A question about 1.6 magnification

[jsfpa@comcast.net]

This is probably a stupid question but, how dose the 1.6 magnification effect the focal length over one rule when setting shutter speed? Do you calculate the focal length or do you just go with the lens focal length?

[stupot]

well first of all rules are there to be broken:) if you have steady hands or are feeling risky then ignore it..!

but yes, the cropped sensor cameras require you to multiply your focal length by the crop to give you a suitable handholdable speed.

so, if you're at 200mm then you can expect sharp results at 1/320. however i'll happily go down to 1/125 or so, depends how much coffee ive had!

[mrfourcows]

do you need to multiply? i usually go according to the lens' focal length.

[GyRob]

on a 1.6 crop you need to go to the field of view to be safer from camera shake .
so if your lens is 200mm - times that by 1.6 = 320 50 mm x1.6 = 80 .
Rob.

[Woolburr]

The general rule of thumb is to factor in for the crop as well as the focal length. Much safer to take an image at 1/320 with your 200mm lens than it is to try taking one at 1/200....Photography lends itself well to erring on the side of caution.

[Sam]

My experience with this has lead me to believe that it is all up to the focal length, despite the crop factor. You are not magnifying the image with a 1.6 crop camera, you are cutting the edges off of the image. So the focal length is the same, your view is just similar to that of the focal length x 1.6.

If you take a picture of an object with the same lens twice one with a FF camera and one iwth a 1.6 you get the same magnification, just less image on the 1.6.

That's my understanding. I'm no expert though so take it for what it's worth.

[dontblink]

The whole 1/focal length thing is just a rule of thumb, not based on anything really. But yes the 1.6 FOVCF does increase camera shake, and shutter speed should be increased accordingly.

[Sam]

I can hand hold my shots at 1/200 the same on my 1.6 @200mm as I can on my film cameras. I don't notice any more shake. It's a crop, not a magnification.

[Wilt]

I can hand hold my shots at 1/200 the same on my 1.6 @200mm as I can on my film cameras. I don't notice any more shake. It's a crop, not a magnification.

I disagree. If you consider the fact that for the same framing at the same shooting distance, you would use 100mm on FF vs. 62mm on 1.6, it *is* correct to factor in the format in the calculation!

Yes, shake is an angular related thing. Consider this...Hand holding 400mm lens on FF is like hand holding 250 mm lens on 1.6 crop format for the same framing at the same shooting point. You would use 1/400 on FF, and you would use 1/400 on the 1.6 crop camera also, n = 250 * 1.6. The same relative angle displacement occurs with 400mm on FF, as 250mm on 1.6 format.

[mysubaruimp]

How about this:

Go try it yourself and see what works.

Different people use different speeds. Find what works for you and ignore all the "rules".

[CorruptedPhotographer]

the 1.6 is not magnification. Your 100mm lens is not magically 160mm on your 1.6x crop sensor.

[Wilt]

the 1.6 is not magnification. Your 100mm lens is not magically 160mm on your 1.6x crop sensor.


This is true...to an extent!
If you stand with a FF camera and a 100mm lens and aim at a sign, for example, you see a certain amount of that sign. If you stand in the same place and use the 100mm lens on the 1.6 crop format, the framing is the same as if you had used 160mm on the FF camera. So the 1.6 is a 'equivalence' conversion factor so you can equate the Field of View with a lens on 1.6 crop camera to what you might have grown accutomed to over many years of 35mm film camera usage. REPEAT: 100mm lens on a 350D/20D/30D results in the same framing as 160mm lens on 35mm film camera. So in a sense, it is 'magnification'. (If you could see the focal plane area of both cameras, the image would be identical in physical size, however, and is NOT resulting in any real 'magnification' physically.)

[BryanP]

Try this:

Try moving your head a slight angle (without messing with your field of view) then use your hands aftewards to form a "binocular shape" to limit your field of view. Which "seems" more sensitive?

That's essentially what you're doing. Our eyes are still retaining the same amount of magnification but because you're limiting the field of view, things "appear" to be more sensitive. You can probably get away with the rule of thumb but technically, giving yourself more shutter speed than 1/focal length at long lengths is recommended if you really want to freeze action (in fact, it doesn't even matter what length you're at... you know what shutters will freeze action at particular lengths).

This is the problem with some people, as they think their pictures are sharp at 33% or whatever Photoshops zooms in and when they go to 100%, you can see some blur.

[corinto]

Another crop-factor war! I love this.

I am with Solinger in this.

Esentially, you must not think of the crop factor in terms of what's happening in the camera and sensor but in terms of the end result which is the photo. In order to get the same framing with a 1.6 camera, what you will do is step back (of course, you could always use a wider lens but that would mean even lower shake). Consequently, for the same framing, hand-holding rule of thumb is exactly the same; you should not factor in for crop.

Crop factor affects amplification (CoC and all that; smaller image --> requires bigger amplification for same final print) but it does not affect camera shake.

[runninmann]

I've thought about this often in the short time I've been on this board and, for the life of me, I cannot get it through my thick skull how the crop factor plays into the 1/f rule-of-thumb. If the lens magnification is the same on a crop camera as on a FF camera (and I believe most will agree that it is) and the crop is just that -- cropping some percentage of the top, bottom, right and left off of a FF view, I struggle to understand how that increases the effects of camera shake, requiring faster shutter speeds for the same lens magnifications. I tried to illustrate this for myself using the crude illustration attached. For the sake of simplicity, I assumed that the camera shake/movement is generated at the center of the sensor plane. Based on that, it is apparent that the farther away from the sensor plane (lens focal length), the greater the effect of the movement. However, it is not at all intuitive to me how the crop has any effect (edit)on camera shake, regardless of focal length.

[Sam]

Imagine it this way. If you have a projector and you are looking at slides they are coming out at a fixed size. In my example this will represent the job of the lens.

You have a white screen set up so that you can view the picture. In my example this represents the sensor.

If you take an image and view it on the screen and then reduce the size of the screen (or crop it 1.6) the image technically stays the same size but the area on which you view it is smaller, making whatever the subject of the shot is fill the frame more and appear "magnified". The image was actually cropped not enlarged.

Same this happens with lens/film(sensor) combinations.

A 75mm lens on a medium format is a whole other thing compared to a 35mm camera.

[runninmann]

Imagine it this way. If you have a projector and you are looking at slides they are coming out at a fixed size. In my example this will represent the job of the lens.

You have a white screen set up so that you can view the picture. In my example this represents the sensor.

If you take an image and view it on the screen and then reduce the size of the screen (or crop it 1.6) the image technically stays the same size but the area on which you view it is smaller, making whatever the subject of the shot is fill the frame more and appear "magnified". The image was actually cropped not enlarged.

Same this happens with lens/film(sensor) combinations.

A 75mm lens on a medium format is a whole other thing compared to a 35mm camera. I wasn't clear in my post, so I've edited it. I meant to convey that I don't understand how crop factor as any effect on camera shake.

[Sam]

I wasn't clear in my post, so I've edited it. I meant to convey that I don't understand how crop factor as any effect on camera shake.

I actually typed this before you posted, I just got sidetracked for a few minutes and didn't hit post...

[remixity]

I would think that pixel density is more important for this rule of thumb than apparent field of view (the crop factor). Then again, the rule is just a guideline.

If you have steady hands, then 1/physical focal length might do you just fine. Customize the rule of thumb to your own shooting style.

[Wilt]

I have put together an illustration. This shows that regardless of format, you use the same shutter speed for the handholding speed. The speed is stated relative to the 35mm film format norm, calculated by (speed= 1 / F.L.) But the speed, relative to the 'normal' focal length used for different formats (in the drawing, medium format vs. 35mm film vs. 20D) can be calculated by the common formula using a factor to normalize for differences to equivalent lens FOV.

http://i69.photobucket.com/albums/i63/wiltonw/lens_shake.jpg

[SkipD]

I wasn't clear in my post, so I've edited it. I meant to convey that I don't understand how crop factor as any effect on camera shake.Here's why the "crop factor" (actually a difference in "film size" format) affects the camera shake rule of thumb:

Let's assume that you take two photos of the same subject and from the same camera position - one with a "full-frame" digital camera (such as a 5D) using an 80mm lens and another with an APS-C camera (such as a 30D) using a 50mm lens. Both images will have the same amount of the subject material, and will have the same perspective, etc. Now for the difference: When you go to print a 8inch x 12inch, you need to magnify the image from the APS-C camera more to fill the print than you do for the image from the "full-frame" camera. The image formed by the lens and captured by the sensor was physically much smaller for the APS-C camera. Thus, any shaking of the APS-C camera is being made more noticeable by the greater amount of enlargement to make the print than that made from the "full-frame" camera.

The bottom line is that if you want to use the old "rule of thumb" (that was originally stated for 35mm film cameras) for an APS-C, you need to multiply the focal length by the "crop factor" value.

However, a "rule of thumb" is just that. Some folks can hold a camera very steady and can use much slower shutter speeds effectively, and some need much faster shutter speeds.

I feel the need to re-emphasize something. Lenses do NOT change their focal length or aperture range when you put them on cameras with different "film" formats. Nothing changes except the field of view, and that is totally a function of the size of the film or digital sensor.

[Jon]

THANK YOU SKIP!
You can't neglect post-capture magnification. Ignore the "crop factor" in your 1/f.l. calculations only if you make 8x10 prints full frame and 5x7 from your crops. Note also that if you want to go for larger prints, you need to allow for this in figuring both safe hand-holding speed and depth of field calculations.

[KennyG]

Oh my goodness, so those of us shooting sports at 600m at 1/320 on a 1.3 crop factor have commited come cardinal sin and our photographs are useless? I must re-wind years of work.

It is a GUIDE, not a rule and is overplayed to the point where some people ruin what could be a good shot by using it. Go with what works for you to give you the look you are after in the shot or what circumstances dictate.

[Wilt]

THANK YOU SKIP!
You can't neglect post-capture magnification. Ignore the "crop factor" in your 1/f.l. calculations only if you make 8x10 prints full frame and 5x7 from your crops. Note also that if you want to go for larger prints, you need to allow for this in figuring both safe hand-holding speed and depth of field calculations.

Assuming typical viewer reaction to larger prints, they view the larger print and the same old viewing distance, and what you're saying is true! If viewers stepped back to view large prints at the same relative angle of view to one another, they would not notice differences due to print size.

[Jon]

It's a rule of thumb, as we've said repeatedly around here. What's wrong is to asume that you don't need to consider the crop factor on the body in figuring it. I don't suppose you're using the 600IS on a monopod, as in your avatar?

[Jon]

Assuming typical viewer reaction to larger prints, they view the larger print and the same old viewing distance, and what you're saying is true! If viewers stepped back to view large prints at the same relative angle of view to one another, they would not notice differences due to print size.And if they're typical POTN members they'll move in closer!
:{)#

[Wilt]

It's a rule of thumb, as we've said repeatedly around here. What's wrong is to asume that you don't need to consider the crop factor on the body in figuring it. I don't suppose you're using the 600IS on a monopod, as in your avatar?

And the evidence it is a 'rule of thumb' is that it was invented in the days of medium format, preceding the 'miniature format' as 35mm was originally known even into the 60's. 35mm film photographers had been using the exact same rule (and successfully getting away with it!) on the smaller format without the multiplication factor applied at all. The margin for error was reduced, but in spite of that, the same old rule of thumb continued to work. Makes one wonder about when the margin for error will have reached to 'too small of a margin for error' point!

[runninmann]

Here's why the "crop factor" (actually a difference in "film size" format) affects the camera shake rule of thumb:

Let's assume that you take two photos of the same subject and from the same camera position - one with a "full-frame" digital camera (such as a 5D) using an 80mm lens and another with an APS-C camera (such as a 30D) using a 50mm lens. Both images will have the same amount of the subject material, and will have the same perspective, etc. Now for the difference: When you go to print a 8inch x 12inch, you need to magnify the image from the APS-C camera more to fill the print than you do for the image from the "full-frame" camera. The image formed by the lens and captured by the sensor was physically much smaller for the APS-C camera. Thus, any shaking of the APS-C camera is being made more noticeable by the greater amount of enlargement to make the print than that made from the "full-frame" camera.
Eureka! Of course! The reason some people shoot photos -- the print;) !

[curiousgeorge]

How about this:

Go try it yourself and see what works.

Different people use different speeds. Find what works for you and ignore all the "rules".

Isn't photography supposed to be about understanding the theory first.

I will test this, but I agree with Sam's logic simply because it is a crop and not a magnification

George

[curiousgeorge]

If you consider the fact that for the same framing at the same shooting distance, you would use 100mm on FF vs. 62mm on 1.6, it *is* correct to factor in the format in the calculation!

Just because you would use a 100mm to get the same effect, it doesn't mean you are. It's still a 62mm lens.

On a FF body, when applying the 'rule', you wouldn't multiply your focal length by 1.6 to keep the centre part of the image still, would you.

[RuggerJoe]

I think the best answer is to test it yourself. Every one is different. Some people can hand hold down below the "rule", some people aren't as lucky and have to go faster. Start at the 1/focal length and see for yourself and adjust accordingly.

[Jon]

Isn't photography supposed to be about understanding the theory first.

I will test this, but I agree with Sam's logic simply because it is a crop and not a magnification

GeorgeYes, but you need to magnify it more to get the same print size. And that'll enlarge the shakes just as much as everything else in the image.

[curiousgeorge]

you need to magnify the image from the APS-C camera more to fill the print than you do for the image from the "full-frame" camera
and

Yes, but you need to magnify it more to get the same print size.

I totally disagree with this (which is what your arguments are based on).

You still get the same number of pixels (8 million or so) on both bodies so why do you need to magnify the image??

[SkipD]

I totally disagree with this (which is what your arguments are based on).

You still get the same number of pixels (8 million or so) on both bodies so why do you need to magnify the image??An image produced inside a 30D is physically smaller than the image produced inside a 5D. This has absolutely nothing to do with the resolution of the digital image.

It's just like using an old Olympus half-frame 35mm camera versus a conventional "full-frame" 35mm film camera. The Olympus produces a negative (or slide) that is half the area of that from a conventional 35mm camera. In order to make the same sized (call it 8x10 inch) prints from both cameras, you have to enlarge the Olympus half-frame negative more than you do the negative from the conventional 35mm camera. Thus, any aberration on the smaller negative (caused by camera shake, dust on the negative, or whatever) is going to make more of an impact in the print than a similar aberration in the larger negative. This is true even if you used a super-fine-grain film in the little Olympus half-frame camera and a coarser high-speed film in the conventional 35mm camera (evening out the equivalent of digital resolution for the two formats).

There's nothing magic about the current generation of "crop" DSLR's - they are simply a DIFFERENT FORMAT camera from 35mm cameras, even though they are designed to use the same series of lenses that fit on 35mm film cameras. Cameras with different film formats have always had differences that one either chooses to deal with or avoid the format.

[CorruptedPhotographer]

So which format (print size) is perfect for 1.6x and 1.3x crop?

Perfect meaning, no need to "stretch" the image at that print size.

From the above, needing to print any size larger than 8x10 requires it to be stretched.

[curiousgeorge]

An image produced inside a 30D is physically smaller than the image produced inside a 5D. This has absolutely nothing to do with the resolution of the digital image.
But when you're producing an enlarged print, the one thing you take into account is resolution as this will affect the dpi of the print. NOT the sensor size. As you get the same resolution from both bodies, I don't see what the difference is.

[curiousgeorge]

So which format (print size) is perfect for 1.6x and 1.3x crop?

Perfect meaning, no need to "stretch" the image at that print size.

From the above, needing to print any size larger than 8x10 requires it to be stretched.

I thought this is purely down to the resolution. And you're aiming to get at least 300dpi on a print.

So if you want a 10 x 8 print, the image need to have a resolution of at least 3000 x 2400 (which is 7.2 megapixels).

[CorruptedPhotographer]

then how come Jon and the others imply or even state that because of the sensor size, the image from a 5D would be perfect @ 8x10 but smaller sensors will have to be stretched to fit 8x10 print size.

[curiousgeorge]

then how come Jon and the others imply or even state that because of the sensor size, the image from a 5D would be perfect @ 8x10 but smaller sensors will have to be stretched to fit 8x10 print size.
Well that clearly can't be right. Cameras are sold on, amongst other things, the number of megapixels. How often do you hear about the sensor size??

[jaymrobinson]

Interesting question, let's think about it using an illustration and no numbers. Numbers just tend to confuse us.

First, let's take a blurry picture of a ball with a APS-C size sensor and Full Frame sensor. (Blur is exaggerated for the purpose of explaining.)

http://www.pbase.com/jaymrobinson/image/64073435/original.jpghttp://www.pbase.com/jaymrobinson/image/64073435/medium.jpg

Then lets make a 4x6 print of the image from both cameras. Of course we will want to print the whole image captured by each camera's sensor (as in most real world examples), consequently, the picture from the APS-C size sensor camera will need to be magnified more compared to the full frame camera.

http://www.pbase.com/jaymrobinson/image/64073446/original.jpghttp://www.pbase.com/jaymrobinson/image/64073446/medium.jpg

As you can see, the blur is indeed "magnified" in the APS-C print. I guess that answers that question. This means you will need a faster shutter speed with the same lens on the 1.6x camera to have the same amount of blur on the print.

[Wilt]

then how come Jon and the others imply or even state that because of the sensor size, the image from a 5D would be perfect @ 8x10 but smaller sensors will have to be stretched to fit 8x10 print size.

That statement,"from a 5D would be perfect @ 8x10 but smaller sensors will have to be stretched to fit 8x10 print size" is totally erroneous.

OK, guys, some elementary arithmetic...the ratio of sides is one side divided by the other side to thelowest whole integer a both sides of the ratio (apologies to mathematicians in the defintion, but I am merely trying to get this point across!)...
CAMERA
35mm film = 24 x 36mm = 2:3
1DsII = 5D = 24 x 36mm = 2:3
20D = 22.5mm x 15 mm = 2:3
large format = 90 x 150mm = 4:5
PRINTS
4x6 conventional drugstore print = 2:3
4x5 conventional portrait studio print = 4:5
8x12 drugtore enlargment = 2:3
8x10 drugstore enlargement = 4:5
5x7 drugstore enlargement = 5:7
11x14 lab enlargement = 11:14
16x20 lab enlargement = 4:5

If your sensor or film is NOT EXACTLY the same ratio as the print size you want, YOU MUST CROP the shot or the lab will do it somewhat arbitrarily for you in printing your shot!

[curiousgeorge]

This is deviating from the original question of whether shutter speed needs to be increased for crop cameras, and instead addressing the point of whether or not a print from a crop camera is inferior to that of a FF one, due to the 'stretching' of the image.

[Wilt]

Interesting question, let's think about it using an illustration and no numbers. Numbers just tend to confuse us.

First, let's take a blurry picture of a ball with a APS-C size sensor and Full Frame sensor. (Blur is exaggerated for the purpose of explaining.)

http://www.pbase.com/jaymrobinson/image/64073435/original.jpghttp://www.pbase.com/jaymrobinson/image/64073435/medium.jpg

Then lets make a 4x6 print of the image from both cameras. Of course we will want to print the whole image captured by each camera's sensor (as in most real world examples), consequently, the picture from the APS-C size sensor camera will need to be magnified more compared to the full frame camera.

http://www.pbase.com/jaymrobinson/image/64073446/original.jpghttp://www.pbase.com/jaymrobinson/image/64073446/medium.jpg

As you can see, the blur is indeed "magnified" in the APS-C print. I guess that answers that question.

OK the logic illustrates only if you use the SAME LENS on both bodies. But few of use would do that...for example you put the 'normal' lens on whose focal length is suited to the dimensions of the frame, e.g. 50mm FF, 32mm 1.6 crop to provide 45 degrees of coverage (number used here for illustration).

In that case the images are proportionally sized to the frame. When your hand shakes in both cases, it shakes a given linear distance (e.g. 1mm), so that in shake is greater percentage of the smaller format (1mm:15mm vs. 1mm:24mm). BUT it is the SAME ANGULAR DISPLACEMENT on both formats because the subject angle in the frame is the same and the camera is being displaced by the same angle of change!

So if your make lens focal length proportional to the crop size, the angular displacement of the shake remains the same, but one has to use the relative crop size to determine the size of equivalent lenses!

[jaymrobinson]

OK the logic illustrates only if you use the SAME LENS on both bodies.
I thought the original question dealt with the same lens on both bodies. :rolleyes:

Your argument has too many numbers and difficult words. Maybe you should make an illustration ?? :)

"I don't like numbers, and I don't like facts. Don't look things up in a book, look them up in your gut." - Stephen Colbert

[Wilt]

I thought the original question dealt with the same lens on both bodies. :rolleyes:

Your argument has too many numbers and difficult words. Maybe you should make an illustration ?? :)

"I don't like numbers, and I don't like facts. Don't look things up in a book, look them up in your gut." - Stephen Colbert

SIX sentences is too much?!?! ;) Read post # 20 which I wrote! Now if you shake every camera by the width of one eyebrow hair as seen on the subject, for example, and the width of that hair is 1/10 of a degree, it is the SAME ANGLE on ALL cameras. So while it is true that you magnify the smallest format the most in order to get the same giant enlargement, the angular displacement stays the same for all prints. So amount of blur is the same...if you size the lens proportionally, which is not what post 40 illustrates.

The OP makes no assumptions about focal length, only about relative crop size, when asking about the shutter speed question.

[Jon]

and



I totally disagree with this (which is what your arguments are based on).

You still get the same number of pixels (8 million or so) on both bodies so why do you need to magnify the image??Because the 8.2 MP of a 20D are crammed into a 15x22.5 mm area. 8.2 MP on a FF sensor are spread over a 24x36 mm area. To get an 8x10 out of the FF 8.2 MP, you need to enlarge it 200 (8") mm/24mm or just over 8x. To get the same 8x10 out of a 20D you need to enlarge it 200/15 or about 13x. (1.6x more than you had to enlarge the FF "negative"). In addition to enlarging the image you're enlarging the [details (including the shake) within that image.

Put another way, the same physical blur from shake on the 8.2 MP 20D original will occupy 1.6x (linear, 2.56x areally) as many pixels as that exact same amount of blur from that exact same lens on a FF 8.2 MP sensor camera because the same number of pixels are much smaller on the 20D sensor. So when you enlarge to your final print size you've got more "blurry" pixels in the 20D print.

[Jon]

then how come Jon and the others imply or even state that because of the sensor size, the image from a 5D would be perfect @ 8x10 but smaller sensors will have to be stretched to fit 8x10 print size.We never "imply or even state" any such thing. All we say is that you don't need to enlarge a FF original as much to achieve a given print size as you do a smaller "negative" (sensor capture).

[foxbat]

It's all about the angles.... Field of view is measured in degrees. The crop factor decreases your field of view.

If I have a field of view of 100 degrees and I shake the camera such that it moves by 1 degree then the image displaces by 1%.

If I have a field of view of 50 degrees and I shake the camera such that it moves by 1 degree then the image displaces by 2%.

Easy to understand?

[CorruptedPhotographer]

Jon, ok sorry, that is what I meant, an image needs to be stretched. What you said.

[Albert Harrison, Jr.]

Is there an engineer in the house?

[runninmann]

Is there an engineer in the house?At least one, but he got a satisfactory answer way back in post #21:)

[In2Photos]

Is there an engineer in the house?

Yes, but this has nothing to do with my field of expertise.:D A good read though.

[curiousgeorge]

Because the 8.2 MP of a 20D are crammed into a 15x22.5 mm area. 8.2 MP on a FF sensor are spread over a 24x36 mm area. To get an 8x10 out of the FF 8.2 MP, you need to enlarge it 200 (8") mm/24mm or just over 8x. To get the same 8x10 out of a 20D you need to enlarge it 200/15 or about 13x. (1.6x more than you had to enlarge the FF "negative"). In addition to enlarging the image you're enlarging the [details (including the shake) within that image.

So you're saying prints from crob bodies are inferior to those of FF bodies. Is this widely accepted as a fact?

I take your point for prints, but when viewing the image on the screen there should, in theory, be no more visible blurring or loss of sharpness?

[René Damkot]

Prints are not inferior, but with a given amount of motion, you will see more motion blur on the print from the crop body.

Put another way, the same physical blur from shake on the 8.2 MP 20D original will occupy 1.6x (linear, 2.56x areally) as many pixels as that exact same amount of blur from that exact same lens on a FF 8.2 MP sensor camera because the same number of pixels are much smaller on the 20D sensor. So when you enlarge to your final print size you've got more "blurry" pixels in the 20D print.

If you have 0,01" of movement on the sensor, it will show up as approx. 0,08" on a 8x10 print off a FF, and as 0,13" on an 8x10 print off a 20D.

[jaymrobinson]

Look, if you are going to put a long lens on a crop body, in most situations, you are going to want to take advantage of that extra "crop factor" as it is called. So, lets say you want to take a picture of a baseball player, and he fills up the frame on your 20D. Used on a 5D, he isn't going to fill up the whole frame. You would need to use a longer lens to get the same shot on the 5D. Which means the required shutter speed to stop blur is going to need to be shorter.

Now, some might argue that if you took the same picture, and cropped it on your 5D, you would get the same shot as if you took it on the 20D with the same lens. But you have to remember that the pixel density is much higher on the 20D, so it does give you a longer effective reach.

[curiousgeorge]

Prints are not inferior, but with a given amount of motion, you will see more motion blur on the print from the crop body.
But then it's not just when you have a blur caused by movement. Any image has boundaries (eg, the edge of a rock against a sky) and across these boundaries there will be some blur.

This too will be magnified therefore it can be said that all images look inferior from crop bodies.

[René Damkot]

I think I don't see your point: what blurr across boundaries?
Then again, since the two images are taken with the same framing, the objects in the picture will be equally large in the final prints. So there will be no difference there I guess.

[Jon]

So you're saying prints from crob bodies are inferior to those of FF bodies. Is this widely accepted as a fact?

I take your point for prints, but when viewing the image on the screen there should, in theory, be no more visible blurring or loss of sharpness?More like the more you want to enlarge an original, the more critical the quality of that original becomes. And that is widely accepted.

I'm not clear why you think viewing an image on your monitor should be any different from viewing it on a print. If you're viewing the entire image, I'll grant you that the reduction necessary to cram a, say, 3072x2048 image onto a 1024x768 screen will obscure a lot of flaws. If you're talking about pixel-peeping at 100%, you'll see the limits of the lens/sensor combination, and if you're not using the same lens for both, the differences in lens qualities will also come into play.

Remember, the original question was whether you need to allow for the "crop factor" in applying the 1/f.l. steadiness rule of thumb. And the answer is "Yes", because you're enlarging the shake along with everything else.