Shootout GAS attack: Oly EM1mkii vs Sony A7iii

[ooheadsoo]

Yeah, agree on parts and have even stated similar, but I ask and investigate certain findings and arguments. There is nothing wrong with that or that it implies I don't do other stuff. You can think what you want about my or others inquiry. I think it's interesting and I learn at the same time.

Btw, the weather in Southern Finland is quite horrid for any kind of landscape photography at the moment at least for my taste.

It's great for you to ask questions, but your questions are best answered with experience. Your questions show you are looking for some objective truth. None of those guys are wrong. Some care more about diffraction than others, that's all. You wont find some kind of universal truth from these general photography guides - they're just guidelines. The only "universal guide" you will find is to not stop down "more that you have to" because of diffraction unless you haven't reached your lens's sweet spot. Diffraction effects can vary based on lens and sensor format, so how can anyone write one guide that rules them all? Some photographers want to get it all in one shot, so you have to stop down, diffraction or not. Focus stacking isn't always an option, say if you have moving water or leaves. No one can tell you how much diffraction is acceptable to you. Why do lens manufacturers make lenses that stop down to f/32 even for crop sensors? Did Olympus and Tamron not understand about diffraction, or did the marketing department overrule the engineers and say, "no trust us, our market research shows ppl will buy more lenses if they go to f/28"? On the other hand, sometimes you may want less dof and the lens has not yet reached its sweet spot. What will you do? Stop down anyway because a guide or review says so?

You have studied enough to get started. It's time to find out for yourself. It's not that you shouldn't ask questions. It's that I don't think people can give you a meaningful answer other than "it depends".

The biggest wrinkle is that most of the technical faults of gear are not even particularly visible at normal viewing distance. How perfect you want the images is, again, up to you. Most of the reviews you are reading are for single samples no one can tell if anything you personally buy will perform the same when it comes to the fine details. The reviews are good because most don't have the discipline or equipment to do their own testing (thanks, crazy150) but rarely represent an absolute truth. It's hard to have a rigorous and academic discussion about something as vague as "decency level."

Sorry the weather isn't working out.

 

[Crazy150]

That's interesting, center sharpness at lenstip for these seems wrong, first clear "mistake". Edge sharpness seems quite right (point about 24 lpmm vs 0 lpmm over decency, but that was for 12mm) as I see in comparison the edges worse for Tamron than the center for Olympus. Probably sample variation.

Also that Olympus f/8.0 vs Tamron f/16.0 is interesting because it clearly goes against what lenstip says with their decency levels. Tamron in center is at decency and below at edges and Olympus is clearly above in center and still above in edges both at all focal lengths. With Tamron f/16.0 it should be a bit worse than edges on that 35mm f/2.8 picture.

I think edge sharpness is not so important for tele-end like portraiture? But isn't it very often important in landscapes and similar even at longer focal lengths? Or if you shoot street for example and with m43 you want to use fastest aperture?

I'm a beginner so what you said about the aperture interests me too. I was of the impression that one would usually use f/5.6 for landscapes with m43, maybe when shooting at 12-25mm. But this might have to do with the optimal sharpness for most lenses.

I found these three articles about it:

In this article Don Smith says for the last picture that he could not get the whole scene in focus even with f/22 (m43 f/11) with Sony 16-35mm (m43 8-17.5mm) so he focus stacked two pictures.

https://www.donsmithblog.com/2016/10/15/is-there-a-perfect-aperture-for-landscape-photography/

Spencer Cox: "Set your aperture to balance depth of field and diffraction – typically, at a full-frame equivalent of f/8 to f/16 (but a larger aperture for nighttime photography, when you have no other choice)"

https://photographylife.com/landscapes/best-camera-settings-landscape-photography

Jim Harmer: "If you're using a wide lens (10-25mm) on a crop frame camera, and you have a foreground element close to the camera, you might consider an aperture of f/18. This will allow for solid depth-of-field and acceptable sharpness on many wide-angle lenses.

If you're using a wide lens (10-25mm) on a crop frame camera, and you do NOT have a foreground element, consider an aperture of around f/11 or f/16."

https://improvephotography.com/580/the-ideal-aperture-for-landscape-photography/

Are these guys wrong? Two of them mention wide focal lengths with smaller aperture so at least they are not talking about long focal length landscapes. Harmer's advice seems a bit odd though with such apertures for crop frame cameras. But Smith (with great pictures in his article) says with foreground elements and big landscape even f/11 with m43 isn't enough.

As far as diffraction and DOF on MFT is concerned, while I wouldn’t avoid shooting at f16 if necessary (think macro, extreme landscapes), F8 should suffice for 99% of shots on MFT. Just put a DOF calculator on your phone and read about hyperfocal distance. For example, shooting at 12mm on MFT if you focus on something ~4ft from the camera will have everything from 2ft to infinity in focus. At really wide (say 7mm) F5.6 will do the same. So I can't see too many instances where you'd need to do F16 or focus stacking, but I don't do a lot of landscapes so I'll defer to people that get paid for their landscape images. Also keep in mind that many landscape photographers are from the film days when you didn't have instant feedback on the image and nothing is worse than schlepping 15kg of gear on a long hike and getting the film developed only to see your foreground is blurry or your edges are soft or vignetted. Nowadays, you can simply take the image and check it, or take dozens of images at differing exposures, or turn on focus peaking, or zoom in across the frame to check, etc.

As for the lens comparison, while I wouldn't get too caught up with the MTF charts, I won't tell you like so many to "just go out and shoot" or "focus on the basics of photography and stop looking at MTF charts." {rant} Photography is a hobby for 95% of those that own a camera. Of those 90% of those will never sell an image in their lifetimes. Part of a hobby is to dive into the weeds and get out of it something that brings you joy and entertainment. As much as I'd like to, I can't always go out and shoot pretty landscapes or street or astro or whatever and even if my skill allows it, I'd never be able to support my family with photography as a job and have zero desire to deal with bridezillas and deadlines. I'm a nerd and I love exploring the technical aspects and physics of image making, MTF charts and lens/system comparisons are big part of that for me. {/rant}

My comparison isn't as "scientific" as lenstip I'm sure, so I wouldn't discount them for mine. The point of my posts here is just to document my experiences with a serious competitor to the MFT systems espcially with the more compact zooms on offer form Tamron and some cheaper/vintage primes. The advantage of cost, weight, features that MFT has enjoyed is not what it used to be and many here are getting tempted by the dark side. I hope I've reinforced that while there are some advantages that simply can't be replicated on MFT (shallow DOF, DR, AF, low light, legacy wide lenses), for most shooters the gap on weight and cost is still such that it probably isn't the time to switch unless you really crave/need those distinct advantages.

 

[RS86]

It's great for you to ask questions, but your questions are best answered with experience. Your questions show you are looking for some objective truth. None of those guys are wrong. Some care more about diffraction than others, that's all. You wont find some kind of universal truth from these general photography guides - they're just guidelines. The only "universal guide" you will find is to not stop down "more that you have to" because of diffraction unless you haven't reached your lens's sweet spot. Diffraction effects can vary based on lens and sensor format, so how can anyone write one guide that rules them all? Some photographers want to get it all in one shot, so you have to stop down, diffraction or not. Focus stacking isn't always an option, say if you have moving water or leaves. No one can tell you how much diffraction is acceptable to you. Why do lens manufacturers make lenses that stop down to f/32 even for crop sensors? Did Olympus and Tamron not understand about diffraction, or did the marketing department overrule the engineers and say, "no trust us, our market research shows ppl will buy more lenses if they go to f/28"? On the other hand, sometimes you may want less dof and the lens has not yet reached its sweet spot. What will you do? Stop down anyway because a guide or review says so?

You have studied enough to get started. It's time to find out for yourself. It's not that you shouldn't ask questions. It's that I don't think people can give you a meaningful answer other than "it depends".

The biggest wrinkle is that most of the technical faults of gear are not even particularly visible at normal viewing distance. How perfect you want the images is, again, up to you. Most of the reviews you are reading are for single samples no one can tell if anything you personally buy will perform the same when it comes to the fine details. The reviews are good because most don't have the discipline or equipment to do their own testing (thanks, crazy150) but rarely represent an absolute truth. It's hard to have a rigorous and academic discussion about something as vague as "decency level."

Sorry the weather isn't working out.

Yeah, it's good advice to go out and shooting, it just came out as if it was bad to compare these things and talk about it with others in the internet. I agree that pixel peeping is not very good spending of time, but also I like to know things before I get to that perfect moment so I don't ruin it by making bad mistakes.

And one point is that this "argument" if f/4.0 gives always everything in focus with m43 (which as a beginner made me think) also made me research and find different perspectives and update my information about it. Others can also read them and enjoy the photos.

From the Don Smith article who says similar things as you, which I liked very much along with those awesome photographs:

"I could show many more examples of varying apertures for landscapes but I think you get the picture (pardon the pun). The key is to ask yourself how you wish to render your composition. Once I get the proper aperture that renders the scene the way I previsualized it in my mind, then I can always change my ISO to get a suitable shutter if movement of any element within the composition is an issue. So think Composition then Aperture. Make choosing the aperture a conscious thought, and if you need a smaller aperture, then don’t get so hung up on diffraction. Don’t let the fear-mongers influence you to the point that you don’t properly record the image at the proper aperture needed to match your previsualization of the finished image!"

I have especially found this true with macro.

 

[RS86]

As far as diffraction and DOF on MFT is concerned, while I wouldn’t avoid shooting at f16 if necessary (think macro, extreme landscapes), F8 should suffice for 99% of shots on MFT. Just put a DOF calculator on your phone and read about hyperfocal distance. For example, shooting at 12mm on MFT if you focus on something ~4ft from the camera will have everything from 2ft to infinity in focus. At really wide (say 7mm) F5.6 will do the same. So I can't see too many instances where you'd need to do F16 or focus stacking, but I don't do a lot of landscapes so I'll defer to people that get paid for their landscape images. Also keep in mind that many landscape photographers are from the film days when you didn't have instant feedback on the image and nothing is worse than schlepping 15kg of gear on a long hike and getting the film developed only to see your foreground is blurry or your edges are soft or vignetted. Nowadays, you can simply take the image and check it, or take dozens of images at differing exposures, or turn on focus peaking, or zoom in across the frame to check, etc.

As for the lens comparison, while I wouldn't get too caught up with the MTF charts, I won't tell you like so many to "just go out and shoot" or "focus on the basics of photography and stop looking at MTF charts." {rant} Photography is a hobby for 95% of those that own a camera. Of those 90% of those will never sell an image in their lifetimes. Part of a hobby is to dive into the weeds and get out of it something that brings you joy and entertainment. As much as I'd like to, I can't always go out and shoot pretty landscapes or street or astro or whatever and even if my skill allows it, I'd never be able to support my family with photography as a job and have zero desire to deal with bridezillas and deadlines. I'm a nerd and I love exploring the technical aspects and physics of image making, MTF charts and lens/system comparisons are big part of that for me. {/rant}

My comparison isn't as "scientific" as lenstip I'm sure, so I wouldn't discount them for mine. The point of my posts here is just to document my experiences with a serious competitor to the MFT systems espcially with the more compact zooms on offer form Tamron and some cheaper/vintage primes. The advantage of cost, weight, features that MFT has enjoyed is not what it used to be and many here are getting tempted by the dark side. I hope I've reinforced that while there are some advantages that simply can't be replicated on MFT (shallow DOF, DR, AF, low light, legacy wide lenses), for most shooters the gap on weight and cost is still such that it probably isn't the time to switch unless you really crave/need those distinct advantages.

It was this I was commenting to: "F4 gets just about everything in focus @17mm." and which made me wonder and research. But now you say wider 7.5mm needs f/5.6 to have everything in focus, so maybe there was a mistake in the number F4? And that now you mentioned F8 suffices for 99% with m43 to get everything in focus.

Yeah, I'm very thankful for your contribution. I just don't like to trust single review sites, but have the impression that lenstip.com should be one of the best. So it's interesting to see what can be found with your testing and also can their "decency levels" be compared across systems. Of course it is tough to make any conclusions with the lens sample variations. Also wanted to know about the m43 diffraction disadvantage, because it's not very obvious by looking at the reviews or my macro stuff.

I think it's very natural to compare Olympus 12-40mm and Tamron 28-75mm for their similar price point, like you said, it interests many. For me it looks like Olympus is better for landscapes (especially because it's wider) and some other stuff, and likewise Tamron with 24MP FF sensor for others. One just needs to know what they need, and this is one way to find out.

 

[Darmok N Jalad]

A very good read, research, and discussion—one that is certainly over my head at times. Like many, I’ve felt the pull of FF, but I shoot tele a lot, and I couldn’t bear to haul (or finance) FF tele around, so m43 portability gets points right away. I also hear about the complaints of Sony ergo’s, and I just love the feel of my G9 so much that it would be hard to go backward there.

I’m actually a little surprised that APS-C is mostly a mid-tier system. It seems like a great middle ground, but I haven’t seen much in that class that interests me. Fujifilm seems to take it the most serious at the high-end, as the XT3 is very well reviewed and appears to have a good layout for me. If I were to leave m43 today, it would probably be to the XT3. I think one thing that I like about m43 is that there is some work to getting the rewards. It’s like driving a manual—an automatic is easier, but you are not as bonded to the machine.

 

[Pluttis]

Yeah, it's good advice to go out and shooting, it just came out as if it was bad to compare these things and talk about it with others in the internet. I agree that pixel peeping is not very good spending of time, but also I like to know things before I get to that perfect moment so I don't ruin it by making bad mistakes.

Nothing bad or wrong about being interested in lerning techical things regarding photography...

but experience from practices/taking pictures and experiment will be by far the best and most important thing for you when it comes to be prepared and not make mistakes for thise "perfect" moments.

 

[Pluttis]

I think it's very natural to compare Olympus 12-40mm and Tamron 28-75mm for their similar price point, like you said, it interests many. For me it looks like Olympus is better for landscapes (especially because it's wider) and some other stuff, and likewise Tamron with 24MP FF sensor for others. One just needs to know what they need, and this is one way to find out.

I would say Sony + Tamron combo will offer best overall image quality, especially if you want/need to push the files much in post.

 

[RS86]

Nothing bad or wrong about being interested in lerning techical things regarding photography...

but experience from practices/taking pictures and experiment will be by far the best and most important thing for you when it comes to be prepared and not make mistakes for thise "perfect" moments.

Yeah, that's the way I went into it. Mostly reading forums and some internet stuff and then trying out. Quite happy, but I have some photos with 45-150mm photographing nice cloud formations and having f/11. And a few similar landscape mistakes with too small aperture for pixel peepers from first year of owning a camera.

They are still surprisingly good looking and actually great to my eye, but for a person with any perfectionist tendencies these are the things that make you want to improve the most for the next time, and maybe do some studying.

Happy New Year, I'm off!

 

[Crazy150]

It was this I was commenting to: "F4 gets just about everything in focus @17mm." and which made me wonder and research. But now you say wider 7.5mm needs f/5.6 to have everything in focus, so maybe there was a mistake in the number F4? And that now you mentioned F8 suffices for 99% with m43 to get everything in focus.

Yeah, I'm very thankful for your contribution. I just don't like to trust single review sites, but have the impression that lenstip.com should be one of the best. So it's interesting to see what can be found with your testing and also can their "decency levels" be compared across systems. Of course it is tough to make any conclusions with the lens sample variations. Also wanted to know about the m43 diffraction disadvantage, because it's not very obvious by looking at the reviews or my macro stuff.

I think it's very natural to compare Olympus 12-40mm and Tamron 28-75mm for their similar price point, like you said, it interests many. For me it looks like Olympus is better for landscapes (especially because it's wider) and some other stuff, and likewise Tamron with 24MP FF sensor for others. One just needs to know what they need, and this is one way to find out.

Okay, gotcha. Sorry, I wasn't intending to be technical or literal in that comment, but just hilighting that on the MFT system for general use I rarely find the need to go beyond F4 for DOF. You should explore the DOF calculators that are free online. For example, here is what I'm talking about with the 17mm. If you focus at or just beyond the hyperfocal (~5m in this example), then everything from about 2.5m from the lens to infinity will be in focus. So, unless you have foreground elements that are closer than that F4 should suffice.

Subscribe to see EXIF info for this image (if available)

F8:

Subscribe to see EXIF info for this image (if available)

You'd only need F16 if you want to reduce that to 60cm but as we see, the lens gets softer so maybe this is dimishing returns. As you go wider, it's even more favorable.

In your links, this image is an extreme example because he wants the flowers/grass in focus as well as the mountain. Looks to be 35mm on a FF. The grass looks to be less than 50cm away from the camera.

Subscribe to see EXIF info for this image (if available)

Using the DOF calculator even F22 will only get 60cm to infinity in focus. So, he needed to do stacking to get this exact image. Alternatively to the focus stacking is that he could have moved a little back or went a little wider on the focal length to increase DOF and simply cropped in for the composition (it is an A7r2 he's using), but that's up to the user and what he intends with the resolution. Personally, I don't favor the stacking because it can look a bit like a composite image like in this image the grass in the foreground looks simply added onto a nice landscape.

 

[Crazy150]

Little update on this thread. Still using the A7iii a bit. I've decided to shoot both systems for now and still working on my lens kits for the two, but I've recently been playing around with a Canon FD 500mm F8 mirror lens that I got for just over $100. On the MFT that's a lot of reach for $100. Anyway, decided to compare it on the Sony vs the Oly. Here is the result with the A7iii cropped in 200% vs 100% on the Oly. Oh and the bird is perched in an 80ft palm about 100-200 yards away I'd guess.

Subscribe to see EXIF info for this image (if available)

I was really expecting the Oly to have at least a slight advantage considering the crop, but actually the Sony has just the slightest advantage here. Same thing when I was shooting the moon the other night, once I got both images in lightroom and cropped them the same there really was no noticeable difference in detail. I'm not a huge birder or anything, but does anyone else's experience contradict this?

One thing I will say is that the superb stabilization made manual focusing while zoomed in much easier on the Oly.

 

[ooheadsoo]

How does the focusing compare with each other? I'm primarily interested in autofocus differences, face/eye autofocus in particular, but if you've noticed differences in focus peaking, that would be interesting, too.

 

[Crazy150]

How does the focusing compare with each other? I'm primarily interested in autofocus differences, face/eye autofocus in particular, but if you've noticed differences in focus peaking, that would be interesting, too.

On AF I'd say "it depends". In general the AF on sony is quite good. The algorithm is different though and it definitely prefers the nearest thing in the frame, so fences, tree limbs, etc. in the foreground need to managed more than i remember with the Oly. If I were to shoot in an AF intense situation I think I would reach for the Sony because the accuracy seems better. Although the Oly with many lenses has almost instant AF lock which seems faster than the sony if less precise.

The eye-detect of the Sony is very good. I have it assigned to a button to activate when needed. That said for portraiture type shots, I rarely if ever miss with the Oly and say the 45 1.8 or 75 1.8. The larger DOF makes it so much easier to hit the eye, I don't think eye detect is really needed. In summary, I'd say the sony is slightly better but it needs to be bc of the shallower DOF.

As for manual focussing, I'd say they are comparable. I haven't really had issues with either. I typically rely more on the zoom though for manual focussing than peaking and they both are similar here But if you are doing street or other run/gun manual focus the sony has a more visible peaking but haven't tested the accuracy.

As just a slight update to the post above with the birds, I noticed I didn't have the Sony IBIS set to the right focal length so maybe it would have been better at aiding focus when zoomed if it was able to stabilize better.

 

[Jeroen81]

Little update on this thread. Still using the A7iii a bit. I've decided to shoot both systems for now and still working on my lens kits for the two, but I've recently been playing around with a Canon FD 500mm F8 mirror lens that I got for just over $100. On the MFT that's a lot of reach for $100. Anyway, decided to compare it on the Sony vs the Oly. Here is the result with the A7iii cropped in 200% vs 100% on the Oly. Oh and the bird is perched in an 80ft palm about 100-200 yards away I'd guess.

View attachment 801852

I was really expecting the Oly to have at least a slight advantage considering the crop, but actually the Sony has just the slightest advantage here. Same thing when I was shooting the moon the other night, once I got both images in lightroom and cropped them the same there really was no noticeable difference in detail. I'm not a huge birder or anything, but does anyone else's experience contradict this?

One thing I will say is that the superb stabilization made manual focusing while zoomed in much easier on the Oly.

I like the way the Oly is processing the bird much more. Look at the haircut, it's totally washed away on the Sony.
But without joking, I do like the Left picture more. The colors attract me way more. For IQ I don't see much of a difference.

 

[Cederic]

It's not just the colours, the tree has far more texture in the Oly shot - to the extent that I'm wondering if the tree was in focus rather than the bird, which is why the bird isn't quite as sharp as the Sony picture.

The Oly does get a bit confused when focussing on scenes like that, and ironically has the dynamic range and detail to show it.

 

[ooheadsoo]

Thanks for the notes on AF. Too bad you don't have a G9 for comparison.

 

[Darmok N Jalad]

It's not just the colours, the tree has far more texture in the Oly shot - to the extent that I'm wondering if the tree was in focus rather than the bird, which is why the bird isn't quite as sharp as the Sony picture.

The Oly does get a bit confused when focussing on scenes like that, and ironically has the dynamic range and detail to show it.

I’ve found such photos are hard to get much out of anyway, as the surroundings can take away from the subject. Maybe it’s just me, but it’s too busy for my eyes. I would want to have more zoom so less cropping is needed.

 

[Crazy150]

It's not just the colours, the tree has far more texture in the Oly shot - to the extent that I'm wondering if the tree was in focus rather than the bird, which is why the bird isn't quite as sharp as the Sony picture.

The Oly does get a bit confused when focussing on scenes like that, and ironically has the dynamic range and detail to show it.

It was manually focussed with a 500mm vintage lens so no camera error here. It was a burst and this was the one most in focus on the bird, but yes it's possible the focus is slightly off.

I’ve found such photos are hard to get much out of anyway, as the surroundings can take away from the subject. Maybe it’s just me, but it’s too busy for my eyes. I would want to have more zoom so less cropping is needed.

It's not meant to be a great shot, just a bird in my backyard while I was testing the lens/camera. Not sure how you going to get more zoom though. That's 1000mm equivalent, but of course I get your point. But my point was that at least in terms of detail, the Sony with the same lens cropped manages just as much detail. So, the MFT effective focal length "advantage" here was lost.

Here is another example of the Moon the other night. Both with the same 500mm adapted lens. Basically, I get the same shot by cropping the Sony image. While the true resolution of the EM1 image is greater, the increased detail with the Sony means that for this lens at least any advantage on the Oly is lost.

Subscribe to see EXIF info for this image (if available)