(Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII

Started Jul 12, 2018 | Discussions
MEDISN
MEDISN Contributing Member • Posts: 866
(Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII
15

A discussion broke out within our local camera club recently on the topic of real-world differences between modern camera formats. DxO measurements, hyperbolic epithets, and personal bias aside, how much of a difference in image quality am I going to see by using format A vs format B? After a few weeks of gathering test shots, several members presented blind comparisons (FF vs APS-C, APS-C vs Pixel2, FF vs MF and even a MF vs Ricoh GR). That last comparison may have sold a few Ricoh GR’s that day as it really surprised a lot of people unfamiliar with it. I presented many of the 35 shot comparisons below for FF vs mFT and I thought it was worth posting the unblinded 1:1 shots here for discussion.

If you don’t want to read, and just want the verdict – the FF sensor in the A7R is better. But that wasn’t the real question.How different are the images shot under the same conditions?

There are numerous ways to address a question like this based on what matters most to the user. I primarily shoot FF (Canon and Sony) and mFT (Olympus and Panasonic), plus a mix of older film formats for fun (but not in any significant quantity). More on what/how I shoot here if interested.

I have, for years, carried an A7R with me for wide landscape/urban work alongside my “general use” Canon or mFT camera’s, as the results have been fantastic, especially given the size/portability of the A7R with small primes. As I was traveling and happened to have both the EM1mkII and A7R with me in overlapping focal lengths, I made some side-by-side shots for comparison. I later included some indoor shots in lower (if not poor) lighting. Since most of these shots employed a tripod or solid surface to maintain shot-to-shot consistency, I have included high-res (pixel shift) output in the mix as well, as many are curious about this feature.

Nuances

There are numerous limits to these types of comparisons. While I have tried to keep parameters as similar as possible, these are two different size, shape, resolution, technology, sensors with two different imaging pipelines shooting through three different lenses. I’ve tried to match focal length and physical aperture to yield the same AOV and DOF.

  • Sony + 15mm Voigtlander and 25mm Batis
  • Olympus at 7mm or 12mm with mZD 7-14mm PRO

I kept ISO the same for many comparisons except where base ISO is being evaluated. All shots are 100% center crops. I made minor adjustments to WB to keep colors somewhat similar. I set sharpening to 0 for consistency which makes the files look a bit soft in general (especially with high-res) but I included some examples with sharpening and processing applied at the end for comparison.

Noise at Base ISO

At respective base ISO’s, when AOV and DOF are matched, the mFT is at a 3-stop detriment. This is probably the biggest difference I see when using both side-by-side. A true base ISO of 100 (or lower) is really nice to have.

Taken in low natural lighting from a window on the opposite side of the room

Hazy summer day

Center stack of a rental Jeep with bright light pouring through the windshield resulting in flare of the mZD 7-14.

Nearby parking structure

Coffee break. The cup was rotated slightly between shots - it was an active prop

Clouds were moving overhead - the sun came through for the Olympus shot. Didn't notice the discrepancy until later.

Airline terminal waiting for next flight.

Full sun, summer day, clear sky focused on the very edge (thin rubber) of the wheel.

Extended (LOW) ISO

I use those “LOW” options sometimes for cleaner shadows/skies and to eek out longer exposure times when a ND filter isn’t quite enough for the effect I’m looking for at base ISO.

Low natural light backlit from window

Same scene lower down focused on fan speed knob

Natural light from nearby window

This final "extended LOW ISO" shot has Shadows +100 to see the boards under the platform better.

Handheld

One of the caveats however of base (and extended LOW ISO) is managing the slower shutter speeds. This presents a problem for non-IBIS cameras like the A7R (rectified in subsequent "R" versions) and certainly knocks pixel shift out of the equation if shooting handheld.

Inside of a doll house lifted 1EV

Fortunately for the A7R, you can raise ISO to compensate. For high-res, the only option is tripod or stable surface.

I was able to get a steady shot with the A7R and 15mm Voigt at 1/15th, raising the ISO to accommodate. High-res file is predictably unusable.

Finally, the standard 20MP EM1mkII shot edited to taste next to the edited A7R shot downsized to match.

Inside of a doll house - images edited to taste.

Scaling for Fun

What happens when we upscale the standard Olympus 20MP or downscale the high-res file to match the A7R?

20MP Olympus file scaled to A7R size using Photoshop Preserve Detail 2.0

High-res file scaled to A7R size

Sunny day with passing clouds. High-res scaled to 48MP (original 80MP)

20MP Olympus file scaled up to A7R size. 80MP Olympus file scaled down to A7R size

One final 80MP high-res file scaled to A7R size. Both files sharpened and processed to taste. My "usual" A7R sharpening could probably be more aggressive by comparison. The high-res files hold up to a lot more sharpening than I'm used to.

Exposure Lifting

One of the benefits of gathering more total light on a FF sensor is the ability to lift exposure in post and still have clean detailed images. We’ve all seen the benefits in various reviews and with our own images. Whether intentional or unintentional, this is a need/benefit for many – let’s look at how they fare.

1EV

No surprise here, both look very good. With good exposure (#21) I don’t see a difference when lifting 1EV.

#21 light from window perpendicular to scene.

In more difficult light (backlit) (#22), the white wall behind the fan (below the window) looks cleaner with the A7R.

#22 bright light from window causes some shadow noise on white walls

I know, I know, 1EV, big deal. Let’s push on.

3EV

A difference more visible here, especially in poor lighting. Still, good result from the standard 20MP RAW mFT even without processing/noise reduction. In the Jeep photo (#23) there is some flare going on with the mZD 7-14. A7R and high-res look very good in these shots even without noise reduction.

#23 Center stack of a rental Jeep with bright light pouring through the windshield resulting in flare of the mZD 7-14.

Doll house in low natural lighting from nearby window.

Dirty wheel, low light.

Natural light from window perpendicular to scene.

5EV

The FF A7R pulls away from the 20MP mFT RAW here. A7R and high-res appear to control shadow noise well here despite the 5EV lift.

Doll house in low natural lighting from nearby window.

Dirty wheel, low light.

Center stack of a rental Jeep with bright light pouring through the windshield resulting in flare of the mZD 7-14.

Bright light from window causes some shadow noise on white walls

A7R vs High-Res

With a little processing for both, and scaling high-res down to A7R size, all look very good, even when underexposed and lifted in post.

Sunny summer day in a shady alley. Processed to taste.

Dirty wheel processed to taste

Doll house, processed to taste

Matching Total Light

No obvious difference in sample below aside from resolution. A7R file is underexposed 1EV, EM1mkII is overexposed 1EV. Merely an academic exercise to illustrate “total light”.

High-res file scaled to A7R size.

Final Thoughts

No one will make any claims about a mFT sensor going toe-to-toe with a FF sensor and “winning” on the merits of image quality, but that wasn’t the focus of the discussion. The original question was “what differences am I going to see?”. I’ve tried to provide some contextual examples that illustrate the differences here, with albeit mundane shots, but beyond the typical studio scenes available on this site and others.

My opinion is that both of these sensors produce great results and will largely depend on the user’s threshold for noise and detail in the final output. In some cases, IBIS, pixel-shift can attenuate the disadvantage in sensor size, in some cases they cannot. There are clear benefits of a larger sensor (certainly beyond this simplistic comparison). Yet, there are numerous variables beyond side-by-side 100% crops to weigh when choosing a format/system. Ultimately, I’m thrilled to see so many good options on the market – the various comparisons we saw in camera club illustrated that nicely. Many of us shoot multiple systems/formats so these results may seem “predictable” within the limitations stated.

I’m open to criticism, and welcome critique and commentary. This was a funexercise meant to stimulate discussion among club members and drain some cocktails while pixel peeping a 5K display and taking guesses at which format is which. Yes, the pictures are boring and don’t push the limits of either sensor tested. I have my own reservations with these types of comparisons, but I do think they havesome intrinsic value. I only ask that we keep discussion civil and not devolve into format/brand wars.

-IM

Olympus E-M1 II Sony a7R
If you believe there are incorrect tags, please send us this post using our feedback form.
goodbokeh
goodbokeh Contributing Member • Posts: 731
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII

That was a lot of good work you did! The one thing that was a red flag for me is the a7R's shutter shock problem. How did you work around that issue when you were using shutter speeds below 1/500 sec?

 goodbokeh's gear list:goodbokeh's gear list
Panasonic GH5 Sony a7R III Fujifilm X-T30 Fujifilm XF 14mm F2.8 R Fujifilm XF 27mm F2.8 +18 more
Iantau Regular Member • Posts: 101
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII
3

nicely documented work  

I did a similar exercise with a EM-1, 5D mkII and A7RII....  came to very similar conclusions. ...

Then i printed exhibition 16"x24"  prints of the same shots from all 3 cameras (24 in total)...  apart from some rendering differences between the lenses NO ONE could consistently tell which print came from which camera....

A local camera store borrowed the prints for a while - even after they had had them for a while and used them every day they still had to look at the back of the prints to tell which was which.

The only consolation that i had (id just bought the A7RII and a suite of lenses !) was that the A7 had more latitude for recovering from "operator errors" and was much less demanding of good technique in low light.

-- hide signature --

Ian
'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´¯'·.¸.·´
> > > > >

MEDISN
OP MEDISN Contributing Member • Posts: 866
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII

goodbokeh wrote:

That was a lot of good work you did! The one thing that was a red flag for me is the a7R's shutter shock problem. How did you work around that issue when you were using shutter speeds below 1/500 sec?

In my experience shutter shock is lens dependent.  I noticed it more on big tele lenses I had adapted (EF mount) than I did on normal and wide primes.  I can't say I've ever experienced it with the 15mm Voigtlander or 25mm Batis.  Some of my adapted legacy lenses (OM mount, M mount etc) are up to 100mm but then they never produce the razor sharp results of modern lenses so I'd be hard pressed to say if subtle softness was optics or shutter shock at 1/500 or 1/1000.   Something to be aware of but it really is unpredictable (which is why it can really frustrate some).

Keith Schmidt Veteran Member • Posts: 7,142
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII

Great work on this, and I like the way you've described the differences and scenarios.  There are definitely positives and negatives with each system.  I've used the EM1 and more recently the EM1mkII for the last 5 years, and had an A7II for most of 2017.  I couldn't see enough differences for me, and the positives of m43 were more important to me, so I finally let go of the A7II.

Fast forward to now, and I've added the A7III to my lineup, and can see myself parting with the m43.  There's enough of a difference in certain situations from an IQ standpoint, and the addition of the 24-105 lens to the mix resolves some of the issues/challenges I had with my previous FF experience.

We're lucky to have some compelling choices up and down the sensor/system spectrum these days.

 Keith Schmidt's gear list:Keith Schmidt's gear list
Olympus E-M1 Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Leica Summilux DG 25mm F1.4 Olympus M.Zuiko Digital 45mm F1.8 Olympus M.Zuiko Digital 17mm F1.8 +4 more
Lan Senior Member • Posts: 1,141
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII
1

Thank you for posting that! As you say a fun exercise I might have been tempted to tweak the white balance of the Sony shots a bit more, as some are rather green (fans/wheels/laundry)?

I did something similar a few years back; I took a shot of a local landmark with my cameraphone at the time (a 5MP Motorola ZN5), and I took the same shot a few seconds later with my 22MP Canon EOS 5D2 and 24-105 f4/L. Both shot full auto as SooC JPEG, as the Motorola didn't offer RAW or any exposure adjustments.

I then printed both at 8" x 10", and asked my photographer friends and the staff of my local enthusiast camera shop which they preferred. The results were interesting, and somewhat unexpected. On the whole they preferred the cameraphone shot. Why? Greater DOF, and more vibrant colours. D'oh!

Obviously if I'd performed the same test in low light conditions, the dSLR would have won, but I was surprised it didn't win the popular vote in good light!

Time has moved on, and I wouldn't be surprised if people preferred my current cameraphone (Huawei P20 Pro) in twilight now too, due to handholdability via computational photography. In darkness the "proper" camera will still win, but the boundaries are being pushed further every day.

That's one area that continues to perplex me; on average the mainstream camera manufacturers seem to be ignoring computational photography, whilst the phone manufacturers are embracing it in a big way. I can't help but think that if the camera manufacturers applied the same tricks, they'd be able to maintain more clear space between themselves and the cameraphones...

Case in point, my Huawei has a handheld night shot mode, where it takes a run of underexposed shots and aligns and averages them - the result is that I can reasonable shots handheld in cases where I'd struggle to get reasonable results handheld with my real cameras, even with IBIS, without the ISO going stratospheric.

Clayton1985 Veteran Member • Posts: 8,384
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII
2

Iantau wrote:

The only consolation that i had (id just bought the A7RII and a suite of lenses !) was that the A7 had more latitude for recovering from "operator errors" and was much less demanding of good technique in low light.

And yet no one could see this in your prints?

Iantau Regular Member • Posts: 101
Re: (Fun) Camera Club Comparison: Sony A7R vs Olympus EM1mkII

Clayton1985 wrote:

Iantau wrote:

The only consolation that i had (id just bought the A7RII and a suite of lenses !) was that the A7 had more latitude for recovering from "operator errors" and was much less demanding of good technique in low light.

And yet no one could see this in your prints?

ah...  all of these prints were done from well focused, well exposed, stably held/tripod shots, so "operator errors" wern't a factor.

the reality is that in most cases, when considering well exposed, well focused shots, the minor differences between cameras that we can see (zoomed in) on screen are not visible when the files have been put through the printing process - even with reasonably large prints.

Of course my point is irrelevant if your endpoint isn't a physical print

I see that the quote above can be a bit unclear - the 'A7' that I'm referring to is an A7RII.

-- hide signature --

Ian
'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´'·.¸¸.·´¯'·.¸.·´¯'·.¸.·´
> > > > >

Keyboard shortcuts:
FForum MMy threads