M4/3 vs APS-C vs FF vs Medium Format

5th street

Well-known member
Messages
108
Reaction score
8
In my search for a camera/lens to take sweeping ultra wide landscape photos (and printing sizes like 12 x 18" and occasionally 2 x 3 ft) I've asked questions on forums and read lots of discussions and reviews. I'm still puzzled about the importance of sensor area (in real world applications).

In general, there seems to be a consensus that the area of the sensor is very important for collecting the photons. (It seems to me that physics and common sense support this consensus).

Therefore medium format would be best except for the fact that it is cost prohibitive, is too large and heavy, and requires too much in computer resources.

The same arguments are used for ff vs aps-c. However the differences in cost, size and weight etc between ff and aps-c are not as great as medium to ff. Therefore there are folks who do use ff and who recommend it.

What about aps-c vs m4/3? As in the other comparisons there are advantages of the larger sensor area of aps-c vs m4/3. However, the medium format vs ff and ff vs aps-c sensor area arguments no longer appear to be very important (they are rarely mentioned in discussions).

And in the aps-c vs m4/3 comparison there is rough equivalence in cost (in some cases m4/3 is more expensive). The size and weight is more equivalent (after all, you still can't put m4/3 in your pocket).

I am leaving out smaller sensors since practically no one recommends them for my purpose (probably because there areas are too small to do a good job).

Why is it that the sensor area advantages tend to be downplayed in the aps-c vs m4/3 discussions when these same arguments become so important in medium vs ff vs aps-c discussions and the small sensors are ignored entirely?

There must be some trade offs that clearly favor the m4/3 size and make the sensor area advantages less important.

What am I missing here? My attempt at logical analysis fails me. Thanks for your input...I do want to make a good choice an am open to all suggestions.

(I'll also post this question on the Landscape forum.)
 
Thanks for your reply. I'm still trying to understand. Carrying your points one step further does this mean the Nikon 1 smaller sensor would have similar advantages compared to m4/3? At what size sensor does this line of reasoning stop?
If the aim is to get the greatest depth of field, then the smaller the sensor, the better. The reverse holds true for shallow DoF. The tradeoffs are in things like lens selection, resolution, diffraction limit, etc. With the very small sensors you can't stop down very far before getting diffraction, which is why premium compacts have a built-in variable ND filter to control exposure.

I like M4/3 because the lens selection is good, the lenses are typically sharpest at F4 or F5.6 where I do a lot of my shooting, and the gear is lightweight.

Since you want to make landscape posters I would start by looking at the lens selection of various systems and see what your options are. Then look at some full size images on Flickr and see which cameras / lenses are suitable. Start from the image and work back to the sensor.
 
Thanks for your input. Can you comment on your experience with your 7-14mm lens?
I am happy with the lens. It's certainly good for landscapes. While there are UWA zooms for other systems that may be just as good optically, such as the Sigma 8-16 for APS-C and the Nikon 14-24 for FF, those will also be bigger and heavier.

Here's a modest example of what the 7-14 can do. I happen to have this one online at full resolution so that you can view it at 100 percent if you want. Perhaps I could have sharpened it a bit more and the close foreground is not fully within the DoF, but I think it gives you an idea of what the lens can do even under difficult light conditions (not much contrast in the shadow).





Another good option for landscape photography with MFT is the Samyang 7.5/3.5 fisheye. Very good and very cheap for what it delivers. Here are a couple of examples. The first with the original fisheye distortion and the second partly defished.











 
Ever done a few days hike/climb into a place like the Rockies with a rugged (DX/FX D300/D800) body, a few high end lenses (14-24 2.8), tripod and your camping/hiking gear, you would not hesitate to lighten that load with something like an EM-1/EM-5 and a 7-14 or a couple of primes. And as a bonus, no tripod required.

Especially as you get older ;-)
 
What about aps-c vs m4/3? As in the other comparisons there are advantages of the larger sensor area of aps-c vs m4/3. However, the medium format vs ff and ff vs aps-c sensor area arguments no longer appear to be very important (they are rarely mentioned in discussions).

There must be some trade offs that clearly favor the m4/3 size and make the sensor area advantages less important.

What am I missing here? My attempt at logical analysis fails me. Thanks for your input...I do want to make a good choice an am open to all suggestions.

(I'll also post this question on the Landscape forum.)

At one time I owned MFT, APS-C and FF cameras and I did a lot of comparisons between them all on screen and in prints up to and including A3.

My finding was that at low to medium ISO settings all of my cameras produced good images when shooting raw and processing each image for best effect and clear differences were only really visible when shooting at the highest ISO settings or pixel peeping on screen at high magnification. In prints differences were less noticeable.

When pushing the cameras to their limits I found that the FF camera produced the best image quality at the highest ISO settings and the APS-C was better than the MFT camera although these days I'd expect new model MFT cameras to be more competitive with APS-C cameras.

The most interesting thing to come out of my little tests was that people I asked to pick MFT from FF images couldn't reliably do so and guessed right at no better than chance. Looking at images on my sceen I can often only tell what they were taken with by remembering the shot and what gear I used, by looking at the file name and guessing from that or by reading the exif.

So, IMVHO unless you are ging to shoot at very high ISO settings it really doesn't matter all that much what format you choose between MFT, APS-C or FF) if you are only printing to a max of A3 or so.

One advantage I've found for MFT (apart from bulk and weight) is that when shooting hand held there's often a clear advantage over full frame as you can get the same DoF (at / or smaller than aperture f2.8 on FF) when using a faster shutter speed... this is often an advantage when shooting hand held.
 
What about aps-c vs m4/3? As in the other comparisons there are advantages of the larger sensor area of aps-c vs m4/3. However, the medium format vs ff and ff vs aps-c sensor area arguments no longer appear to be very important (they are rarely mentioned in discussions).

There must be some trade offs that clearly favor the m4/3 size and make the sensor area advantages less important.

What am I missing here? My attempt at logical analysis fails me. Thanks for your input...I do want to make a good choice an am open to all suggestions.

(I'll also post this question on the Landscape forum.)
At one time I owned MFT, APS-C and FF cameras and I did a lot of comparisons between them all on screen and in prints up to and including A3.

My finding was that at low to medium ISO settings all of my cameras produced good images when shooting raw and processing each image for best effect and clear differences were only really visible when shooting at the highest ISO settings or pixel peeping on screen at high magnification. In prints differences were less noticeable.

When pushing the cameras to their limits I found that the FF camera produced the best image quality at the highest ISO settings and the APS-C was better than the MFT camera although these days I'd expect new model MFT cameras to be more competitive with APS-C cameras.

The most interesting thing to come out of my little tests was that people I asked to pick MFT from FF images couldn't reliably do so and guessed right at no better than chance. Looking at images on my sceen I can often only tell what they were taken with by remembering the shot and what gear I used, by looking at the file name and guessing from that or by reading the exif.

So, IMVHO unless you are ging to shoot at very high ISO settings it really doesn't matter all that much what format you choose between MFT, APS-C or FF) if you are only printing to a max of A3 or so.

One advantage I've found for MFT (apart from bulk and weight) is that when shooting hand held there's often a clear advantage over full frame as you can get the same DoF (at / or smaller than aperture f2.8 on FF) when using a faster shutter speed... this is often an advantage when shooting hand held.
Your experience exactly mirrors mine. One must print quite large before a noticeable difference emerges. If the photo is properly exposed and composed, the difference in sensor size is very marginal. Print 4'x6' and it becomes a different story.
 
Thanks for your reply. I'm still trying to understand. Carrying your points one step further does this mean the Nikon 1 smaller sensor would have similar advantages compared to m4/3?
In principle yes, although the Nikon 1 sensors don't shine as examples of their size. See here regarding the general pattern:

http://www.dpreview.com/forums/post/52452668
At what size sensor does this line of reasoning stop?
It probably doesn't. But there are other reasons to say stop at a certain point. One is that the DoF eventually becomes greater than you want it to be. In other words, with a sufficiently small sensor you increasingly often find yourself in a situation where a) you could accept less DoF in exchange for less noise or b) you actually want less DoF for subject isolation by means of background blur but cannot reduce the DoF as much as you would want to even if the lens is fast.

For example, even if you could find a small-sensor compact with a lens as fast as f/2, that may be equivalent to f/11 or f/16 on FF with respect to DoF and there may be many situations where you don't need quite that much DoF but could make do with less (in exchange for less noise) or actually want less (for better subject isolation).

Furthermore, you will run into problems with sharpness because lenses do not do as well at the very wide f-stops you would normally want to use on a very small sensor (for reasons mentioned above) and/or because sharpness becomes limited by diffraction (which has a greater impact at the same f-stop the smaller the sensor gets).

Personally, I am happy with MFT, but I am not tempted to go any smaller than that in terms of sensor size for the reasons I have mentioned above.
 
Last edited:
Thanks for your input. Can you comment on your experience with your 7-14mm lens?
I am happy with the lens. It's certainly good for landscapes. While there are UWA zooms for other systems that may be just as good optically, such as the Sigma 8-16 for APS-C and the Nikon 14-24 for FF, those will also be bigger and heavier.

Here's a modest example of what the 7-14 can do. I happen to have this one online at full resolution so that you can view it at 100 percent if you want. Perhaps I could have sharpened it a bit more and the close foreground is not fully within the DoF, but I think it gives you an idea of what the lens can do even under difficult light conditions (not much contrast in the shadow).



Another good option for landscape photography with MFT is the Samyang 7.5/3.5 fisheye. Very good and very cheap for what it delivers. Here are a couple of examples. The first with the original fisheye distortion and the second partly defished.



Where are these taken. Great shots.
 
Thanks for your input. Can you comment on your experience with your 7-14mm lens?
I am happy with the lens. It's certainly good for landscapes. While there are UWA zooms for other systems that may be just as good optically, such as the Sigma 8-16 for APS-C and the Nikon 14-24 for FF, those will also be bigger and heavier.

Here's a modest example of what the 7-14 can do. I happen to have this one online at full resolution so that you can view it at 100 percent if you want. Perhaps I could have sharpened it a bit more and the close foreground is not fully within the DoF, but I think it gives you an idea of what the lens can do even under difficult light conditions (not much contrast in the shadow).



Another good option for landscape photography with MFT is the Samyang 7.5/3.5 fisheye. Very good and very cheap for what it delivers. Here are a couple of examples. The first with the original fisheye distortion and the second partly defished.



Where are these taken. Great shots.
Thanks. All three are in or around the (US) Sierras. The first is at Mono Lake, which you hit when you leave Yosemite going eastwards. The second is the Ubehebe crater in Death Valley. The third is Tenaya Lake, located at the Tioga road within Yosemite.
 
I have read this forum every single day for years, but I haven't posted much. I came from Canon long ago, but have since had an Oly EP-2, Pana GH2 and now an Oly EM1 (with 12-40mm, which is outstanding). So I'm an m4/3 convert through and through.

However, when I want to do highly detailed landscape work, the Sigma DP1M and DP3M come out of the pack. Nothing short of medium format beats the DPXM (Merrills) for high resolution. I love the Oly EM-1 and I'm blown away by its capabilities with the 12-40mm. However, whenever I take landscape pics with either the DP1M or the DP3M, both exceed my wildest expectations for resolution. Like medium format.

So, I'm in full agreement with previous posters in this thread who have mentioned the Sigma DP Merrills... they are (at least for me) the go-to option when I really want super high resolution (eg., a landscape with lots of tree branches, leaves, distant trees, grass and other fine details). I paid full price for mine, but I've seen lately that they are discounted a bit.

The Sigma Merrills with their quirky Foveon sensors are almost a 'technical camera', very much 'specified purpose', and they are somewhat finicky and difficult to use... but the results are hard to beat.
 
In principle yes, although the Nikon 1 sensors don't shine as examples of their size. See here regarding the general pattern:

http://www.dpreview.com/forums/post/52452668
At what size sensor does this line of reasoning stop?
It probably doesn't. But there are other reasons to say stop at a certain point. One is that the DoF eventually becomes greater than you want it to be. In other words, with a sufficiently small sensor you increasingly often find yourself in a situation where a) you could accept less DoF in exchange for less noise or b) you actually want less DoF for subject isolation by means of background blur but cannot reduce the DoF as much as you would want to even if the lens is fast.

For example, even if you could find a small-sensor compact with a lens as fast as f/2, that may be equivalent to f/11 or f/16 on FF with respect to DoF and there may be many situations where you don't need quite that much DoF but could make do with less (in exchange for less noise) or actually want less (for better subject isolation).
Which brings us back to FF, as the f/5.6 mFT lenses are indeed f/11 equivalent. And there is only that much of a deep DoF you will ever need in most cases. Past that point it becomes destructive more than helping. For example, f/11 is way too closed for most landscapes shots with FF cameras, and unless you have some very close foreground stopping lens down to that mark is simply not needed. More typical f/5.6 to f/7.1 does just fine. What do you do with mFT on the other end though, like where you do need to open it up and bring the main point of interest forward. Well, not much, really.
Furthermore, you will run into problems with sharpness because lenses do not do as well at the very wide f-stops you would normally want to use on a very small sensor (for reasons mentioned above) and/or because sharpness becomes limited by diffraction (which has a greater impact at the same f-stop the smaller the sensor gets).
Exactly! And that is why by comparing the images I see that latest FF cameras pull more than two stops ahead, mainly because you have more room to deal with noise (when needed) before down-scaling those ready to be shown images.
Personally, I am happy with MFT, but I am not tempted to go any smaller than that in terms of sensor size for the reasons I have mentioned above.
Personally, I am happy with FF. But it does not mean I would not prefer mFT either. I would simply use it for different things, and where FF would see less action for all practical purposes.
 
Probably the best compromise of all.

FF sensor in a small body. Really small.

lots of options with lenses.

Not a fast camera like the EM-1 but easily fine for landscapes.

Lots of dynamic range and resolution advantages over MFT in case you need them. High resolution is always good to have if you want to crop.

One of these and a couple of prime wide lenses ( Nikon, Canon lenses or whatever ) from eBay and you have the perfect landscape rig IMO.
 
for large landscape prints, i would select a camera with the sharpest output, either one without AA-filter or one with foveon-sensor, or a medium format camera, which mostly don't have any AA-filter either.

if you still want something small but with great dynamic range and sharpness you could try one of the merill sigmas, DP1 or DP2 or even DP3 when you want to stitch. several stitched shots from the DP3 should give you quality equivalent to medium format, but costing a fraction and being portable in the jacket pocket.

i know the sigmas are a pain to use, but the output will make up for this.
 
Last edited:
In principle yes, although the Nikon 1 sensors don't shine as examples of their size. See here regarding the general pattern:

http://www.dpreview.com/forums/post/52452668
At what size sensor does this line of reasoning stop?
It probably doesn't. But there are other reasons to say stop at a certain point. One is that the DoF eventually becomes greater than you want it to be. In other words, with a sufficiently small sensor you increasingly often find yourself in a situation where a) you could accept less DoF in exchange for less noise or b) you actually want less DoF for subject isolation by means of background blur but cannot reduce the DoF as much as you would want to even if the lens is fast.

For example, even if you could find a small-sensor compact with a lens as fast as f/2, that may be equivalent to f/11 or f/16 on FF with respect to DoF and there may be many situations where you don't need quite that much DoF but could make do with less (in exchange for less noise) or actually want less (for better subject isolation).
Which brings us back to FF, as the f/5.6 mFT lenses are indeed f/11 equivalent.
Which f/5.6 MFT lenses are you talking about? To my knowledge, there isn't one.
And there is only that much of a deep DoF you will ever need in most cases. Past that point it becomes destructive more than helping. For example, f/11 is way too closed for most landscapes shots with FF cameras, and unless you have some very close foreground stopping lens down to that mark is simply not needed.
Since I have somewhat higher standards for sharpness, I often shoot landscapes at f/8 (f/16 on FF) and rarely go wider than f/5.6 (f/11 on FF).
More typical f/5.6 to f/7.1 does just fine. What do you do with mFT on the other end though, like where you do need to open it up and bring the main point of interest forward. Well, not much, really.
You open up.
Furthermore, you will run into problems with sharpness because lenses do not do as well at the very wide f-stops you would normally want to use on a very small sensor (for reasons mentioned above) and/or because sharpness becomes limited by diffraction (which has a greater impact at the same f-stop the smaller the sensor gets).
Exactly! And that is why by comparing the images I see that latest FF cameras pull more than two stops ahead, mainly because you have more room to deal with noise (when needed) before down-scaling those ready to be shown images.
As you know, FF is actually a bit behind MFT with regard to noise for equivalent images (same DoF, same shutter speed, and, consequently different f-stops and different ISOs).
Personally, I am happy with MFT, but I am not tempted to go any smaller than that in terms of sensor size for the reasons I have mentioned above.
Personally, I am happy with FF. But it does not mean I would not prefer mFT either. I would simply use it for different things, and where FF would see less action for all practical purposes.
Different from what? MFT works great for landscapes.
 
Depends on the budget, the best quality would be a Nikon D800e with some sharp glass...
 
I have no interest in taking sweeping ultra wide landscape photos, but if I did, I would stitch. The camera used would be irrelevant and the DR with bracketed exposures would for all practical purposes be unlimited.
 
Which f/5.6 MFT lenses are you talking about? To my knowledge, there isn't one.
From the lenses you own, it is in your gear page,

Panasonic Lumix G Vario 45-200mm F4-5.6 OIS
Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS
Panasonic Lumix G Vario 14-42mm F3.5-5.6 ASPH OIS
Panasonic Lumix G Vario 100-300mm F4-5.6 OIS
Olympus M.Zuiko Digital ED 40-150mm 1:4-5.6 R
Olympus M.Zuiko Digital 14-42mm 1:3.5-5.6 II R
Since I have somewhat higher standards for sharpness, I often shoot landscapes at f/8 (f/16 on FF) and rarely go wider than f/5.6 (f/11 on FF).
Ah, c'mon Anders, let's not go into that, shall we :-) !?
You open up.
You know, there is nothing better than shooting through native (not equivalent) focal length and apertures. Like f/2.8 = f/2.8. When you need deeper scene, you stop the lens down. When you need to bring the point of interest forward, you open it up. Well, you know .. all that un-compromised good stuff I am talking about :-) .
As you know, FF is actually a bit behind MFT with regard to noise for equivalent images (same DoF, same shutter speed, and, consequently different f-stops and different ISOs).
No, I did not know that. Here is high ISO with two stops in between. Opened and down-scaled the larger image. How is it looking?





original.jpg


Different from what? MFT works great for landscapes.
Yes, but for better output larger format is always better. It is just the way it is. Unless they stop developing larger formats, and mFT finally catches up, it will always be so.





--
- sergey
 
Depends on the budget, the best quality would be a Nikon D800e with some sharp glass...
Sony A7R a much better option. Same sensor inside the camera as the D800e. Much smaller and lighter. Can take all the lenses you want.

D800e has better AF and stuff but this isn't needed here anyway.

A7R would be fun too, goofing around on ebay and second hand camera stalls and trying out all sorts of glass on it with adaptors.

Focus peaking and a great EVF offers lots of advantages over D800e as well.
 
Which f/5.6 MFT lenses are you talking about? To my knowledge, there isn't one.
From the lenses you own, it is in your gear page,

Panasonic Lumix G Vario 45-200mm F4-5.6 OIS
Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS
Panasonic Lumix G Vario 14-42mm F3.5-5.6 ASPH OIS
Panasonic Lumix G Vario 100-300mm F4-5.6 OIS
Olympus M.Zuiko Digital ED 40-150mm 1:4-5.6 R
Olympus M.Zuiko Digital 14-42mm 1:3.5-5.6 II R
I don't see a single f/5.6 lens there. All I see are zooms with variable max aperture, reaching f/5.6 only at the long end.

Now speaking about my gear page, let me know at what FL I can't go wider than f/5.6 and the relevance of that FL for landscapes.
Since I have somewhat higher standards for sharpness, I often shoot landscapes at f/8 (f/16 on FF) and rarely go wider than f/5.6 (f/11 on FF).
Ah, c'mon Anders, let's not go into that, shall we :-) !?
Why not?
You open up.
You know, there is nothing better than shooting through native (not equivalent) focal length and apertures. Like f/2.8 = f/2.8. When you need deeper scene, you stop the lens down. When you need to bring the point of interest forward, you open it up. Well, you know .. all that un-compromised good stuff I am talking about :-) .
The lenses and apertures I shoot are of course just as native as those you shoot. When I need deep DoF I stop down and when I need less, I open up, just like you.
As you know, FF is actually a bit behind MFT with regard to noise for equivalent images (same DoF, same shutter speed, and, consequently different f-stops and different ISOs).
No, I did not know that. Here is high ISO with two stops in between. Opened and down-scaled the larger image. How is it looking?
No idea how that was processed. And, as always, you should look at the weakest link (the shadows). Here's the real story. Click on "show signature" if the images don't show:

http://www.dpreview.com/forums/post/50462878

BTW: How many stops difference do you see here and in which direction when a) ISO is held constant and b) DoF is held constant (the D800 at two stops higher ISO)?

DR-EM1-D800.jpg

Different from what? MFT works great for landscapes.
Yes, but for better output larger format is always better.
Of course not. Why would that be? Smaller sensors tend to be more efficient.
It is just the way it is.
Tautology of the year?
Unless they stop developing larger formats, and mFT finally catches up, it will always be so.
I am afraid it is impossible to catch up when you are already ahead. ;-)
 
Last edited:
As you know, FF is actually a bit behind MFT with regard to noise for equivalent images (same DoF, same shutter speed, and, consequently different f-stops and different ISOs).
No, I did not know that. Here is high ISO with two stops in between. Opened and down-scaled the larger image. How is it looking?
Close, with the D800 "a bit behind MFT with regard to noise..". Would have expected the downsampled D800 image to have a resolution advantage (better 'pixel-level sharpness'), but that doesn't seem to be the case here.
 

Keyboard shortcuts

Back
Top