4/3 vs APS-C sensor size

[stevesanacore]

Well I'm ready to go out and finally buy my next compact camera and this discussion seems to not have helped at all  

I think have it narrowed down to either the Panasonic GX-1, Olympus Penn, or the Sony Nex-7

Yes the OM-D is a winner but it seems much bigger and bulkier and defeats the purpose of a very low profile - consumer looking - camera. It would seem to me that the lens choices are paramount to the comparison. My first choice for a lens would be a 24-28mm equivalent. Then a 35 and 50. Any opinions on the best optics for these? I have no interest in the bulky zooms - I want to only stick with the smallest glass I can find- maybe M mounts? I also have no need for autofocus. Optical quality and bulk are more important.

Thanks for any opinions.

[Ken Bennett]

Why the GX-1 or the Pen series? Will you use it held out in front, or will you purchase the optional eye-level EVF and hold it up to your face? If so, you lose quite a bit of the sleek shape and low profile of these cameras.

Having used most of the Panny and Olympus m4/3 cameras, I much prefer the built in viewfinder of my GH2 (and I expect I would like the OMD.) For lenses, I like the Panasonic 14/2.5, the Panasonic 20/1.7 (though many love the 25/1.4), and the Olympus 45/1.8. All are terrific lenses. Some folks like the Oly 12/2, which is 24mm-equivalent and fast, though larger and heavier than the tiny Panny 14. For now I use an old Tamron 90 macro for a long lens.

If I had to do it over again, I would buy the Panasonic 7-14/4 zoom for my wide angle, and use it with the 20 and 45. Next on my list is the Panasonic 100-300 zoom, which also gets high marks.

The reality is that any of these cameras would be a fine choice. Pick one and start shooting, and just ignore all the talk of new cameras on the market for several years. Good luck.

[stevesanacore]

I'm not crazy about the miniature SLR shape of the OMD or the GF-2 - I like the smaller slimmer bodies. The Sony has the benefit of a digital finder but without the shape of the slr. Yes the 12 f2 would be one of my chosen lenses if I went with olympus. I with the Sony had a better selection of lenses.

[Chris Pollock]

I bought a Lumix GF1 not long after it came out, along with the Panasonic 20mm F1.7, 7-14mm F4, and 14-45mm F3.5-5.6 lenses.

I was reasonably happy with it, but found that the lack of an EVF was a bigger pain than I expected. It's awkward to hold the camera in front of me, and hard to keep it steady. I looked at the optional attachable EVF, but the the picture quality was just too poor for me to part with my money.

The lack of in-camera image stabilization was also a limitation. The 20mm F1.7 is a terrific lens, but the lack of image stabilization and the difficulty of holding the camera steady somewhat limited its low light potential.

This Monday I yielded to temptation and upgraded to an Olympus OM-D E-M5. My experience so far is limited to a few low light shots and a lunch time stroll around central Sydney, but my initial impressions are very favourable. Here is a quick summary of my main observations, if they're of any interest to anyone:

Shooting with the EVF is vastly superior to using the rear display. I would go so far as to recommend avoiding any camera without an EVF or a reflex mirror.

The electronic level in the viewfinder is great. I hate even slightly crooked photos, but I'm not good at judging when the camera is level. My only complaint is that the level could be a little more sensitive. It seems to let you tilt the camera a few degrees before it indicates that it's not level.

I haven't had a chance to properly test the in-camera image stabilization, but it promises to be a killer feature in combination with fast optics.

The metering definitely tends to underexpose slightly, but it's very consistent - large areas of shadow or sky didn't seem to confuse it. In most cases I think 1/3 or even 2/3 of a stop of exposure compensation would be a good idea.

The image quality is good for the sensor size. With default settings in Lightroom I saw some noise even at ISO 200, but it has an inoffensive quality, like film grain. I expect it would be unnoticable at realistic print sizes, or easily removed with a few adjustments in Lightroom.

The 14-45mm lens did pretty well considering its low price, but it wasn't really worthy of the sensor. I've ordered an Olympus 45mm F1.8 and a Panasonic 25mm F1.4. Combined with my 7-14mm and 20mm they should make a great team. The Olympus 75mm F1.8 looks interesting, but pricey. I considered the 12-35mm F2.8, but decided to get the cheaper, faster, and probably sharper 25mm F1.4 instead.

[Chairman Bill]

Just to add to the confusion (there is confusion, isn't there?), the Canon G1X has a sensor slightly larger than the 4/3, but of course, smaller than APS-C. I've been considering it as an alternative to the Fuji X100 (APS sensor) & some of the various 4/3 offerings.

I want something lighter & smaller than my FF DSLR, no need for interchangeable lenses (and a fixed 35mm equivalent as on the X100 would do), don't want to be holding the camera at arm's length (so viewfinder please), want good ergonomics, and image quality is paramount.

It'll be a travel camera as much as anything else, so landscapes, city-scapes, 'street', family holiday snaps/portraits etc.

Thoughts on the merits of the Canon sensor could be enlightening & useful

[AJSJones]

People stating that DOF doesn't get altered with format, just don't understand the definition of DOF.

Regards

Indeed.
DoF only "exists" when a print (screen) size is defined and the image is viewed from a defined distance by a human who perceives which parts are "sharp".  If those conditions are not met, DoF remains merely a qualitative concept.

To compare DoFs from different situations, the parameters that are fixed and those that are changed need to be stated.  For example, if there is only a change in the viewing distance, we can say the DoF changes. 

To compare the same image using different formats, the FL must be changed to compensate for the difference in AoV from the sensors.  To use the same lens from the same camera position means the sensors will record different images. Why would one want to compare the DoF of two different images?  What would we learn?

To the question: I am happy with my Lumix GH2, the 7-14, the Leica 25, the 14-140 (walkabout and video) and the 100-300 for wildlife.  The more recent  Olympus sensor appears to have some advantages over the Panasonic, by all accounts, so I will evaluate a body change later.  The whole kit is so much lighter and easier to travel with and produces some great images.  Can I print them as large as those from my 5D2 - of course not, but the trade-off is one I am willing to take under some/many circumstances (I have yet to stitch some panos I shot with the 25 but I'm optimistic  ).

[MrIconoclast]

I have no problem making very good 16x20 prints from my OLY PL5.  I do use PRIME lenses most of the time.  The Oly 45mm f/1.8 and a Panasonic 14mm f/2.5. I think being able to carry around several very small and light prime lenses is the secret.  They beat out all but the best of the full-frame zooms while being being far lighter and somewhat faster.   

This above combo sure beats carrying around my old DSLR with it's bulky zoom lens.

[Ray]

It seems we are falling into the typical confusion of comparing the technical qualities of images of different scenes.

If one uses the same lens at the same F/stop with different size sensors, one gets different images. It doesn't make sense to compare the 'depth of field' in two images when the field itself is different.

Using the same lens and F/stop, the larger sensor produces a wider field. If that wider field is cropped to the same width as the image from the smaller sensor, the DoF will be the same, excluding minor variations due to differences in sensor pixel density and sensor resolution.

We experienced a similar confusion some time ago on this forum in relation to perspective. There were those who adamantly argued that focal length of lens has nothing to do with perspective. It's only the position that counts. To demonstrate this misconception, such people would present the argument that cropping a shot taken with a wide-angle lens to the same field of view as a shot taken from the same position using a telephoto lens, resulted in images that appeared to have the same perspective, thus demonstrating the principle that cropping an image is effectively the same as increasing the focal length of lens and that two images shot from the same position, using 'effectively' the same focal length of lens, will have the same perspective in all respects.

[elf]

It seems we are falling into the typical confusion of comparing the technical qualities of images of different scenes.

If one uses the same lens at the same F/stop with different size sensors, one gets different images. It doesn't make sense to compare the 'depth of field' in two images when the field itself is different.

Using the same lens and F/stop, the larger sensor produces a wider field. If that wider field is cropped to the same width as the image from the smaller sensor, the DoF will be the same, excluding minor variations due to differences in sensor pixel density and sensor resolution.

We experienced a similar confusion some time ago on this forum in relation to perspective. There were those who adamantly argued that focal length of lens has nothing to do with perspective. It's only the position that counts. To demonstrate this misconception, such people would present the argument that cropping a shot taken with a wide-angle lens to the same field of view as a shot taken from the same position using a telephoto lens, resulted in images that appeared to have the same perspective, thus demonstrating the principle that cropping an image is effectively the same as increasing the focal length of lens and that two images shot from the same position, using 'effectively' the same focal length of lens, will have the same perspective in all respects.

Perhaps a rigorous definition of perspective would be in order

I would submit that you are correct when you state the focal length has nothing to do with perspective, but it doesn't follow from there that the focal lengths are equal. 

[Ray]

Perhaps a rigorous definition of perspective would be in order

I would submit that you are correct when you state the focal length has nothing to do with perspective, but it doesn't follow from there that the focal lengths are equal. 

The definition that I use, as found in all the decent-sized English dictionaries that I've checked, is as follows.

"The appearance of objects, buildings, etc, relative to each other, as determined by their distance from the viewer, or the effects of this distance on their appearance."

What I've found, without exception, is that scenes shot with wide-angle lenses result in the appearance, on the print or monitor, of close objects being larger and closer than they seemed to the eye in the real scene at the time the shot was taken, and distant objects appearing further away than they seemed to the eye in the real scene at the time the shot was taken.

What I've also found, without exception, is that scenes shot with a telephoto lens result in the appearance, on print or monitor, of distant objects being much closer than they initially appeared to the eye in the real scene. In fact, if the lens is a long telephoto, objects that are too small for the eye to notice, or plain invisible, can appear in great detail and clarity on print or monitor.

I can therefore only conclude that focal length of lens has a strong influence on perspective. How anyone could argue against this, beats me.

However, that is not to say that position does not affect perspective. Obviously it does. And so does the focal length of lens used.

[Slobodan Blagojevic]

I guess our friend Ray subscribes to the marketing theory that "perception is reality"

[aizan]

there is separate term for what you are talking about: perspective distortion.

the ideas are related but they are not the same.

[BJL]

The definition that I use, as found in all the decent-sized English dictionaries that I've checked, is as follows.

"The appearance of objects, buildings, etc, relative to each other, as determined by their distance from the viewer, or the effects of this distance on their appearance."

What I've found, without exception, is that scenes shot with wide-angle lenses result in the appearance, on the print or monitor, of close objects being larger and closer than they seemed to the eye in the real scene at the time the shot was taken, and distant objects appearing further away than they seemed to the eye in the real scene at the time the shot was taken.

What I've also found, without exception, is that scenes shot with a telephoto lens result in the appearance, on print or monitor, of distant objects being much closer than they initially appeared to the eye in the real scene. In fact, if the lens is a long telephoto, objects that are too small for the eye to notice, or plain invisible, can appear in great detail and clarity on print or monitor.

And as has been observed many times in these forums, the effect like change in the relative sizes of nearer and farther objects that you see is due to some combination of (a) the typical usage of wider angle lenses closer to the main subject and telephoto lenses from further away, with the camera position then being the cause of the perspective differences, and (b) the presence of a more objects in a wider angle image taken from the same location, so that a great range of distances and apparent sizes are present.

If instead one compares wide and telephoto images taken with a camera at the same location pointing in the same direction and so at the same main subject (but with more peripheral subjects in the wide angle version), for example by using a standard zoom lens on a tripod and taking successive shots at both its shortest and longest focal length, then there is no significant change in the "appearance of objects, buildings, etc, relative to each other" for objects that appear in both images. Instead the telephoto image is essentially a crop of the wide angle image.

[Craig Lamson]

If instead one compares wide and telephoto images taken with a camera at the same location pointing in the same direction and so at the same main subject (but with more peripheral subjects in the wide angle version), for example by using a standard zoom lens on a tripod and taking successive shots at both its shortest and longest focal length, then there is no significant change in the "appearance of objects, buildings, etc, relative to each other" for objects that appear in both images. Instead the telephoto image is essentially a crop of the wide angle image.

+1

[theguywitha645d]

The change in perspective because of different FoV is a viewing distance issue. Perspective changes with viewing distance because it comes down to a projection and angular size problem. When the first experiments in linear perspective where done, the idea of the "point of view," not only for the point where the painting was made, but also where it was viewed, was very important. We know know that the 3-D illusion created with linear perspective is a very robust illusion--we can even look at an image from the sides and our visual system will interpret the image as "normal looking."

There are two basic viewing distances; standard viewing distance and the correct viewing distance. The standard viewing distance is equal to the diagonal of the image/print--so a 16x20 print should be viewed at about 25". The correct viewing distance is proportion to ratio of the focal length to the format diagonal, so a 21mm lens on a 35mm camera (diagonal: 43mm) would be about 0.5X and so a 16x20 prints would be viewed at 12.5" for the correct perspective (a perspective we would have experienced). So when the correct viewing distance is less than the standard, the perspective increases; when greater, it decreases.

A "normal" lens is a called that, not because it has the same angle of view as the human eye (it does not), but because the correct and standard viewing distance is the same and the perspective appears normal to the viewer.

Now some folks don't get viewing distance--how can the image change because of that? Well, think about this. We can see if the image was taken by wides or telephotos. We can really notice if the photographer gets close to a wide. But yet, as we actually wander through the world, we don't actually feel our visual perspective change--your significant other does not look anymore strange across a room than when you have your faces close. It only becomes strange when we photograph.

[AlfSollund]

It seems we are falling into the typical confusion of comparing the technical qualities of images of different scenes.

If one uses the same lens at the same F/stop with different size sensors, one gets different images. It doesn't make sense to compare the 'depth of field' in two images when the field itself is different.

Using the same lens and F/stop, the larger sensor produces a wider field. If that wider field is cropped to the same width as the image from the smaller sensor, the DoF will be the same, excluding minor variations due to differences in sensor pixel density and sensor resolution.

We experienced a similar confusion some time ago on this forum in relation to perspective. There were those who adamantly argued that focal length of lens has nothing to do with perspective. It's only the position that counts. To demonstrate this misconception, such people would present the argument that cropping a shot taken with a wide-angle lens to the same field of view as a shot taken from the same position using a telephoto lens, resulted in images that appeared to have the same perspective, thus demonstrating the principle that cropping an image is effectively the same as increasing the focal length of lens and that two images shot from the same position, using 'effectively' the same focal length of lens, will have the same perspective in all respects.

First; if there should be any point in comparing one should do that for different images. It doesn't make much sense to compare the same image (to itself)

The DoF as defined and used in photography depends on objective measures (CoC), so its quite easy and reasonable to compare the 'depth of field' in two images when the field itself is different.

We experience a similar confusion now for DoF as perspective.  Some adamantly claims that a argument (perspective) that can be demonstrated and supported by proof in fact is a misconception. Well, it should be fairly easy to prove this wrong, not only make the claim  .

[Ray]

First; if there should be any point in comparing one should do that for different images. It doesn't make much sense to compare the same image (to itself)

Of course! You've misunderstood me if you think I'm recommending comparing identical images. I'm recommending comparing images of identical scenes. There's a difference. The purpose of such comparisons is find out how the images differ from each other.

The DoF as defined and used in photography depends on objective measures (CoC), so its quite easy and reasonable to compare the 'depth of field' in two images when the field itself is different.

No it isn't. Take a photograph of a brick wall that fills the whole frame, using a good lens at F2.8 that's sharp from corner to corner. Take another photograph of the same wall and same field at F5.6, and compare DoF. Take a third photograph of the same brick wall at F11, but increase the size of the field so there are a few distant, rather blurry trees visible around the borders of the wall. Which image has the largest circles of confusion?

I'd suggest that the images at F2.8 and F5.6 will appear to have the same DoF, and that the image shot at F11 will appear to have a shallower DoF because it's the only image out of the three that has blurred components, the distant trees.

However, when comparing the effects of lens focal length on perspective, it is necessary to compare different fields because that's the purpose of using a different focal length, to create a wider or narrower FoV from the same position. I would be very foolish to use a 14mm lens to shoot a distant bird, thinking I could just crop the wide-angle shot to get the same perspective that a 400mm lens would provide.

Imagine such a comparison. I produce on the one hand a lovely, A3 size print from the 400mm shot, showing the birds feathers, beak and eyes in sharp clarity. On the other hand, I produce an A3 size print from a very small crop resulting in a hugely interpolated print file that gives a very blurred impression of a few daubs of color that might be loosely interpreted as a semi-abstract rendition of a few leaves. The bird is not discernible.

What! Can't you see that these two images have the same perspective? 

[AlfSollund]

No it isn't. Take a photograph of a brick wall that fills the whole frame, using a good lens at F2.8 that's sharp from corner to corner. Take another photograph of the same wall and same field at F5.6, and compare DoF. Take a third photograph of the same brick wall at F11, but increase the size of the field so there are a few distant, rather blurry trees visible around the borders of the wall. Which image has the largest circles of confusion?

In order to have a meaningful comparison the elements compared must be within the most "tele" crop. The definition and use of CoC is already agreed, so lets not waste time on discussing this. The bottom line being that CoC can be used to decide DoF for any given photography irrespectively of aperture, field of view etc. The purpose is just that; to be able to decide DoF in a objective manner transparent from equipment. Its not a question about "Which image has the largest circles of confusion?", but up to the analyzer to choose an appropriate CoC and then compare with the same CoC across images to check DoF.

From Wiki and other sources “acceptable sharpness” in the final image (e.g., print, projection screen, or electronic display) is that the blur spot be indistinguishable from a point" and

"Visual acuity. For most people, the closest comfortable viewing distance, termed the near distance for distinct vision (Ray 2000, 52), is approximately 25 cm. At this distance, a person with good vision can usually distinguish an image resolution of 5 line pairs per millimeter (lp/mm), equivalent to a CoC of 0.2 mm in the final image.
Viewing conditions. If the final image is viewed at approximately 25 cm, a final-image CoC of 0.2 mm often is appropriate. A comfortable viewing distance is also one at which the angle of view is approximately 60° (Ray 2000, 52); at a distance of 25 cm, this corresponds to about 30 cm, approximately the diagonal of an 8″×10″ image. It often may be reasonable to assume that, for whole-image viewing, a final image larger than 8″×10″ will be viewed at a distance correspondingly greater than 25 cm, and for which a larger CoC may be acceptable; the original-image CoC is then the same as that determined from the standard final-image size and viewing distance. But if the larger final image will be viewed at the normal distance of 25 cm, a smaller original-image CoC will be needed to provide acceptable sharpness.
Enlargement from the original image to the final image. If there is no enlargement (e.g., a contact print of an 8×10 original image), the CoC for the original image is the same as that in the final image. But if, for example, the long dimension of a 35 mm original image is enlarged to 25 cm (10 inches), the enlargement is approximately 7×, and the CoC for the original image is 0.2 mm / 7, or 0.029 mm."

Imagine such a comparison. I produce on the one hand a lovely, A3 size print from the 400mm shot, showing the birds feathers, beak and eyes in sharp clarity. On the other hand, I produce an A3 size print from a very small crop resulting in a hugely interpolated print file that gives a very blurred impression of a few daubs of color that might be loosely interpreted as a semi-abstract rendition of a few leaves. The bird is not discernible.

What! Can't you see that these two images have the same perspective? 

I say that, and further more they have the same perspective irrespectively of what you or I say! The "blurred impression of a few daubs of color" is simply due to that the small crop has been enlarged so much that details are lost. But the perspective will be exactly the same. You are mixing perspective with resolution. If you take your A3 print and aply the same resolution as to the small crop they will be identical. This is simply nothing to discuss unless you claim to change the laws of physics. If this is the case your results should be published elsewhere than here.

[Ray]

I say that, and further more they have the same perspective irrespectively of what you or I say!

So in other words, you think that perspective is determined by some mathematical theory or imaginary model that may have nothing to do with what we see? If you see a difference in the perspective of elements in two images and such differences seem at odds with your interpretation of the mathematical model, then you assume that your vision must be faulty. Right? 

The "blurred impression of a few daubs of color" is simply due to that the small crop has been enlarged so much that details are lost.

Sorry! You've got that wrong. Enlargement does not result in loss of detail. It's downsizing that results in loss of detail.

But the perspective will be exactly the same. You are mixing perspective with resolution.

No I'm not. Resolution comes first. It is the basis of all images. No resolution, no image. No image, no perspective.

If you take your A3 print and aply the same resolution as to the small crop they will be identical. This is simply nothing to discuss unless you claim to change the laws of physics. If this is the case your results should be published elsewhere than here.

But it is you who are attempting to break the laws of physics by applying the same resolution to a very small crop of a wide-angle shot. The laws of physics put a limitation on lens resolution. It's known as diffraction. The laws of physics also put a limitation on sensor resolution. Such limitations are photonic shot noise, thermal noise, read-out noise, and even  the frequency of the light waves themselves which would eventually put a practical limit on the smallness of pixels.