Beginner How does a crop sensor "magnify"

[Trekmeister]

Could somebody explain how a crop sensor works in comparison to a full frame sensor.Im not talking about the fact it has a smaller sensor and crops the image. I am asking how it is able to effectively provide a longer focal length on a lens on a crop sensor. I can understand a smaller image however does it cause a magnification of the image and if so how is that achieved?

 

[dean messenger]

l. a crop sensor has a tighter field of view so it does exactly what it says on the tin and crops the equivilant full frame image.
This means that a 50 mm lens used on a sensor with a 1.6X crop factor would produce the same field of view as a 1.6 x 50 = 80 mm lens on a 35 mm full frame sensor.
so your not actually zoomed. you just need a smaller length lens on a crop sensor to get the same shot on full frame due to the field of view on full frame being wider.

 

[Sam Tip]

It doesn't.

 

[Nod]

Wiki explains it very well... http://en.wikipedia.org/wiki/Crop_factor

 

[Teflon-Mike]

It doesn't. whatever you view the picture on does.
OK.. a 'Crop'; grow a field of corn then cut it down; what you have cut down and keep is your crop. So, if you took a photo, and 'cut it down' sliced some off the top bottom and sides to make it smaller.. THAT is your 'Crop' or 'Crop Section', and the resultant field of view as the 'same' as if you had used a longer lens than you did.. only you didn't. and you haven't 'Magnified' your image, you've made it smaller.
Now, a Camera lens 'shrinks' a scene you see down to an image small enough to fit in the box. To view that image you have to 'enlarge' it again, to whatever size you want to look at it.
Here's the mind-warp.. longer lenses don't magnify the image, they shrink it, and you get more 'zoom' simply from them shrinking less of the scene, into the box.
"Magnification" comes when you take whatever the camera put in the box and 'enlarge' it again for viewing...
Lets do this in film... and take a photo with a full-frame 35mm film camera and standard 50mm lens. Develop film, pop negative into an 'enlarger' and blow it up to... 5x8" to make a print.
Image is a bit cluttered and we only really want the middle, and rue not having a longer lens to tale our original photo with.. so we crop it down to get the bit we want.. only its now rather small... Oh-Kay..... get another bit of paper, pop it under the enlarger, and crank the head up to get more 'enlargement', and we could print the full-frame of our original negative say, 10x8".. only we don't have a piece of paper that big... A-N-D we'd only chop it down after anyway.. so lets stick the 5x8" paper under the head and 'just' get the bit we'd 'crop out'.... what you get is a 'sectional enlargement'.. same 'Print' size but, the bit of the original image we've printed has the same field of view as if we'd taken a full-frame image with a 75mm lens.
IF we had used a 75mm lens to start with, and printed the full-frame, we'd have got the same field of view.. but that lenses wouldn't have provided any more 'magnification'... remember longer lenses simply 'shrink' less scene, and when we made the print, we'd have used less actual 'magnification' from the enlarger to make it.
So 'Cropping' gets this effective 'magnification' simply from over-enlarging the captured image to the SAME viewing size.
Now, in the example we cropped a full captured image down.... and 'wasted' image on the negative - what if we didn't put so much film in the camera to begin with?
Not commonly done with 35mm, but is often done with medium format cameras that take 6cm wide 120 roll film, is to put 35mm through them instead... NOW the camera is still projecting 'scene' over the full 4x4cm 'trap' in the camera, but like putting a smaller piece of paper under the enlarger, its spilling over the edges of the 35mm film...
Same lens on the camera, less 'scene' caught on celluloid, 'like' using a longer lens... magnification effect obtained simply by enlarging that smaller 'section' to the same viewing size.
No scissors required to physically 'crop' anything, whether its still actually a 'crop' is probably a matter of semantics; which is where confusion comes from; because the term 'crop' is being used in this sense to describe how a smaller sensor is producing images 'equivalent' to ones made with a larger sensor and then 'cropped'.. which they aren't.. the smaller sensor camera is making a full-frame image.. just not one as big as the full frame image a larger sensor would!
Before digital, we didn't talk of the 'crop factor' of 110 cartridge cameras that has a 13x17mm 'frame-size', or the crop factor of 35mm cameras that usually have a 24x36mm 'frame-size' or the crop factor of Medium format cameras that could have 'frame size's from 6x4.5cm to perhaps 6.9cm.. whatever size your 'sensor (or film trap!)', it IS 'Full-Frame' .. for THAT camera!
Interesting to note, that while the 'frame trap' in a 35mm film camera is commonly 24x36mm, the film was stolen from movie cameras that ran it through the camera vertically, and had a trap 16x24mm.. (same size, incidentally as a contemporary 'crop-sensor') and to get twice the pictures on a roll, some still camera makers did the same; notably Olympus with their "Half-Frame" Pen series among others....which again, is a bit of a mis-noma, as that was the 'full frame' intended to be put on that format of film; and the 24x36 size by that convention ought to have been named 'Double Frame'.. but, either way, the frame size is the frame size and if the IMAGE the lens puts on it covers the FULL-FRAME.. its a full-frame! Like a lot in photography, its dogged by common-mis-use of terminology! But that's another debate.
Back up to first principles and summerising
1/ A camera lens isn't magnifying a scene its shrinking it.
2/ Longer lenses then don't produce 'more' enlargement, they merely produce 'less' shrinkage.
3/ Captured images are 'Magnified' post capture to viewing size.
4/ The 'Key' then is in the angle of view we obtain, from a Lens AND Camera combination
5/ The focal length of a lens is a single 'specification', from which the angle of view it should give, on any particular 'frame size' MAY be inferred; but it is NOT 'the' angle of view.
6/ Camera Talk suffers hugely from jargonese!

 

[woof woof]

I think I can explain using very short sentences...

What you see is a change in the field of view.
Imagine a photo taken with a 35mm camera fitted with a 50mm lens.
Cut a section out of the middle of the picture.
That's it.
That's what a crop is.
So a 50mm lens ends up giving a much tighter field of view when you mount it in front of a smaller sensor.

The equivalent crop factor to FF 35mm is...
APS-C Nikon = x1.5.
APS-C Canon = x1.6.
Micro Four Thirds = x2.

And out of interest...

A 50mm lens on a 35mm camera...
= 75 or 80mm field of view on Nikon or Canon APS-C.
= 100mm field of view on Micro Four Thirds.

To get the same field of view as a 35mm camera with a 50mm lens will give...
= 33 or 31mm Nikon / Canon APS-C.
= 25mm Micro Four Thirds.

 

[Nawty]

And the magnification then occurs when you take that smaller section and turn it into a normal sized picture.

edit:

And in fact, that's how all cameras magnify and you need less magnification to make a full frame capture into a (for example) 5x7 print than you do a M4/3 capture.

This is why medium and large format have even better image quality than so called full frame - you have to magnify them less to get a useful size print.

 

[Trekmeister]

edit:

And in fact, that's how all cameras magnify and you need less magnification to make a full frame capture into a (for example) 5x7 print than you do a M4/3 capture.

This is why medium and large format have even better image quality than so called full frame - you have to magnify them less to get a useful size print.

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

 

[AJQS]

The simplest way is perhaps to print a picture at A4. Now put it in an A5 frame. That's more or less the difference between FF and APS.

 

[weybourne]

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

I think you are trying to understand too much at once - and unwittingly are introducing other factors - such as pixel count of sensor and dot per inch of printer (not to mention circle of confusion). Try to think only of one size of printed photograph, how a lens makes an image smaller to fit on a sensor and how a smaller sensor means that less of the image produced by that same lens is captured. Look at the wikipedia link posted by Nod for illustrations of this. When you have got your head around that, then start to think about circle of confusion, ppi and dpi.

 

[Nawty]

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

If I understand you correctly I could maybe explain it thus:

Forget about the sensor resolution. For any given lens the more you magnify the image through the lens the harder the lens has to work to make the image sharp as you also magnify the microscopic imperfections in the lens. You can show this easily yourself - forget about the sensor - when you zoom into an image to 100% the image doesn't look as sharp/distinct. It's the same effect.

So if you have less magnification (FF) then the image has better quality, if you magnify the image (DX) then the quality drops slightly - all else being equal.

The problem is, all else is rarely equal but that's a different matter.

 

[StephenM]

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

Yes and no. Let's pluck an imaginary figure out of the air and say that your image resolves 100 lines per millimeter. Now if your image is 3/4in x 1in to make a 6x4 print you magnify it 6 times, so the resolution is now 100/6 lines per millimeter. If you image is 1in x 1.5in, the magnification is 4 times, and the resolution in the print 100/4 - more detail. The "no" part comes in when we consider how much detail can be reproduced in a print. If the print can't manage the smaller resolution of 17 lines, then there's no advantage; if it can, then there is an advantage to the larger sensor. Print at A3 or A2 and the advantage of more detail becomes clearer (pun intended). And if your lens can't keep up, then again, not a lot of advantage.

Above a certain magnification, the print resolution matters less and the image more; for small prints, the image from the camera will "outperform" the print.

Hence, print size matters when you consider whether a full frame (or larger) sensor is needed.

I could go in to details on typical resolutions, visual acuity and related things, but that would only muddy the waters.

 

[woof woof]

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

With smaller formats you have to blow the image up more to get the same sized picture you'd get with a larger format and blowing the picture up more can reveal any shortcomings. This is why lenses used on smaller systems are often designed to be...better, sharper than lenses mode for larger format systems, so that after you've blown the smaller image up more you still end up with a good print.

After doing comparison prints between "FF," APS-C and MFT I decided that (mostly) format size doesn't matter all that much to me and so I (mostly) forget about it and on the subject of what's good enough and what maybe could be forgotten about...

this piece may interest you and there's a nice little vid...

https://luminous-landscape.com/the-mirrorless-revolution/

And this is good too...

View: https://www.youtube.com/watch?v=PHYidejT3KY#t=45

 

[Teflon-Mike]

What I have been struggling with; does a crop sensor "expand/magnify the image more meaning less resolution than a full frame when viewed/printed. For instance 2 same size prints; crop and FF, the crop has captured a smaller area yet produces a print the same size as the FF although the FF has more area coverage (hope you understand my point?)

utterly accademic.
Again, There are so many specifics that are just not considered that any generality you try and make will be so utterly nebulous as to be ridiculous
.the bigger sensor on a full frame camera is just that, a bigger sensor.
The only significant thing that changes generically with the focal length of the lens is the depth of focus.
The shorter the focal length, the more depth of focus you get around a subject, for the same camera to subject distance.
Therefore, if you have a smaller sensor, that provides a similar angle of view with a shorter focal length, the less easily you will be able to exploit 'selective focus' techniques to de-focus a back-ground with larger apertures; conversely it will be less sensitive of 'critical' focusing.
As I said earlier, a camera lens doesn't magnify an image, it shrinks it, and longer lenses shrink it 'less'.
If it didn't 'shrink' the image, it wouldn't be a lens, it would be a bit of window-glass..... SO, larger format cameras, MIGHT gain greater 'image quality' from any number of things, imparted by the 'scale'; It's easier & to make window-glass than a pair of spectacle lenses..... There's no complex shape to form that might introduce distortions, and its large, so any imperfections there may be will be a small % of overall dimension; so you increase the margin of accuracy, in a number of ways; This then leads on, and a manufacturer can exploit that to make a camera more cheaply, or with better optical performance.. which is up to them, and the SPECIFIC camera and lens combination... NOT the frame-size.
And so the debate develops, and will we discover more questions lurking behind the initial enquirery.... ?

 

[Orville]

Imagine that both cameras are wearing the same full-frame lens. The lens will pass exactly the same image into each camera body, but the full-frame sensor (red box) will capture more of it than the crop sensor (yellow box). Crop sensors do not magnify an image, rather they simply do not capture as much of it due to their smaller size. The final image looks 1.5x (or 1.6x) closer than on full frame, providing the illusion of magnification.

 

[mrgubby]

 

[FlyingShrapnel]

Imagine that both cameras are wearing the same full-frame lens. The lens will pass exactly the same image into each camera body, but the full-frame sensor (red box) will capture more of it than the crop sensor (yellow box). Crop sensors do not magnify an image, rather they simply do not capture as much of it due to their smaller size. The final image looks 1.5x (or 1.6x) closer than on full frame, providing the illusion of magnification.

these, it is much simpler to come to grips with the whole "crop" sensor thing when shown graphically

 

[Trekmeister]

I think where I have been going wrong is that I was under the impression that, the crop sensor is smaller therefore the print size is smaller and has to be stretched/magnified somehow in order to make it the same size as the FF which in turn would degrade the image quality on the crop.

 

[Ed Sutton]

Forget all the technical mumbo jumbo and imagine what you see on the screen on the back of a camera.

Put a fixed focal length lens on a FF camera and the same lens on a crop sensor camera. Point the two at the same thing from the same place and switch to live view. The picture on the screen of the crop camera will look like a magnified view of the central portion of the picture on the screen of the FF camera. because that's what it is, effectively.

If you zoom in on the FF screen the pictures will match - except the zoomed in FF picture will be of lower quality given equal numbers of pixels on both sensors.

It's at the output stage (screen or print) that the magnification appears, not at the taking stage.

 

[andyscott]

it doesn't really matter its just a different point of view.

the FF vs crop argument, is a bit like arguing different religions are better. if it works for you then great, just use it and enjoy it.

 

[StephenM]

I think where I have been going wrong is that I was under the impression that, the crop sensor is smaller therefore the print size is smaller and has to be stretched/magnified somehow in order to make it the same size as the FF which in turn would degrade the image quality on the crop.

Actually, that's a very succinct explanation of exactly what a crop sensor is. Smaller. Hence, for the same degree of magnification as a full frame sensor, the print is smaller. And to get the same size of print, you need to magnify more, enlarging the imperfections.

 

[Trekmeister]

Actually, that's a very succinct explanation of exactly what a crop sensor is. Smaller. Hence, for the same degree of magnification as a full frame sensor, the print is smaller. And to get the same size of print, you need to magnify more, enlarging the imperfections.

So what you are saying is that the crop sensor output image (print) will be of a lower quality (same lens. camera pixel count etc) than an FF output print or am I reading it all wrong what you have said above?

 

[Phil V]

So what you are saying is that the crop sensor output image (print) will be of a lower quality (same lens. camera pixel count etc) than an FF output print or am I reading it all wrong what you have said above?

That's right, in principal. It's the 'all things being equal' part that can throw in the curve ball.

The best lens on the best crop sensor will outperform a crap lens on an old FF camera.

But really you've got the gist of it.

 

[woof woof]

So what you are saying is that the crop sensor output image (print) will be of a lower quality (same lens. camera pixel count etc) than an FF output print or am I reading it all wrong what you have said above?

But how much will all this theory matter in reality?

Most people these days only seem to view images on computers, tablets and phones and when doing this it's (arguably) pretty impossible to detect any significant differences between pictures taken with different camera.

IMVHO you'll only see these potential image quality differences in extreme situations such as...

When printing VERY big.
When doing severe crops and printing big.
When examining pictures at very high magnification on screen maybe at 100%+
When shooting at very high ISO's.

Again IMVHO I think that once you get to sensor sizes of MFT or maybe even a bit smaller the image quality differences to 35mm equivalent "FF" will for many people and in many situation pretty much insignificant.

 

[SsSsSsSsSnake]

Alan thats a fair point because viewing different photos say on flikr not knowing when first viewing what camera/lens was responsible
Ive been shocked at times when viewing a photo that looked wonderful re colour sharpness Etc yet when after looking at the info re camera /lens was say an iphone or old point and shoot say.
Now im not saying im an expert at viewing photos and a more experienced viewer might notice more ,by the way this is on a 27 imac.
Anyway just agreeing generally with your point.
Im reminded of a friend who works at a museum and one well know photographer(dont ask me his name) came to give a talk,he brought one of those £5 throw away cameras and my friend said the Pictures he took were exceptional.so makes one wonder for the non pro who doesnt print big is it really worth considering getting worked up about new cameras and lens's etc.

 

[woof woof]

I think that processing to make the pictures look good can't be overlooked. Whenever I post my pictures here via Photobucket they never look anywhere near as sharp as they do on my pc so some people must be thinking about how their pictures are going to be piped through to the internet and viewed and processing them for the best end result.

I just process mine for how they look on my screen and print now and again and I when sharing in electronic format or posting them on a forum all I do is reduce the size but there must be more to it

 

[Ajf350d]

Interesting conversation of a subject I still fail to grasp in its entirey, although I understand the basics.
However, as in the example above if you do have say two 16mp sensors, one FF and the other a 1.6x crop, why would the 1.6x crop in print produce a lower quality image simply by being a 1.6x crop sensor?
They still have the same image resolution so should still be able to produce equivelant ppi.
In fact, if you wanted identical prints, wouldn't you actually have to addiitonally crop the FF image to match the 1.6x crop sensor ouptut, which would reduce the overal image resolution of the FF image?

I understand larger sensors give better quality, but I was under the impression this was due to larger pixels etc which is slightly outside this discussion?
Or is that in fact the whole point?

 

[lakeland]

In newbie terms from a newbie,please correct me if I'm wrong.

If you take two pics, one on a crop sensor and one on a full frame, both with the same lens, the full frame would have more image to the top, bottom, left and right.

If you were to crop the full frame image to make it match the crop frame image, you'd lose resolution.

Therefore to get the same resolution on a full frame you need to zoom in further so you effectively crop the image optically and don't lose resolution.

 

[StephenM]

There's more than one subject here, although they often seem to become intertwined and confused.

1. Sensor size. If you use the same len, it will produce the same image no matter what sensor you put behind it. But a smaller sensor will only be able to capture the image from the middle of the area that will be captured by the larger sensor. The smaller sensor will therefore give a smaller angle of view than the larger one, given the same lens.

2. Degree of enlargement. Lenses are not perfect (even if we assume sensors are!). That's why a simple plastic one element lens gives superb results with a box camera and large negative because no enlargement is needed. As you enlarge more, the imperfections are magnified. Hence the old saying in film days "a good big un will always beat a good little un". It's exactly the same with sensors.

BUT - if you don't enlarge enough to see the differences, you won't see them. Not quite tautological - the unenlarged contact print from a box camera could still take a small enlargement and look perfect. So, if you don't push a crop sensor too far, you won't see a difference between it and a larger sensor. (In passing - why "crop sensor"? Smaller than a 35mm film frame, yes, but why "crop"?)

As to whether larger sensors have a better ability to hold detail across a large subject brightness range etc., a lot depends on the sensor and software design rather than the size or megapixel count.

 

[chris malcolm]

All the confusions arise from oversimplifications. Anyone who starts off saying "it's quite simple really" is going to go on to explain one of the simple elementary aspects of the situation which in fact have important exceptions and qualifications some of which are imoprtant enough to turn any obvious looking deduction on its head.

The most annoying simplification is the completely unnecessary one of crop factor changing the effective focal length of the lens. It doesn't. The reason people think it does is that people got used to equating the view angle of a lens with a certain focal length on a 35mm film camera, e.g. the old chestnut that a 50mm lens (on a 35mm film camera) gives the same field of view as the human eye. I can't imagine how anyone who has ever owned both a 35 film camera or full frame DSLR at the same time as a working human eye or two could ever have fallen for that one. Just look at something with your eye, and then look through the allegedly "normal field of view" lens at the same scene. The idea is so plainly absurd in so many respects.

Since crop sensor DSLRs can use lenses from 35mm film cameras the helpful oversimplifiying salesman tells us that the crop sensor extends the focal length by around 50%. One misleading oversimplification piled on another.

 

[StephenM]

Surely everyone new to photography finds it much simpler to have the angle of view on the camera they own compared with the angle of view on a camera type they've never encountered? And learn how to convert from a system they've never used to the only one they know (and back again)? In the old days we had to really, really struggle with the difficult mathematical concepts of not only knowing what the "standard" focal length for our selected format was, but being able to handle (in our heads) advanced mathematical concepts like multiplication to work out how much bigger the image would be if we doubled the focal length. Not everyone can be expected to be able to handle multiplication by 2 in their heads. Or so marketing types believe.

I put it down (more seriously) to two things:
a) bigger is better. Small sensors need shorter focal lengths, which is a smaller number and less impressive.
b) it enables digital compacts to look better - how would "massive zoom range from 3mm to 9mm" look? It also helps people to overlook how small some of these sensors really are.

 

[Nawty]

Athe old chestnut that a 50mm lens (on a 35mm film camera) gives the same field of view as the human eye. I can't imagine how anyone who has ever owned both a 35 film camera or full frame DSLR at the same time as a working human eye or two could ever have fallen for that one. Just look at something with your eye, and then look through the allegedly "normal field of view" lens at the same scene. The idea is so plainly absurd in so many respects.

No one (correctly) says it does, it gives a natural PERSPECTIVE similar to how we judge distance. This is so much the truth that law courts have established it to be the focal length for evidential photos where distance is important (RTA's for example).

 

[HoppyUK]

<snip>...if you do have say two 16mp sensors, one FF and the other a 1.6x crop, why would the 1.6x crop in print produce a lower quality image simply by being a 1.6x crop sensor?
They still have the same image resolution so should still be able to produce equivelant ppi. <snip>

There is one major reason. Larger formats don't work the lens so hard, so they are able to deliver better sharpness.

What we call 'sharpness' has two aspects - resolution (the fineness of detail) and contrast (how clearly those details are shown). Fact of physics is that when lenses are asked for deliver greater resolution (ie a smaller sensor) image contrast goes down, and it is actually image contrast that conveys the greater visual impression of sharpness - despite our obsession with squillions of pixels!

It's like a car that accelerates from 0-60 in six seconds, but takes a lot longer to go from 60-120. Same with lenses - the more you ask, the harder it gets optically. It's called Modulation Transfer Function (MTF) that is shown in those squiggly graphs lens manufacturers produce, like this one for a Canon 50mm f/1.4 http://www.usa.canon.com/cusa/consumer/products/cameras/ef_lens_lineup/ef_50mm_f_1_4_usm Ignore the colours and the dashes, just look at the thicker lines that are at 10-lines-per-mm and then the thinner ones at 30-lpmm - the thinner ones are always lower down the chart. Contrast is lower, so the image looks less sharp. In this case, up the left-hand side at the top, contrast is at about 97% at 10-lpmm (thicker lines) but at 30-lpmm (thinner lines) it drops to around 86%. And the higher resolution goes, the greater the gap becomes.

 

[chris malcolm]

No one (correctly) says it does, it gives a natural PERSPECTIVE similar to how we judge distance. This is so much the truth that law courts have established it to be the focal length for evidential photos where distance is important (RTA's for example).

That's because it gives about the same perspective when looking at an A4 print at a distance of around a foot as we'd get when looking through an A4 window a foot away. The so called perspective distortion of a wide angle lend isn't a weird optical property of the lens, it's simply due to the fact that we look at wide angle photographs from the same comfortable reading distance, which creates a disparity between lens angle of view and eye angle of view. Move the eye closer and the wide angle distortion disappears.

 

[Ajf350d]

There is one major reason. Larger formats don't work the lens so hard, so they are able to deliver better sharpness.

What we call 'sharpness' has two aspects - resolution (the fineness of detail) and contrast (how clearly those details are shown). Fact of physics is that when lenses are asked for deliver greater resolution (ie a smaller sensor) image contrast goes down, and it is actually image contrast that conveys the greater visual impression of sharpness - despite our obsession with squillions of pixels!

It's like a car that accelerates from 0-60 in six seconds, but takes a lot longer to go from 60-120. Same with lenses - the more you ask, the harder it gets optically. It's called Modulation Transfer Function (MTF) that is shown in those squiggly graphs lens manufacturers produce, like this one for a Canon 50mm f/1.4 http://www.usa.canon.com/cusa/consumer/products/cameras/ef_lens_lineup/ef_50mm_f_1_4_usm Ignore the colours and the dashes, just look at the thicker lines that are at 10-lines-per-mm and then the thinner ones at 30-lpmm - the thinner ones are always lower down the chart. Contrast is lower, so the image looks less sharp. In this case, up the left-hand side at the top, contrast is at about 97% at 10-lpmm (thicker lines) but at 30-lpmm (thinner lines) it drops to around 86%. And the higher resolution goes, the greater the gap becomes.

Ah, right. i understand now.
Does that apply to all lenses? For example for M4/3 lenses are specifically designed for that format so the whole of the lens is 'used' as opposed to a FF lens on a crop?

As said, at the end of the day in most cicumstances it probably makes no, or little, difference, but I am just curious from thr technical side. i have tended to simply accept it works mostly, but sometimes interesting to stand back and look at the why!

 

[woof woof]

Some small sensor lenses are excellent.

I used to use old film era lenses on my MFT cameras and I still do on my FF Sony A7 and the results do look different. On my A7 the lenses seem to perform well but less so on my MFT cameras. I once took the same shot with my MFT Panasonic G1 and what I thought was a very good film era 24mm and then a modern MFT Voigtlander 25mm and when looking at the results at high magnification on screen the new lens just simply outclassed the "good" old lens despite only the best centre portion of the old lens being used.

The excellent nature of some of these smaller system lenses will no doubt go a long way towards mitigating the negative effect of blowing a smaller image up more.

 

[HoppyUK]

Ah, right. i understand now.
Does that apply to all lenses? For example for M4/3 lenses are specifically designed for that format so the whole of the lens is 'used' as opposed to a FF lens on a crop?

As said, at the end of the day in most cicumstances it probably makes no, or little, difference, but I am just curious from thr technical side. i have tended to simply accept it works mostly, but sometimes interesting to stand back and look at the why!

Yes, the principles of Modulation Transfer Function apply to all lenses, always. That is, higher resolution demands mean lower contrast and reduced sharpness. It's directly related to the crop factor, so let's take a resolution level of say 30-lines-per-mm on full-frame - a good standard, roughly equating to an eye-lash or strand of hair in a head and shoulders portrait. On a crop-format camera, that translates to 45-48-lpmm at sensor level (1.5x or 1.6x crop factor) and to 60-lpmm on M4/3 (2x crop factor). Those are big, significant, and noticeable differences.

So bigger always means better, all things being equal, but it's not an entirely one-sided contest. It's much easier to design lenses for smaller formats, and that gives more freedom to make them sharper. Some lenses for M4/3 for example, if they could be scaled up to cover full-frame, would be the sharpest lenses out there, bar none. Or designers can use that freedom in other ways, to extend zoom range (just look at the massive megazooms that some bridge cameras have), or to run faster apertures, or to make lenses smaller/lighter/cheaper.

There's another way of looking at it too. Sure, bigger is better and that will always be true, but we've now reached the point where the differences are hard to see in most people's everyday on-screen output. So what's the point in spending all that extra money and effort on full-frame when APS-C or M4/3 is plenty 'good enough'?

Edit: crossed post with Alan WW.

 

[woof woof]

There's another way of looking at it too. Sure, bigger is better and that will always be true, but we've now reached the point where the differences are hard to see in most people's everyday on-screen output. So what's the point in spending all that extra money and effort on full-frame when APS-C or M4/3 is plenty 'good enough'?

A bigger format may always give a better image quality but smaller can have real world advantages as long as you don't need to make a very large image or shoot at very high ISO settings. For example one big plus point I see for smaller systems is that you can get deeper depth of field for the same settings or indeed the same depth of field at a wider aperture and with a faster shutter speed,

I very rarely use a tripod and I much prefer hand held shooting and for this MFT and APS-C have advantages for me. For example if I shoot with a FF camera at f8 I could get the same field of view and depth of field with a MFT camera at f4 or APS-C at f5.6 and these settings, especially for MFT, would give a faster shutter speed and less chance of camera or subject movement causing blur.

For hand held shooting I think that the sweet sport for me is more towards MFT or APS-C than FF once everything is considered including the bulk and weight of the kit.

 

[HoppyUK]

A bigger format may always give a better image quality but smaller can have real world advantages as long as you don't need to make a very large image or shoot at very high ISO settings. For example one big plus point I see for smaller systems is that you can get deeper depth of field for the same settings or indeed the same depth of field at a wider aperture and with a faster shutter speed,

I very rarely use a tripod and I much prefer hand held shooting and for this MFT and APS-C have advantages for me. For example if I shoot with a FF camera at f8 I could get the same field of view and depth of field with a MFT camera at f4 or APS-C at f5.6 and these settings, especially for MFT, would give a faster shutter speed and less chance of camera or subject movement causing blur.

For hand held shooting I think that the sweet sport for me is more towards MFT or APS-C than FF once everything is considered including the bulk and weight of the kit.

That's taking the debate beyond the question of sharpness and into 'equivalence'.

There are pros and cons between formats depending on your point of view and application, but equivalence cuts both ways. In your example, with M4/3 you are getting a higher shutter speed with greater depth-of-field, but at the same time on the full-frame camera you would be able to raise the ISO two stops to get that same shutter speed and DoF with the same level of noise due to the larger format's 4x advantage in sensor area. So you end up the same.

Good article on equivalence http://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care

 

[woof woof]

Yes, I've gone beyond the original question but maybe into areas that perhaps make more real world difference for most people.

On your point of FF ending up the same once you've stopped the lens down and upped the ISO I'd argue that it's not necessarily so as eventually you run out of settings to ramp up, for example my GX7 and A7 both max out at ISO 25,600, and when ISO and aperture are maxed out a smaller system will still give you more depth of field for the same field of view (and this may matter to you) and you may still have the advantage of a lighter and less bulky system plus the smaller systems lens could well be sharper wide open than the larger systems lens. All of my MFT primes are relatively sharp wide open when compared to the Canon fit FF and APS-C DSLR primes I had.

All of these theories, blogs and internet reviews are nice and interesting but I think that the real light bulb moments come when you test and compare these systems for yourself and how you shoot and view and for me the image quality differences are mostly of little importance if looking at MFT, APS-C and FF.

But maybe these are debates for another thread.