Micro four thirds and ISO noise. What's the big deal?

If you've ever even vaguely considered changing to the micro four thirds system, you will have composed a list of pros and cons. If it's an Olympus OM-D system like the E-M1, then on the pro side you might have listed features such as; weight-saving, excellent Zuiko optics, 5-axis ibis, weatherproof, evf, etc...

Then on the cons list, there's probably things like; smaller sensor, poor low-light ISO performance, limited video capabilities, larger depth of field (compared to bigger sensors), smaller printing sizes.... stuff like that.

Seems like a genuinely accepted list of pros and cons when weighing up the Olympus OM-D micro four thirds system to other systems and sensor sizes.

Except it's not. Not really. Because when it comes right down to it, most of the 'features' listed above - both the pros and the cons, are purely 'subjective'. And yet most photographers, reviewers and 'influencers' (naming no names) speak about these things as if they are objective truths. They're simply not.

Even the OM-D 'faithful' constantly feel the need to apologise for the 'poor low light performance' of micro four thirds. I think I've heard the phrase 'poor low-light performance' in relation to micro four thirds used about a dozen times this week on Youtube alone. And that's from confirmed Olympus users. Why all the constant apologising over the noise quality from a micro four thirds sensor? Is it really all that bad - objectively? Or, is it purely subjective? To quote Shakespeare - "the lady doth protest too much, methinks".

Olympus OM-D E-M1 with Lumix 25mm. f/5.6 @ 1/125th, ISO 12800. Noise reduced in LR

When I heard the 'poor low light performance' line trotted out again today as a 'matter of fact', this got me thinking. A lot. Because to be absolutely truthful, I've never - and I do mean never - thought that my E-M1 had poor low light performance. And that's a fact. But, it's also subjective.

I'm not stupid (honestly), or ignorant. I realise that the phrase 'poor low light performance' is made in comparison to full frame sensors. So it's relative. And yet, when it is constantly used as a 'con' of the micro four thirds system - a negative aspect of the system as a whole - then it makes prospective owners or potential users of the system stop and seriously question whether to even bother or not.

From the dawn of digital photography we've been 'taught' to fear NOISE in our images. It's to be avoided at all costs. And, to be fair, some of the very early first-gen sensors were pretty awful above ISO 400. But that was 20 years ago - an absolute age in technology. Today's sensor 'noise' isn't even close to what it used to be when noise was a dirty digital word. Somehow (don't ask me how), manufacturer's have managed to make digital noise a lot more 'organic' looking - more (dare I say it) like film grain that hi-fidelity noise.

And besides which, noise is one of those 'features' that is highly subjective. Some photographers don't mind it. Some embrace and use it (much like people embraced and used grain). And some will delete any image that shows even a hint of it! As I've already said, I've never looked at any of the images I've taken with my OM-D cameras and thought they were too noisy. Or suffered from poor low light performance. Just never.

But I have also come from a film photography background. Where ISO 800 film was crazy-fast, and shot only rarely. This kind of thinking around ISO's has been ingrained into my photography. Even with digital. You could hand me a Nikon D3, or a Canon 1Dx with crazy low-light performance, and I'd still only shoot it at an absolute maximum of 3200 ISO (and probably only up to 1600 if I'm honest).

Back in my film days I tried an ISO 3200 Ilford black and white film once. Just once. It was horrible. And I mean horrible! Grain the size of golf balls. The prints I made from that film didn't look like photographs - more like those black and white sand drawings that do the rounds on Youtube. Compared to that, ISO 3200 on my E-M1 micro four thirds camera is practically grainless! So does micro four thirds have 'poor low light performance'? Yes - and No. What are you comparing it too? And what is your individual tolerance for noise in an image?

Above is a comparison of the ISO performance of my OM-D E-M1. I know it will be difficult to tell from this image on the web, so I'll just give my synopsis of what I see at 100% on my computer screen.

With a well-exposed image (exposing for the right), ISO 800 is very clean. Not noise-free, but so close as to not be an issue at 100%. The same is true of ISO 1600, and maybe even 3200 - again depending on your proclivity for noise in an image. While noise is apparent at ISO 3200, it would clean-up very easily in the noise-reduction programme of your choice - so I still don't see it as an issue.

As far as I'm concerned, you can stop right there. A relatively clean, eminently useable (for social media and A3 sized prints) image at ISO 3200 is an amazing result for micro four thirds if you ask me. I'm a happy camper. So no, I don't concur with the 'poor low light performance' appraisal of micro four thirds - at all.

But as you can also see from above - the E-M1 doesn't stop there. ISO 6400 is also useable when cleaned up - as is ISO 12800, remarkably! I'd stare clear of ISO 25600 - but there's no surprises there. The highest ISO setting on practically all cameras is only there for marketing purposes. ISO 25600 sounds good, so it's included as a selling point. That you would never shoot that high is irrelevant.

I did the same ISO test with my E-M10, and the results were identical, so I haven't bothered including them here. Based on what I see in these images, I would have no hesitation in shooting both cameras at ISO 3200 - and even 6400 if I had to.

Hand me a camera - any camera - and I'm always going to shoot it at its lowest native ISO, if I can. If I can't, then I will either use a tripod, add some light into the scene, or start to increase the ISO. Most of the time, for even the most extreme low-light scenarios I shoot, my ISO will never get past 1600 if I'm hand-holding. That's why I have a fast f1.7 prime in my bag. But usually I'm shooting from a tripod if I'm in very dark conditions. And then I can stay at ISO 200. 

You're mileage, and shooting scenarios, might vary. I get that. You might be a coalmine photographer who 'needs' to shoot clean images at ISO 2 million? If that's the case, then no, micro four thirds might not be right for you on those occassions (although it still might make for an excellent travel kit when you go on holiday with your family). 

The most common scenario I hear is wedding photography. "I shoot weddings, so I shoot a lot in low light. So I can't use micro four thirds" is the general gist of the conversation. I'm not here to dismiss their experiences, but I also often wonder if this is said by people who have ever even tried micro four thirds? I shot weddings for years, in churches, without flash, and I wouldn't hesitate to do so now with my E-M1 at ISO 3200. There are plenty (and I do mean plenty) of wedding photographers using micro four thirds who aren't being run out of town by angry clients who are unhappy with the amount of noise that are in some of their images. 

Let's face it. Who cares about noise? Really? It's us - isn't it. Photographers.Photographers who have been told, from the beginning of the digital era, that noise = bad! And so, because micro four thirds might be a little (yes, a little) bit more noisy that APS-C or full-frame sensors, then micro four thirds = 'poor low light performance' = bad. And while this might be 'objectively' true in terms of test charts, formulas and the laws of light and physics - 'subjectively' I don't see noise as an issue with micro four thirds. Period.

Sorry this has been such a long post/rant. But I felt it was an important topic to address, since it's high on the list of 'cons' given against using any micro four thirds system. The next time you hear someone talk about micro four thirds 'poor low light performance', please don't take it as the gospel truth. Try the system for yourself. Do your own ISO tests, and decide what level of noise you feel comfortable with.

For me, I'm happy to go as high as I would with any other camera system. Poor low light performance with my E-M1 and E-M10? Nah. Don't think so.