Crop-factor question

Steep said:
You may well be a lecturer or a technical writer in your other life, you haven't said which you are but here you're just another member who when you haven't been insulting or patronising (and you do continue to do both in post after post) has stated your opinion. You say it's fact, I disagree.
Click to expand...
Neither of those are me? :bang:
Play the ball - not the man; You might see me as an easy target - but it doesn't alter the fact that some things in photography are based on maths and physics. At which point, my opinion, your opinion or anyone else's opinion ceases to be relevant. There are real rules that govern maths and physics.
And at the point when opinion ceases to be relevant, people who get paid for their professional knowledge become the people you'd want to listen to.
Arguing with me might be fun, personally insulting me might be your bag too - I really don't care. But refuting the laws of physics is a bit daft:bonk:

I love the way that the people who are wrong in this thread are clinging to the word 'opinion' as if it makes a difference - You're entitled to believe what you want, but science requires a measurement of evidence, and the measurement of evidence is what people who are right are calling facts, they're different to opinions.

And still, at the end of the day, it's only a rule of thumb, and some people have really steady hands and some people need a tripod to get a sharp picture at any speed.
 
Steep said:
You may well be a lecturer or a technical writer in your other life, you haven't said which you are but here you're just another member who when you haven't been insulting or patronising (and you do continue to do both in post after post) has stated your opinion. You say it's fact, I disagree.
Click to expand...

Hoppy is a well respected technical writer. Pookeyhead works on the academic staff at an FE and I'm a press photographer.

All three of us think that you are wrong. Does that not make you want to pause and think?
 
Steep said:
You mean enlargement don't you? the magnification on a 50mm lens is no different on a 1.6 sensor compared to a full frame.
Click to expand...

Enlargement/reduction, magnification, same difference really.

The bit that some folks are missing is these things can only be compared when the entire imaging chain is included - subject distance, focal length, sensor size, output size/print, and finally, viewing distance.

It's the flipside of the familar depth-of-field/sensor size debate where the same arguments are common. And yes, that changes too.
 
Last edited:
Pookeyhead said:
While the speed you can shoot hand held may be different for different people, I think it's all relative, and therefore relevant. If the lowest speed I could hand hold a 400mm lens on a full frame camera was 250th of a second, then the most I could get away with at 300mm on a crop sensor camera would also be 250th, as the magnification is the same.

It's still important to know this, no matter how steadily you can hold a camera. As you can see in this thread, there can be a great deal of confusion, and many people think it's a product of absolute focal length only and assume that if they can shoot at 250th with a 400mm lens on a D800, then it will be 250th with a 400mm lens on a D7000 too.. and it won't.

I'd say it's far from irrelevant, and something everyone should know. Not the exact speeds, as that changes, as you point out, but the fact that it's NOT focal length that matters, but magnification.
Click to expand...

This is perhaps the most useful post on this thread (y)

Edit: just to add, not only are we all different in how steadily we can hand-hold, our stance/posture can make a big difference. Leaning against a wall is way better than stooping or streching uncomfortably to get a better angle. Good technique is important too, basically taking all the weight of the camera cupped in your left hand, with elbows gently braced against your body, mainly the left arm tucked underneath your boob to form a supporting triangle. Your right hand should do nothing but balance the camera and be free to operate the controls.

When you're at or near the limit, not every picture will be unsharp - luck of the draw. If you can shoot a few frames, that increases the odds and there's a good chance one will be acceptable.

And macro shooters should be aware that all the 'rules' go out of the window when you're close. At 1:1 the image dances around like you're using a super-telephoto (magnification again). One of the reasons flash is so handy with macro.
 
Last edited:
HoppyUK said:
....
And macro shooters should be aware that all the 'rules' go out of the window when you're close. At 1:1 the image dances around like you're using a super-telephoto (magnification again). One of the reasons flash is so handy with macro.
Click to expand...

I thought magnification was irrelevant;) :coat:
 
It's amazing that this has gone on so long and got so heated at times when anoyone who has a digital camera can quickly find out things such as "At how low a shutter speed can I take a hand held shot and get a keeper?"

I once took a 20mm shot at something like 2 seconds and it was a keeper, but I couldn't do it every time (I probably couldn't do it again...) and I know that even at 50mm I want to be heading above 1/100 sec or taking multiple shots (in the hope of getting a steady "keeper") if at all possible :D
 
Last edited:
woof woof said:
I once took a 20mm shot at something like 2 seconds and it was a keeper, but I couldn't do it every time (I probably couldn't do it again...) and I know that even at 50mm I want to be heading above 1/100 sec or taking multiple shots (in the hope of getting a steady "keeper") if at all possible :D
Click to expand...

Well to be honest none of this matters to me anyway. I've cut right down on booze as part of a bit of a health kick, my shutter speed's needing to be around 3 stops faster than my focal length to stop the shaking and that's with IS! :p
 
PMN said:
Well to be honest none of this matters to me anyway. I've cut right down on booze as part of a bit of a health kick, my shutter speed's needing to be around 3 stops faster than my focal length to stop the shaking and that's with IS! :p
Click to expand...

Off Topic


If you are getting shakes due to reduced alcohol intake, then I would very seriously recommend a chat with your GP. It's definitely not something to shrug off lightly!

/Off Topic
 
PMN said:
It was purely a joke but thanks for the advice! :)
Click to expand...

I'm glad that you were joking about reducing your alcohol intake :D

I find that when I drink my keeper rate goes up and my photos look excellent. It's only when I sober up that some shots head to the bin.
 
Last edited:
PMN said:
If I set my 100-400L to 100mm on my 5D2 and set my shutter speed to 1/100th then I'm going along with the 1/focal length guideline, correct? If I then take a scalpel and trim the sides off the 5D2's sensor (a horrific thought but just go along with me) so it's 22.5mm wide then what else, apart from how much of the image the camera sees, has changed?

Absolutely nothing has changed. The centre portion of the sensor is seeing the image in exactly the same way and this is precisely the same difference as mounting a lens on an APC-S body compared to a full frame. There's no need to compensate for the 1.6 crop factor because the small sensor is simply discarding the outside of the image rather than magnifying the centre of it.
Click to expand...

That is indeed true, but what changes is when you consider the output stage rather than the taking stage. Let's say that you want to print the image from both cameras at 12x8. The image from the crop sensor will have to be blown up more to make a print of that size, and this is where the extra magnification becomes apparent. If there is any blurring due to camera shake, it will be more obvious in the print from the crop sensor because it has been enlarged more. Perhaps that's why people advocate a slightly faster shutter speed than 1/focal length when talking about crop sensor cameras.
 
Woah... So it wasn't a simple question then :D

I knew it was a rule of thumb, but was curious about the technicalities of it all.

Thanks for the multiple, often colourful replies!
 
It also depends very much on technique and the person taking the shot. Some people are naturally "shakey". Some have a very steady stance. Just something else to think about.

Oh, i`m not a writer,lecturer nor a staffer, but Hugh, Hoppy,Pookey and Mark are correct in my opinion. I use a 500 f4 on a FF D700 and a crop D300, i have to up my shutter speed with the crop body mate.
 
Durbs said:
Woah... So it wasn't a simple question then :D

I knew it was a rule of thumb, but was curious about the technicalities of it all.

Thanks for the multiple, often colourful replies!
Click to expand...

It is a straightforward question. It's just that some people don't understand that there's a time for opinion and a time for technical facts. ;)
 
I've always wondered about this and never fully understood the reason why a crop sensor affects the recommended minimum shutter speed for a given focal length. Having read through the thread it now makes perfect sense - I was previously overlooking the fact the image produced by a crop sensor is blown-up to fill the frame, therefore making any motion-blur more apparent.

Thanks, that's another mystery ticked off the list. :)
 
What PMN said was entirely accurate and it's only by manipulating the question to talk about blowing an image up to equal print size that you could claim he wasn't correct. Viewing at 100% (all things being equal e.g. pixel density, focal length etc..) on screen there would be no difference.
 
James Hope said:
What PMN said was entirely accurate and it's only by manipulating the question to talk about blowing an image up to equal print size that you could claim he wasn't correct. Viewing at 100% (all things being equal e.g. pixel density, focal length etc..) on screen there would be no difference.
Click to expand...
But that's not how we view images, so we have to compare output at the same size. ie a print or even if you wish a screen image - at the same size.

It's you who is manipulating the question with your '100% view' because that's not something that happens in the real world.
A 6x4 print taken with each of the cameras in my post with a 50mm lens will show the camera movement as per my post - if we show all of those images at 100% we'll then need a discussion about viewing distances - you look like you're setting a standard size but really you're making it more complicated.
 
That's not true, I'm making it as simple as it can be. You like to talk about the maths and physics, well on the same screen at 100% there would be no difference whatsoever. Of course when you start blowing things up then things change, but to imply that I'm wrong because I'm not talking about the 'real world' is simply not true and I don't think you do yourself any favors by not making the distinction between what's physically happening and what the end result might be depending on how you choose to use the image.
 
But the end result is all that matters - forgive me, but that's the 'real world' it's not about a 100% view of the pixels, because that's not what my customers buy, or I print in an album or for a wall or even how we downsize for web view.

In all of those scenario's there's a different amount of magnification depending on the sensor or negative size.

'Real world' isn't an insult - it's where our pictures are actually viewed. So the crop sensor photo with the same pixel density has it's camera movement roughly doubled to produce the same sized output as a FF file. What they both look like at 100% is something only the photographer will ever see!

So if the photographer takes your advice, and is then unhappy when he gets his prints - what's your response to that?
'Oh you can't print them as big as you can the FF files, because your sensor is smaller. If you want them sharper you have to increase your shutter speed if you want prints - my 'correct' advice only holds for 100% viewing':nuts:
 
I'm sorry but you're being difficult here, there's the entire thread for the OP to understand what's going on, my point was that you shouldn't be telling people they're incorrect when they're not. I'm happy to agree that if you wanted to print the same size then of course that would make a difference. However by claiming someone else is wrong when they most certainly aren't only adds to confusion. The difference between crop and FF is so often misunderstood and there is nothing wrong with understanding the underlying principles as well as understanding how you would then use the resulting image afterwards.
 
I really don't understand what you're saying?

Because the whole thread consists of a debate between 2 completely different sides of a coin. So there is no obvious answer for the OP as you suggest.

Most people agree that you would need to increase the shutter speed as sensor sizes decrease, with a vocal few including you appearing to support the opposite view - yet you appear to agree that if output is considerd, then that's different. Notwithstanding you absolutely disagree with me :thinking:.

I'm glad it's clear to you, because I'm utterly clear on the original issue and your post has confused the hell out of me.:help:
 
Ok well I have to say again it feels like you're just being awkward. I'm not in any camp and I'm very clear about what I'm saying. I don't like getting drawn into these debates as it doesn't really serve any purpose so ill make this my last post. What was said earlier regarding viewing on screen at 100% was accurate and not incorrect as you claimed. If your desire is then to print the image and blow up to the same size as would be gained from a FF sensor then certainly you will notice a difference and it would be advantageous to increase shutter speed.
 
A good example of this...

I have a 5D (FF) and a G1 (MFT, x2 "crop") and I recently tried my Zuiko 135mm on both cameras and I very quickly decided that with the G1+135mm I needed a very fast shutter speed indeed :help: :D as I could see the camera shake through the VF.

If the two cameras had the same resolution I'm convinced that when viewing an image (for example an A4 sized print) a 5D shot - cropped to give the same FoV as a G1 shot and printed out at A4 - would show pretty much the same camera shake as a G1 shot printed at A4... although as the cameras are different sizes and weights and one has a flapping mirror and the other doesn't there's alwasy going to be slight differences

Anyway, when doing a real world comparison like this it's easy to see how smaller sensor cameras with decent resolution will show any camera shake with even standard to short tele lenses if the image is big enough to allow it to be seen.
 
James Hope said:
Ok well I have to say again it feels like you're just being awkward. I'm not in any camp and I'm very clear about what I'm saying. I don't like getting drawn into these debates as it doesn't really serve any purpose so ill make this my last post. What was said earlier regarding viewing on screen at 100% was accurate and not incorrect as you claimed. If your desire is then to print the image and blow up to the same size as would be gained from a FF sensor then certainly you will notice a difference and it would be advantageous to increase shutter speed.
Click to expand...
If you didn't want to get drawn into the debate, you could have ignored it, but you decided to 'agree' with someone else before actually changing your mind.

James Hope said:
What PMN said was entirely accurate and it's only by manipulating the question to talk about blowing an image up to equal print size that you could claim he wasn't correct. Viewing at 100% (all things being equal e.g. pixel density, focal length etc..) on screen there would be no difference.
Click to expand...

But PMN never admitted that fact, he stuck to his guns - and only mentioned 100% view once BTW, he claimed it was somehow lens dependant:wacky: And that although crop and FF accepted focal lengths as the same - it wouldn't hold true for other formats because 'they're different systems'.


Re-read his posts, he's not put this as scientifically as you appeared to think, and he definitely doesn't agree with what you just said about output.
 
Phil V said:
But PMN never admitted that fact, he stuck to his guns - and only mentioned 100% view once
Click to expand...

I only mentioned it once because all I was talking about is viewing digital images at 100%, i.e. the native resolution the camera is giving you. My view on this also stems from the fact that to me an image is either sharp or it isn't; there's no 'well it's sharp when viewed at 200 pixels wide'. It's blurry or it's sharp, it's that simple. The vast majority of my shooting has been with telephoto lenses on both APS-C and full frame bodies, I've used the 1/focal length guideline on thousands of frames on both cameras and never once compensated for the crop factor. Never once have I noticed a difference between the two. I don't claim to be the most scientifically minded photographer out there but I trust my experience and the results it's given me. If more technically knowledgeable people than me say I'm wrong then that's fine, I've learned something but my results still don't quite match up to what's considered to be 'technically correct'.

Another quick question though, as we're talking about output size (which I'd love to hear a technical response to). A 5Dmk1 has 12MP, my APS-C 7D has 18MP. Which image is 'physically bigger'?

My point with all this wasn't referring to printing; perhaps I could indeed have made that clearer and I apologise if not doing so has meant my point was less clear.
 
Last edited:
Then that is why you couldn't grasp why you were wrong. Despite your 'thousands of images' your actual test method is small numbers and flawed. As Richard posted earlier, when measuring sharpness, all stages of the process have to be taken into account. And the fact you are simply looking at 100% screen images is completely throwing your perception of the issue.

That's also why you have asked about the 7d and 5d, the 7d may have more pixels, and therefore create a higher res file at any given print size, but it'll still show more camera movement because of the magnification factor of the recorded image.

Think about the movement recorded at the sensor and how it gets magnified for output. At 100% it'll look the same on every size sensor, but as soon as you consider output sizes you're talking about magnification factors. And that is why you need to increase your shutter speed as sensor size reduces, because a tiny amount of recorded blur will be magnified more from smaller sensors and less from larger ones.
 
This thread is great. The answer was provided in the very first reply by Alan. The second reply then caused another 3 pages of 'discussion', it is almost like people are doing it deliberately isn't it?
 
ernesto said:
This thread is great. The answer was provided in the very first reply by Alan. The second reply then caused another 3 pages of 'discussion', it is almost like people are doing it deliberately isn't it?
Click to expand...

You'd think:bonk:
 
reminds me of my 10 x 8 contact prints of 35mm i used to do ,,,i'd have my magnifying glass on them and think yes thats the one ,,,,until i printed at 10 x 8 ,,,,,some of them didn't look quite so good at that size ,,,,that darn magnification ,if only i could make them bigger without it :LOL:
 
donutagain said:
reminds me of my 10 x 8 contact prints of 35mm i used to do ,,,i'd have my magnifying glass on them and think yes thats the one ,,,,until i printed at 10 x 8 ,,,,,some of them didn't look quite so good at that size ,,,,that darn magnification ,if only i could make them bigger without it :LOL:
Click to expand...

You could have....use a 10x8 camera. :love:
Ahh, those were the days.:D

As far as the original question goes, I know if I put a 200mm lens on my D700, I can get away with 1/250th sec, (no VR, and slightly shaky hands), but if I put it on my D7000, it would need to be 1/500th to ensure no shake, all other things being equal.
 
You must log in or register to reply here.
Back
Top