Reading mode: Light Dark

Landscape perfection?

Go to last post
Ken, you seem to be explaining the difference between higher and lower resolution sensors, rather than the difference between larger and smaller sensors.

Is there a reason there can't be a 36mp m43 sensor? If not, why would a 36mp m43 sensor necessarily be any worse than a 36mp FF sensor for landscape work?

--
My Flickr Photostream
 
Last edited:
jkrumm said:
But common enough... First of all the very idea of a "pro" landscape photographer these days is not based on reality. There are a few photographers out there who make a living selling huge landscape fine art prints, but not many, likely less than 100 worldwide. Most of the pros who do that stuff have to teach workshops, endorse gear, produce podcasts, become accountants...whatever it takes to make income.

Second, there are former professional medium format shooters out there doing professional projects on M43 right now, happy with the image quality.
Click to expand...
there are professional formula 1 race car drivers who are happily driving their day to day automobiles too but i'd bet they would not care to race them against their track cars or suggest to others that these day to day cars had a sniff of a chance of beating an F1 track car on a track.

a 2X crop sensor body/system has plenty of advantages over a FF DSLR. IQ is not one of them.

good enough is only good enough.
Member said:
Third, if you look through the handful of photo books I've made for myself (and I'm sure others in this forum) using 4/3 and m4/3 gear, you cannot tell the format, the brand, the lenses used or anything of the sort. You can tell most of the shots were not made with a phone, that's about it.
Click to expand...
 
Ansel Adams gear was good enough. Why don't we all run out and buy that gear? This is getting silly....................





 
FrankS009 said:
This is interesting, but I would be grateful for a slightly greater explanation of the resolution question in particular. Folks who favour FF argue that for landscape shots, foliage and other details are sharper and more defined with FF sensors - that M4/3rds images are "muddier." There are have been more than one thread that deals with this subject. It reminds me of the debate about stereo equipment, it is all about "removing veils" from our capture of reality.

What you seem to be saying is that this can be done through post processing. Can you provide a bit more information for those of us who are relative beginners in the PP area?
Click to expand...
You already got a good answer from Richard. What I can add is that with regard to resolution, it is not the larger sensor per se that is of primary importance but the pixel count. If you take a 16 MP FF sensor, e.g., that of the Nikon Df or D4S, you won't see much of a resolution difference in comparison with a 16 MP MFT sensor as long as the lenses are of similar quality. If, however, you take a 36 MP FF sensor, like for example that of the Nikon D810, it's another story.

One way of achieving the same or higher resolution with MFT is to stitch multiple frames together so that you get the same or higher pixel count for the final image. This involves more work with post-processing but also in the field. You have to get a series of images, not just a single one, and these have to be shot so that they will fit nicely together. You find a few images produced by means of this technique by texinwien here:

Loading…

www.dpreview.com

Those are not posted at full resolution. But you find a crop from the last image here

Loading…

www.dpreview.com

that will give you an idea of the amount of detail it provides.
 
Michael M Fliegel said:
Ansel Adams gear was good enough. Why don't we all run out and buy that gear? This is getting silly....................
Click to expand...
AA was a post processing master but he had the expertise and tools 'required' to achieve the utmost results. i'm almost positive AA did not use cell phone captures to make his prints from.

i'm no AA so i doubt i'll use his comparable gear. i'll stick with my m4/3 stuff since it's good enough for me.

just because any particular image might look fabulous (content) does not strictly mean the equipment used to capture that image was (or wasn't) 'the' best tool for the job.

good enough, fast enough, tasty enough, soft enough is up to each and his their own.

for you or i to instruct others that our opinion of any such matter trumps any other's is odd.
--
Thanks,
Paul
 
Last edited:
richarddd said:
Ken, you seem to be explaining the difference between higher and lower resolution sensors, rather than the difference between larger and smaller sensors.

Is there a reason there can't be a 36mp m43 sensor? If not, why would a 36mp m43 sensor necessarily be any worse than a 36mp FF sensor for landscape work?
Click to expand...
Well, they are all sort of related. Key point being "sort of".

In theory we could make a 36MP m43 sensor, no problem there technically. Pixel densities already there in smaller senors. Probably doesn't fit the market segment though because once you account for read out rates, video capability, processing and so on and so on it isn't a free lunch by any means to go to higher pixel densities. Most m43 offerings are currently tailored to being excellent "jack of all trades" cameras with great video and fast shooting features rather than trying to be a slow resolution machine. But anyway, of course you could make a 36MP m43 camera.

But then we would need to make lenses to keep up with the 36MP sensor. Again quite possible, but starting to get a bit harder due to manufacturing tolerances. Again, not clear cut, because smaller lenses while they require tighter tolerances in mechanical assembly also typically allow for more elements of the lens to use aspherical elements. But in general, the physically larger your optical system the higher resolution you will be able to obtain within the limits of manufacturing. What isn't clear is how much of a practical difference that would come down to in say an ideal m43 vs an ideal FF imaging system, there are so many other variables beyond just the two times sensor size difference. As a real matter of what is available in the market place the difference is quite obvious, m43 tops out a 16MP right now where FF goes up to 36MP and has quite a few lenses that keep up with the 36MP sensor. Move up to medium format and difference is even more drastic as to what is achievable with available rather than theoretical camera systems.

We also should remember that "resolution" for a landscape photographer also usually includes noise - even at base ISO. This is another place where size matters. As has been discussed ad nauseum when we shoot "equivalent" exposures there really isn't much if any noise advantage to FF vs m43. However, landscape photographers rarely shoot "equivalent". They shoot base ISO and if that means a 10 second exposure they don't care, they still shoot base ISO. And FF sensors just because of their larger area effectively have lower base ISO noise than an identical spec m43 sensor. As Anders already pointed out in this thread (and myself in many others) you can get around this by doing bracketed exposures with m43, but that doesn't always work with all subjects.

Lastly there are just additional features and lens selection effects at work in what is practical and the FF and MF camera systems bring more to the table. Once you get to very high resolutions you find DoF and diffraction being severely limiting. You are left with either lens tilts or focus stacking to solve that problem. The vast majority of landscape shots are easily tackled with lens tilts. And the availability of TS lenses means a lot fewer exposures in the field for a given high resolution shot.

For a static subject like most landscapes there are few limitations to m43 that can't be overcome with various field techniques. I pretty regularly do HDR stacks because they are easy to do in the field and easy to handle in post processing. That solves the dynamic range and base ISO noise "problems" of m43. I've done some panorama stitching to and you can get to very high resolutions doing that of course. I've experimented with small focus stacks (just a foreground and infinity stack) but those haven't hit my normal workflow. But put all three of those together to try to match what a TS lens on a high resolution FF or MF camera could do in a single exposure and things will get impractical fast. I personally would do it just for the fun of the challenge. Someone trying to make money would probably be better served just taking the bigger camera and gear...
--
Ken W
See profile for equipment list
 
I remember reading in the Oly SLR forum years back that Olympus originally stated its lenses were good for about 20 mp. This back when their flagship had 5, so it sounded a long ways off. In truth, I suspect it's about right for their best lenses achieving something that looks good at 100%. It would be like a full frame lens dealing with 80 mp, I think. Otherwise you get diminishing returns... as Thom describes with the RX100 lll, which he considers equivalent to a good 12 mp camera in the end.
 
FrankS009 said:
This is interesting, but I would be grateful for a slightly greater explanation of the resolution question in particular. Folks who favour FF argue that for landscape shots, foliage and other details are sharper and more defined with FF sensors - that M4/3rds images are "muddier." There are have been more than one thread that deals with this subject. It reminds me of the debate about stereo equipment, it is all about "removing veils" from our capture of reality.
Click to expand...
Richard and Anders already gave you great answers. I'll add one more wrinkle...

When it comes to foliage the color response of the sensor can become a factor in "muddy" looking results. By color response I mean what is the specific spectral response of the red, green and blue filters on the sensor. On some cameras these filters are wider (i.e. the red lets in a bit of green, the green lets in some red and blue, the blue lets in some green). By making the filters wider the camera can test with better noise performance. Unfortunately the wider filters usually result in poorer color discrimination which can impact the look of slightly different colored foliage.

Since smaller cameras have been competing heavily in the noise department and places like DPR and DxO always test for noise there appears to have been a trend to widen the spectral filters on smaller sensors. This has even happened on some FF sensors apparently, many folks greatly preferring the narrower filters of the old A900 sensor to the more recent ones. Move up into MF sensors where the target audience is studio photographers with unlimited controlled lighting and landscape photographers happy to open a shutter for many seconds and suddenly there is no good reason to widen the filters and most sensors are using narrower spectral filters.

So the result is that there is another variable at play, smaller sensors typically have different color response than larger sensors simply because of the different market forces applied to the different sensor sizes. You could of course make a m43 sensor with the color filters typically used in a MF back - but no one does.

The effects of filter spectral response are difficult to test and demonstrate. As a result they are quite likely to be misunderstood or overstated - just like all the "golden ears" problems with audio equipment. For example, a lot of people seem to have confused the effects of wider filters with the similarly timed move from CCD to CMOS. But the differences because of filter spectral response are there and real to an extent. How much they matter compared to all the other poorly controlled variables when people compare two different camera systems? Really hard to say...
--
Ken W
See profile for equipment list
 
jkrumm said:
I remember reading in the Oly SLR forum years back that Olympus originally stated its lenses were good for about 20 mp. This back when their flagship had 5, so it sounded a long ways off. In truth, I suspect it's about right for their best lenses achieving something that looks good at 100%. It would be like a full frame lens dealing with 80 mp, I think. Otherwise you get diminishing returns... as Thom describes with the RX100 lll, which he considers equivalent to a good 12 mp camera in the end.
Click to expand...
The other big deal for landscape photographers is that edge and corner performance can matter a lot. So when someone designs a lens with super high resolution in the center at wide apertures that is great news for a studio or portrait photographer. It is pretty useless for a landscape photographer though. Designing lenses with high resolution corners and flat fields of focus is really, really challenging. And the required filter stacks on top of digital sensors has not made the job any easier...

So I suspect many current m43 lenses would test to gangbusters resolutions in the center. But at the moment even an expensive dedicated prime like the 12/2 which should be wonderful for landscape is already a bit of a let down in the corners at just 16MP.
--
Ken W
See profile for equipment list
 
And that's part of the reason the regular 4/3 lenses were just about the same size as full frame lenses (at the super high grade level).

If you want a flat field, the 50 f2 is like some kind of lab instrument. The 7-14 and even my 12-60 show some corner funkiness on my EM1. Not sure if it's the adapter or what, but it's irritating when zooming in on the screen. I'll be curious to see how good the new 7-14 2.8 is in this regard.
 
Hi, Paul. I was just venting. I love my EM1. Each to his fate!





 
jkrumm said:
I remember reading in the Oly SLR forum years back that Olympus originally stated its lenses were good for about 20 mp. This back when their flagship had 5, so it sounded a long ways off. In truth, I suspect it's about right for their best lenses achieving something that looks good at 100%. It would be like a full frame lens dealing with 80 mp, I think. Otherwise you get diminishing returns... as Thom describes with the RX100 lll, which he considers equivalent to a good 12 mp camera in the end.
Click to expand...
It doesn't really work that way. MFT lenses are about twice as sharp as FF lenses if we consider resolution per millimeter (on the sensor) and about equally sharp if we consider resolution per image. So the per-image resolution you can expect from an MFT lens on an x MP MFT sensor is about the same as the one you can expect from an FF lens of comparable quality on an x MP FF sensor.

What the larger format has going for it here, however, is that the lens is stopped down more at equivalent f-stops (those that are the same with regard to DoF, diffraction, and total light on the sensor). And the more a lens is stopped down, the better it becomes at keeping those aberrations that limit sharpness in check.

How much this matters, depends on where along the f-stop range you compare. If you compare f/4 on MFT with f/8 on FF, the sharpness difference you can expect is very small. If, however, you compare f/1 on MFT with f/2 on FF, the FF lens can be expected to be much sharper since it is far easier to make a lens perform well at f/2 than at f/1.

If instead you compare at the same rather than equivalent f-stops, the larger format has the advantage of less diffraction. In this case, the difference can be expected to be small at the wide end (say f/2 on MFT versus f/2 on FF) but large at the narrow end (say f/16 versus f/16). In schematic terms, the differences you can expect are depicted in the graphs below:

abe6976012b1421c9456ece5f5e7e237.jpg

35bc1c1968b54aab89564595e7fa4454.jpg
 
Last edited:
kenw said:
jkrumm said:
I remember reading in the Oly SLR forum years back that Olympus originally stated its lenses were good for about 20 mp. This back when their flagship had 5, so it sounded a long ways off. In truth, I suspect it's about right for their best lenses achieving something that looks good at 100%. It would be like a full frame lens dealing with 80 mp, I think. Otherwise you get diminishing returns... as Thom describes with the RX100 lll, which he considers equivalent to a good 12 mp camera in the end.
Click to expand...
The other big deal for landscape photographers is that edge and corner performance can matter a lot. So when someone designs a lens with super high resolution in the center at wide apertures that is great news for a studio or portrait photographer. It is pretty useless for a landscape photographer though. Designing lenses with high resolution corners and flat fields of focus is really, really challenging. And the required filter stacks on top of digital sensors has not made the job any easier...

So I suspect many current m43 lenses would test to gangbusters resolutions in the center. But at the moment even an expensive dedicated prime like the 12/2 which should be wonderful for landscape is already a bit of a let down in the corners at just 16MP.
Click to expand...
But do you really think that MFT is any worse in this department (sharpness across the frame) than larger formats? I can't say I have such an impression.
 
kenw said:
FrankS009 said:
This is interesting, but I would be grateful for a slightly greater explanation of the resolution question in particular. Folks who favour FF argue that for landscape shots, foliage and other details are sharper and more defined with FF sensors - that M4/3rds images are "muddier." There are have been more than one thread that deals with this subject. It reminds me of the debate about stereo equipment, it is all about "removing veils" from our capture of reality.
Click to expand...
Richard and Anders already gave you great answers. I'll add one more wrinkle...

When it comes to foliage the color response of the sensor can become a factor in "muddy" looking results. By color response I mean what is the specific spectral response of the red, green and blue filters on the sensor. On some cameras these filters are wider (i.e. the red lets in a bit of green, the green lets in some red and blue, the blue lets in some green). By making the filters wider the camera can test with better noise performance. Unfortunately the wider filters usually result in poorer color discrimination which can impact the look of slightly different colored foliage.

Since smaller cameras have been competing heavily in the noise department and places like DPR and DxO always test for noise there appears to have been a trend to widen the spectral filters on smaller sensors. This has even happened on some FF sensors apparently, many folks greatly preferring the narrower filters of the old A900 sensor to the more recent ones. Move up into MF sensors where the target audience is studio photographers with unlimited controlled lighting and landscape photographers happy to open a shutter for many seconds and suddenly there is no good reason to widen the filters and most sensors are using narrower spectral filters.

So the result is that there is another variable at play, smaller sensors typically have different color response than larger sensors simply because of the different market forces applied to the different sensor sizes. You could of course make a m43 sensor with the color filters typically used in a MF back - but no one does.
Click to expand...
Not sure I follow you here Ken. Of course, the filter may be made more or less stringent but I can't say I have noticed any tendency for smaller sensors to carry more lenient filters than larger, at least if we compare in the MFT to FF span. Haven't looked all that carefully at what happens further down or higher up but a brief glance at Phase One didn't suggest much of a difference even at that size.
kenw said:
The effects of filter spectral response are difficult to test and demonstrate. As a result they are quite likely to be misunderstood or overstated - just like all the "golden ears" problems with audio equipment. For example, a lot of people seem to have confused the effects of wider filters with the similarly timed move from CCD to CMOS. But the differences because of filter spectral response are there and real to an extent. How much they matter compared to all the other poorly controlled variables when people compare two different camera systems? Really hard to say...
--
Ken W
See profile for equipment list
Click to expand...
 
You must log in or register to reply here.

Similar threads

S
Thoughts on the Nikon Tour....
Replies
1
Views
2K
shuncheung
S
Handiworks
50-400 Flexibility - Surprise Event & Lessons
2 3 4
Replies
68
Views
4K
Handiworks
Handiworks
C
New to DPR. Thinking about older compact cameras!
Replies
3
Views
603
constantlyconfused
C
LongLiveLumia
My brief talk with Brave Leo AI about "L.Monochrome D" and Lumix picture styles and filters
Replies
10
Views
2K
James Stirling
James Stirling
M
Panasonic Lumix 12-60mm f/3.5-5-6 vs 14-140 mm f/3.5-5.6 - optical evaluation
Replies
9
Views
9K
Myrmeleon
M

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top