Mix of Colors, Halcyon – Austin, Texas
I’ve been running an experiment lately, which I’ve allude to on some recent posts. It all started with photographs I shot and posted at the beginning of this month. I was creating pictures from a single image, which I didn’t expect I could make. In the past, I resorted to techniques like HDR to get the look, detail and color that I wanted. Except, I was able to do this now with only a single photograph. You can see what I’m talking about by clicking on my atmtx logo, up top. Look at the urban landscape photos that I posted in early April.
I was getting far better dynamic range that I ever expected to, from the comparably smaller micro 4/3 sensor on the Olympus cameras. I’ve long stopped worrying about high ISO noise, which on modern, interchangeable lens cameras, look great for reasonable shooting. Rather, it’s increased dynamic range that I was craving.
I suspect that modern micro 4/3 sensors have progressed to a point of remarkably capable dynamic range, especially when unlocked by good RAW processing software. When I switched to Capture One, that’s when I really discovered what the Olympus can do. The other important point that many people miss is, getting a good technical image is more than just the sensor. The lens is certainly key as well as the unsung hero, image stabilization.
What I suspected from my casual shooting was confirmed by a Fuji GFX 50S review I read on PetaPixel. Dynamic Range, while better with the bigger sensors, wasn’t as large as I expected. Look at the table below. I got the data from the review.
| Fujifilm GFX 50S | Olympus E-M1 Mark II | Sony a7R II | Fujifilm X-T2 | ||||
| ISO | DR | ISO | DR | ISO | DR | ISO | DR |
| 50 | 11.90 | 63 | 9.80 | 50 | 11.42 | – | – |
| 100 | 11.90 | – | – | 100 | 11.42 | 100 | 9.79 |
| 200 | 11.07 | 200 | 9.82 | 200 | 10.24 | 200 | 9.81 |
| 400 | 10.41 | 400 | 9.32 | 400 | 9.61 | 400 | 9.08 |
| 800 | 9.39 | 800 | 8.38 | 800 | 9.32 | 800 | 8.57 |
| 1600 | 8.60 | 1600 | 7.47 | 1600 | 8.52 | 1600 | 7.60 |
I can’t confirm how accurate these measurements are. I just copied the values from the review. However, regardless of the numbers, there are known facts — larger sensors have better dynamic range and dynamic range falls with increased ISO.
Assuming these numbers are reasonable, here’s a couple of interesting tidbits:
1. Dynamic range on the larger medium format Fuji GFX 50S is not that much higher than the Sony a7R II.
2. A big surprise is, the Fujifilm’s X-T2 with a APS-C sensor, is no better in dynamic range than the Olympus with the smaller micro 4/3 sensor. This is sure to be controversial. Again, I’m not sure if this is true, but I’m just referencing the data. Perhaps the X-Trans sensor yields lower overall dynamic range than Bayer sensors?
But here’s the important point, which is the crux of this blog post. Notice that dynamic range drops for all sensors, regardless of size, when the ISO goes up. Sure, if you want to shoot the highest quality image, use the Fujifilm GFX 50S at ISO 100 and put it on a tripod. But, what if you can’t use a tripod or you prefer a lighter weight setup that yields slightly lower but still excellent results?
Notice that the Olympus, with the much smaller micro 4/3 sensor, shot at ISO 200, has better dynamic range than the medium format Fuji shot at ISO 800. ISO 200 to 800 is just a two stop difference. A difference that can easily be overcome with in-body image stabilization. Which means, if I didn’t use a tripod, the Olympus can beat the dynamic range of the medium format Fuji, in darker conditions.
The Olympus currently has the best image stabilization in the business. The OM-D E-M5 Mark II, that I used to take the photo above, is rated with a 5-stop 5-axis image stabilizer. In real world use, I often get at least 3 – 4 stops of stabilization. I shot that image with a 50mm equivalent lens at 1/6 of a second, tack sharp. The newest Olympus OM-D E-M1 Mark II, with certain lenses has up to a 6 1/2 stop IS rating.
So what does this mean for real world image making? I can shoot images like the one above, or yesterday’s blue hour skyscraper photo without a tripod. Between the image stabilizer challenged Fuji GFX 50S with slower lenses, I can probably outshoot the medium format camera with my Olympus, at night.
Image stabilization is not going to work for every type of photo, of course. But it goes to show how critical it can be for certain types of photography. And with a lighter and smaller camera, without a tripod, it’s a heck of a lot more fun.
So what’s the experiment I’ve been running? I’m shooting pictures, with slow shutter speeds, using the in-body image stabilization (IBIS) to get a lower ISO. If the situation permits, and I can use IBIS to lower my ISO to a 200 – 400 range, I can create very high quality images. My tests indicate that I’m getting very satisfying photos using this technique. I’ll post more examples, over the next few days.
Please support this blog by clicking on my Amazon Link before buying anything.
atmtx photo blog 
I’m a little confused by the experiment you mentioned. Haven’t you always used the Olympus in-body IS for night shots? What’s the new part – this particular Olympus body?
Thanks for the question, Mike. Yes, I have always used IS for night shots but usually accepted higher ISOs (800 or higher) to get faster shutter speeds.
Now I’m pushing for even slower shutter speeds to try to bring down ISOs to a 200 – 400 range, to optimize dynamic range.
There’s been exceptions, for example, when I was purposely trying to create motion blur.
Of course, in dark conditions, I’ll need to use big apertures too, in addition to extremely slow shutters, in order to shoot at ISO 200 – 400.
Ah, OK. That’s cool that you are able to do that with the Olympus in-body IS but I wonder how meaningful the isolated variable of dynamic range really is when comparing m4/3 sensors to medium format. There are a lot of other variables that add up to “image quality”, something that ends up being a very subjective construct anyway. For example, pixel size is 5.3 microns on the GFX vs 3.36 microns on an Olympus Pen-F. That’s a 60% increase in size of the sensor’s light gathering elements and there are a lot more of them. This means that you end up with a higher resolution file and, presumably, lower signal-to-noise ratio (I’m not enough of a camera geek to chase down the numbers). Would the GFX end up looking better at ISO 800 than the Olympus at ISO 200 or 400? I dunno. Numbers can tell one story but I don’t think dynamic range alone makes for an apples to apples comparison. The important thing is that Olympus appears to be giving you what you want in low light without carrying a tripod.
Yes Mike, there are other considerations to technical image quality than dynamic range. There’s color, resolution etc. For portraits, shallower depth of field might also be a plus.
Notice on the post I didn’t say that the Olympus will beat the Fuji GFX 50S’s image quality. I just said it can match the dynamic range under real world conditions, without a tripod.
As for deeper depth of field, in landscapes, I think having inherent greater depth of field, with a smaller sensor, can be a plus. Ditto for street photography.
Much of this depends on what you are shooting and the parameters of what you want to do to create the image.
Hmm…I wonder if the stabilized camera in my iPhone 6S Plus matches the dynamic range of a medium format camera at higher ISOs.
If it were only that simple. Maybe someday.