A year or so ago I sold my Nikon D90 and collection of fairly old lenses, left the Nikon fold which I’d been in since my film shooting days, and bought an Olympus OM-D EM-5 kit. I wanted something that I could put in a rucksack and take up into the hills when out walking or covering events with RAYNET. The Nikon was bulky and I was worried about bad weather, so the small weatherproof Olympus seemed ideal.
The kit, shown here with the 17mm f/1.8 lens, is small though fairly heavy with its metal construction. This is the kit used to take many of the other photographs used in this post.
My other consideration at the time was the Fuji X-T1 with its lovely user interface. I don’t know whether I’d have enjoyed the shutter dial with its 1 stop settings, but the controls are reminiscent of my old film SLR. I shoot RAW, so all the controls I care about at time of capture are exposure and focus. Other things come in post processing.
The X-T1 at the time I was looking had focus performance issues, later fixed in a firmware update. It lacks in-body stabilisation which is a great feature of the Olympus. I really noticed its absence when I left it turned off once. The X-T1’s viewfinder is amazing and I especially liked the depth of field indication when manual focusing. The simple controls would be great for teaching my son about cameras. Olympus’ two customisable dials with functions that change depending on mode are not beginner friendly.
The X-T1 and the then new EM5 MkII were expensive. My plan was to try the cheaper EM-5 and wait for an update to the EM-1. That update has since arrived, but the cost! The EM1 MkII currently sells for £1850 for just the body! And why not use a back illuminated sensor for that price?
For £1600 I can buy the Sony a7ii and 14-70mm kit lens. Sony were running a £250 rebate bringing that to £1350 and may do so again. I’d likely prefer to buy the body and a better lens such as the 24-70 f/4 or 55 f/1.8. This would give me a full frame camera with image stabilisation, 1 stop equivalent boost in speed on zoom lenses, and still quite compact form.
Why stay? What are the advantage of Micro 4/3?
Beyond the cost of changing…
A big advantage of Micro 4/3 is size. This is mitigated to some extent by the Sony which brings full frame into something of a similar size to the old 35mm film cameras. With modern mobile phones producing pretty good images there is less need for a small camera. More and more of my photos are taken on my phone. It can’t come close to the EM-5 in challenging lighting or allow the creative freedom of interchangeable lenses, but it can work well and is always there. An Olympus PEN with a small prime is really quite pocketable, giving amazing results for its size and a second body to an EM-1.
The other advantage of Micro 4/3 is the lenses. I have the Olympus 12-40 f/2.8 PRO which is a lovely lens. It gives me the same range as 24-80 on the Sony in a similar space to the 24-70 f/4. The 17 and 45mm primes are also very nice, making those three my lenses of choice.
The Sony’s smaller f/4 aperture allows a lens of similar size and cost to the Olympus. This pattern appears consistent. For example compare the new Olympus 25 PRO f/1.2 with a full frame 50 f/1.8. However the micro 4/3 primes are generally quite small.
Noise, Dynamic Range and Equivalence
Full frame has better noise performance at a given ISO setting, and more dynamic range than Micro 4/3. I’d hoped that improving sensor technology, for example back illuminated sensors, would keep Micro 4/3 competitive. Adorama claim that this technology can give a stop or so improvement. As it works by increasing the effective light collecting area, that improvement acts on shot noise. That is the unavoidable limitation of small sensors. The results in mobile phone tiny sensors are impressive. I’ve seen digital sensors improve significantly since my first DSLR.
So what do I have now with the EM5?
In Low Light
This shot perhaps benefits from the in camera stabilisation and slow moving subject allowing a relatively long exposure. I believe I asked the camera to spot meter on the vendor’s face. It has captured the wide range well. Yes there is noise in the shadow areas, but there is a lot of detail too.
I’ve recently taken a series of black and white photos in pantomime rehearsal, choosing black and white due to less interesting colour. Black and white allowed me to use techniques reminiscent of my work in film, throwing up the contrast and using dodge and burn to bring in details. This tests the dynamic range of the “film”, but the photos printed incredibly well on 9×6 matt paper.
At base ISO
This is a 100% crop representing a small part of an image near the centre. It is testament to the lens that I have this much detail, which is also replicated at the edges. This foggy low contrast scene emphasises noise, and the frost upset the auto metering. (I thought I’d dialed in 1 stop, so auto was 3 stops down!).
This is the full image. The man and dog are in the centre, very small.
Grain is visible in skin tones, and I wonder if the grain gives a flatter sense to the photos, though it is not clear at the sizes I print.
An f/2.8 full frame lens is huge, and incredibly expensive. But it may not be needed thanks to the effect of equivalence. My f/2.8 lens gives a similar out of focus blur to an f/5.6 lens on full frame, but that f/5.6 lens captures only a quarter of the density of light so the full frame camera will have to use a higher ISO setting to maintain a desired shutter speed. The full frame sensor has four times the surface area so the total light captured will be the same as the more intensely illuminated Micro 4/3, so if sensor technology is equal and neither system is saturated then noise performance should be the same.
So the considerations:
- 45mm f/1.8 in Micro 4/3 is much like owning a 90mm f/3.6 in full frame.
- A 90mm f/1.8 in full frame would be like a 45mm f/0.9 in Micro 4/3.
If I’m shooting for minimum depth of field then full frame wins here. It’s easy to get hold of an f/1.8 prime. Amongst the zooms the f/4 enthusiast offering is a similar cost and size to the f/2.8 Olympus PRO but effectively a stop faster. How do I calculate this? The following settings give equivalent depth of field, field of view and exposure. I’ve assumed that I want to keep the shutter speed constant to freeze subject of camera movement.
- Olympus 40mm f/2.8 1/100s ISO 200
- Sony FE 80mm f/5.6 1/100s ISO 800
The increase in ISO of the full frame system increases noise in that system, so compensating for the full frame advantage. But the Sony lens will give me f/4 if I desire or am willing to accept shallower depth of field, or I can reduce shutter speed.
It’s not all win for the Full Frame. I’ve found that f/2.8 gives quite good results and have found myself stopping down when photographing groups to ensure that all of them are in focus.
- 17mm f/4 Micro 4/3, hyperfocal range 2.16m – infinity
- 34mm f/8 full frame, hyperfocal range 2.16m – infinity
These two give the same field of view and impressive hyperfocal range. In both cases you can turn your focus dial to 4.5m and have anything more than a couple of strides away appearing in focus. f/4 and f/8 are chosen as the sweet spot in both systems. If you can afford to reduce shutter speed then you can take advantage of the lower ISOs on professional cameras and the improved performance at base ISO of the Sony. Though the modern Olympus cameras are very good at base ISO, and the temporal averaging of Olympus’ High Resolution shot gives very competitive noise performance.
Another reason to reduce aperture is to achieve low shutter speed, either because light is very bright or to deliberately blur motion. The limit in both systems is diffraction, and it occurs at different points. Here Full Frame has the edge with a diffraction limit two stops narrower than Micro 4/3. The depth of field is the same for both systems at their diffraction limit, but the Micro 4/3 system may have you reaching for the neutral density filter sooner. (Or the electronic shutter on modern cameras).
Lens Useful Aperture Range
Thanks to equivalence, the useful range of micro 4/3 lenses would have f numbers half the size of those for full frame – all else being equal. At the upper end diffraction becomes significant around f/11 for Micro 4/3, so should be around f/22 for full frame. The sweet spot would be f/4 and f/8 respectively. At the wider end it is possible to buy f/0.9 in Micro 4/3, but f/1.8 is common for primes, f/2.8 for premium zooms. Full frame has the edge here with readily available primes at f/1.2 (f/0.6 equivalent in Micro 4/3).
On the other hand, the premium Olympus lenses really are very good. Those promoting Micro 4/3 credit tele-centricity, an advantage of projecting the image onto a smaller sensor giving more parallel rays. This, they say, allows them to make lenses that give great quality across the image and do so at their widest aperture setting. I can use my 12-40 PRO at f/2.8 and on my current 16MPixel sensor it’s as sharp as can be. My 17/1.8 can be used at 1.8 with a little chromatic aberration, easily removed, but it’s amazingly clear.
It is said that the full frame lenses tend to need stopping down to be good, but the Olympus lenses can take advantage of that wide aperture and the ability to use a lower ISO setting. Here less shallow depth of field can be a help if you need more subject in focus in your shot. I remember encountering other aspects of a premium lens when I bought a second hand manual lens for my Nikon and noticed improved colours over my kit lenses.
I can look at the line-up of premium Micro 4/3 lenses in the shop window and not that I could fit a fair number of them into my old camera bag. Yes in some cases the Olympus are large, because making an f/1.2 25mm lens requires some size and ends up similar to the f/1.8 50mm it competes with. But I can carry a premium 150mm (300mm equivalent) zoom in a small bag. A budget version which isn’t that bad but a little too slow for indoor sports fits in a large coat pocket.
In the end I have to ask what the photos are for. I do enjoy the technical aspects of photography, but a picture depends on so much more than the camera. What I have now with my EM-5 is far more advanced, looking at old photos, than the film I had in the 80s and 90s. I believe my technique has improved since then too, at least in the dark room. When I look back at old images I don’t look at the technical aspects. I just see the photo. I’m only hyper-critical when I take them. Would I enjoy full frame more? I don’t know. I do enjoy the Olympus lenses.
My current plan is to wait and see. I’d like to hire a Sony to see what the difference is, both in terms of usability and results. The viewfinder on the older a7 I tried is nice, but any modern Olympus will also be better than the old EM5. I don’t plan to buy any more Olympus equipment at the moment (which means resisting a tempting offer on a second hand lens at the local camera shop!). I may instead look to investments like a lighter weight tripod. There is a limit to image stabilisation’s power and my old Manfrotto is heavy and bulky and tends to stay at home. I can still enjoy my EM-5 and practice trying to take great images with it. I see Micro 4/3 images on FlickR which are really good.
The EM-5 MkII is already a lot cheaper than when released. The same may happen for the EM-1 MkII. Phase based auto-focus is a requirement for my next camera, making the EM-5 MkII not a likely upgrade.
The EM1 is now available as a kit with the 12-40 and 40-150 for about £2000. That’s cheaper than buying both of those lenses without the camera! It doesn’t have the teleconverter sadly.
1 thought on “To stay with Olympus, or move to full frame Sony?”
I was able to download sample RAW files for the EM1, EM1 Mk II and Sony A7ii and A7rii from https://www.photographyblog.com. Full Frame does not eliminate noise. That should be obvious. I could bring faces out of shadow on the Sony files but they showed the kinds of noise I see on my E-M5. Maybe the Sony has more dynamic range, but moving to full frame is not going to eliminate it.
It would be nice to show some of the JPEGs from my RAW explorations, though that would be unfair to PhotographyBlog.