 1/125, because I can see a difference from 1/60. I've still gotten images that we're plenty acceptable out of that lens at 1/40, all depends on the situation.

So, my suggestion is to just shoot lots of frames with all your lenses and learn how each one responds when you get into sketchy-shutter-speed territory. Some do better than others, and as long as you pay attention, you'll figure out what they all prefer eventually.

Very well put!

That rule of thumb is just that: A rule of thumb. It is not cast in stone and there are other variables that play a part as well.

I have a shot of an owl that I took hand-held with a 300mm lens at 1/30 second with VR and it is sharp enough to see tiny specs in his eyes, but the distance was only something like 8 feet. I also have shots that I've taken at 300mm at over 1/1000 that are lousy with blur from me shaking. Is it windy? Are you cold? Did you drink too much coffee this morning? There are lots of things that are going to have an effect on whether that "Rule" works today or not.

Very true, this wasn't quite what I was getting at though. The rule of thumb can always be broken, I strongly agree. I have got shots with my 70-300 VR the same as you that have stunned me, sharp shots set to 300mm from 1/40 of a second.

My point is more based around advice given to people who use crop cameras in particular. The focal length stays the same, but I have seen alot of people saying you need to match the EQUIVALENT focal length with the shutter speed. This thread is helping me understand why! :thumbup:

 

[bazooka]

I have a different take that I thought was explained by a combination of Tyler and Mike's comments, but I still don't think it's clear. Let me propose an example....

Full frame camera with a 50mm on tripod, shooting a car in the distance travelling from right to left through the frame. Given an amount of time, the image of the car will travel across the sensor a certain distance. Let's say in half a second, the car travels half the length of the sensor.

Take a 2x crop (exaggerating for effect and ease of conversion) with a 50mm on a tripod, same car, same speed, same distance. The car is magnified in the frame (same size on a smaller sensor). Now, in half a second, the car travels the entire length of the sensor (speed is the same). The motion blur is twice as much given the same print res.

The same applies if you shoot handheld... instead of a car moving across the frame, the entire scene is moving at the speed that your hand is wobbling. So, in my thinking, you absolutely need to take into account the crop factor if you are trying to replicate the same amount of shake... or lack or. That's how I understand it anyway. If you magnify the scene via cropping it, you are magnifying the speed the scene moves.

Am I missing something?

 

[o hey tyler]

Third, pixel density has nothing to do with it. It's a simple relationship between angle of view and person's angle of movement.

I beg to differ.

 

[o hey tyler]

If your basis for differing is this...

Then the answer is still that the pixel density has nothing to do with it.

Lets say that in a 6MP image there was motion blur of 3 pixels in the direction of movement. If the same image, with same blur, would have been taken with a 24MP camera, then the blur would be twice as much, or 6 pixels.

However...when both images are printed on the same paper size, a width of 3 pixels from the first image take the same space as a width of 6 pixels from the second image. So the blur is the same...pixel density has no effect.

That's true... If you never ever ever intend to crop your images. Which is extremely uncommon in the photographic field. So yes pixel density is relevant.

 

[Robin Usagani]

Look at it this way.. I have a lot of shots that look good. Then I crop it (zoomed closer), then I start seeing it is softer due to camera shake.

Yes... it is correct.

 

[IByte]

Ugh why can't there be a simple mathematics or physics formula.