15
\$\begingroup\$

I am thinking about either buying a Nikon d3400 (APS-C or an Olympus OM-D EM-10 mkii (M4/3).

In principle I would prefer the Olympus, as it is much smaller and could fit in a jacket pocket. However I am concerned this lack of size may come at the cost of image quality due to the smaller sensor size. So how noticeable would the difference be?

\$\endgroup\$
9
  • 3
    \$\begingroup\$ Can you be more specific about what you mean by "image quality"? This term often gets bandied about as if there were a standard understanding, it's really a complicated thing with a lot of nuance and subjectivity. (What's the image quality of Rouen Cathedral: The Facade at Sunset?) \$\endgroup\$
    – mattdm
    Jan 21, 2018 at 23:13
  • 1
    \$\begingroup\$ You have two big tradeoffs with smaller sensors - low light performance and DOF control. All else being equal, a bigger sensor will bag you some shots in low light that you would have a harder time getting good exposure for with a smaller sensor. If you're into shooting with strong DOF contrast, these shots are harder to achieve with smaller sensors. Are these factors for you? \$\endgroup\$
    – J...
    Jan 22, 2018 at 12:56
  • 3
    \$\begingroup\$ @J... Please put your answers in the answers section, even if they're short \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 0:12
  • 2
    \$\begingroup\$ No, it's an answer followed by a question for clarification that could have stood on its own without the previous explanation. A request for clarification is something like, "In what conditions do you usually shoot?", "Do you shoot in low light often?", or "Do you want to get very shallow depth of field?" Answering the question and then tagging on, "Are these factors for you?" is not a request for clarification. \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 0:41
  • 1
    \$\begingroup\$ @Belle Camera body size is only half of the total size/weight comparison. Larger sensors require larger lenses to get the same field of view and maximum f-number. \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 17:23

11 Answers 11

10
\$\begingroup\$

I'm sure you've heard the old saying, "The best camera is the one you have with you."

Some of my favorite photos are shots I've taken with my three-year-old Samsung Galaxy Note 4, a phone with a decent camera but not a spectacular one. But it's in my pocket all the time, and when there is only a moment to grab a shot, there it is.

You can certainly get some fine shots with a Micro 4/3 camera. Here's one I like:

Pumpkin Brownie

Pumpkin Brownie

(Olympus E-P5, Olympus 75mm f/1.8 lens, crop of JPEG straight from the camera.)

One thing I would ask yourself, especially for mountaineering, is what kinds of lenses you plan to carry. Do you plan to bring several prime lenses, or just one zoom or two? Even more than the sensor, the glass is what will make the difference.

And what's your tolerance for extra bulk and weight vs. convenience and light weight?

One nice thing with Micro 4/3 is some of the specialty lenses you can get for it that are both light and compact and inexpensive too. The Olympus 60mm macro is a real treat to use, as is the Samyang fisheye (also sold under a few other names; mine is Rokinon) that sells for less than $300.

I would even consider going a notch smaller. You can get some very nice compact cameras like the little Fujis or Sony RX100 series that fit into a tiny case and take great photos. I have an RX100M2 that I use a lot - great little camera with a not so great menu system. For example you have to dig around to get to the place down in the menu where you can aim the camera at a gray card and shoot it to set the white balance - something I do all the time.

You may want to try renting some equipment so you can really give it a good workout and see how you like it. BorrowLenses.com and LensRentals.com are a couple of reputable rental outfits (they do rent cameras as well as lenses).

\$\endgroup\$
20
\$\begingroup\$

A great deal here depends on when you (generally) take pictures.

In particular, with bright light, a smaller sensor makes little or no difference in quality. As the light level drops, however, a large sensor (generally) gains a greater advantage.

So, if you're mostly taking pictures of the view from a mountain top in broad daylight, chances are that the Olympus will work beautifully (and even the much smaller sensors in most cell phones will also work quite nicely).

On the other hand, if you were to take some pictures at night by the light of a camp-fire (to give only one obvious example) the differences due to sensor size will be much more noticeable--quite possibly to the point that you want to think hard about the larger sensor.

If it were up to me, I'd think hard about a mirrorless camera with an APS-C sensor. For one example, the Sony A6300 has an APS-C sensor, but is still very close to the size and weight of the Olympus (12.7 ounces vs. 12.4 ounces for the Olympus).

As a disclaimer: no, I'm not really trying to push Sony in particular--as it happens I own a Sony camera, so I'm a little more familiar with what they offer than the other brands. Canon and Nikon (for the two most obvious possibilities) offer mirrorless cameras as well. I believe the Canons use an APS-C sensor (though Canon's version of APS-C is a bit smaller than everybody else's) and the Nikons use one that's substantially smaller (even smaller than four thirds).

That does translate to smaller size and weight though--for example, the Nikon 1 series bodies are around 11 ounces apiece. If you shoot (at least primarily) in situations where a smaller sensor will work well (and given that size and weight are at an extreme premium for you), it might be worth considering something even smaller than four thirds.

\$\endgroup\$
7
  • \$\begingroup\$ I think you missed a word on your disclaimer: "no, I'm (not) really trying to push Sony in particular"? \$\endgroup\$
    – Andrew T.
    Jan 22, 2018 at 7:06
  • 1
    \$\begingroup\$ I thought the Nikon 1 was out of production? If you don't need interchangeable lenses, Sony makes a couple of 1" models. \$\endgroup\$ Jan 22, 2018 at 18:07
  • 1
    \$\begingroup\$ @markransom Panasonic and Canon also make 1" sensor cameras. \$\endgroup\$ Jan 22, 2018 at 20:43
  • \$\begingroup\$ @MarkRansom: Nikon still lists 1 series cameras on their web site, but I don't know beyond that. Some appear to be discontinued ("this product has been archived") but at least a couple seem to be current. \$\endgroup\$ Jan 22, 2018 at 22:41
  • \$\begingroup\$ @StephenG I only mentioned Sony because I have personal experience. I suspected Canon did as well but wasn't sure, didn't know about Panasonic. Thanks for filling in. Do any of those have interchangeable lenses like the Nikon 1? \$\endgroup\$ Jan 22, 2018 at 22:44
16
\$\begingroup\$

In practice this is not a concern unless you have very demanding needs.

Now I would preface this by saying that my view of "image quality" is that many people, particularly beginners, tend to make the mistake of thinking of that in terms of pixel level quality or technical tests of particular parameters (like ISO performance).

In practice an "image of quality" is a result of knowing how to take a good photo. The best photos are about using light, available and artificial, using the right shutter speed and aperture, using the right focal length and framing. It's generally not about pixels - people don't look at pixels, they look at complete images.

So if you want good photos, learn the basics of technique.

But even in terms of technical capability, the gap between ASP-C and m4/3 sensors is not particularly large these days. In fact the capability of any current m4/3 sensor is better than most of the ASP-C DSLRs I have used over the years.

The m4/3 sensors are, in practice, large enough to provide good high ISO and modern ones have good dynamic range. They are not quite as good in terms of allowing very strong out of focus shooting, but in practice they are very good (and far, far better than any ordinary compact). Most beginners to large sensor systems will find the m4/3 a shock to the system in terms of having to deal with narrow depth of field as a potential issue when shooting - all that out of focus blur is a double edged sword !

And there are pros using m4/3.

I do a lot of mountaineering, and a small camera would be a big advantage.

Either system will work, but I'd edge the m4/3 if your aim is mountaineering, not because of the camera size, but because the lenses tend to be smaller on average than equivalents on the APS-C. This is because lens size scales with sensor size - bigger sensor, bigger lens. If all you ever use is a kit lens, then this is not an issue because the Sony APS-C MILCs have compact and reasonably capable small kit lenses. However for more serious work, this becomes an issue.

\$\endgroup\$
9
\$\begingroup\$

There is without a doubt a noticeable difference. The smaller sensor size, as you mentioned, gives a Micro Four-Thirds camera a disadvantage when it comes to low-light performance. The real question is: How much is this difference?

Let me preface this by saying that I have seen and reviewed nearly every Micro-Four Thirds cameras on the market, as well as most APS-C offerings from entry-level ones such as the D3300 to the latest D500 (plus Full-Frame DSLRs and one Medium Format System), so I know what the difference is by experience. You will find most of my reviews on my own Neocamera, although I also review for third party websites and print publications.

What I have been observing, regarding Micro Four-Thirds vs APS-C, is that the performance has been closing until about a year ago when high-end APS-C DSLR took a significant leap. This means that the difference between an Olympus OM-D E-M10 Mark II and a D3400 is less than between an E-M1 Mark II and a D500. Here is an image illustrating the point:

enter image description here

On the left you have the current top-of-the-line Micro Four-Thirds camera, the Olympus OM-D E-M1 Mark II, in the middle the Nikon D5500 which entry-level DSLR and, on the right, the top-of-the-line Nikon D500. As you can see, the first row is taken at ISO 200 and noise levels is very similar. The E-M1 Mark II has a tiny bit more than the others but it would not be noticeable on a typical print. The next row is taken at ISO 1600 where you can see that both the E-M1 Mark II and D5500 become visibly noisy, although you can make out details better on the D5500. On the other hand, the D500 still looks nearly impecable. The final row is taken at ISO 12800 where it is easy to see that the Olympus is way behind the rest. The D5500 got softer which is Nikon's way of hiding its noise. There is some of this on the D500 but its image remains much more usable.

There are different aspects to image-quality and certain circumstances show issues more than others such as handheld low-light photography which requires high sensitivities. When it comes to noise, the current crop of Micro Four-Thirds camera behave similarly to APS-C ones at low sensitivities, showing virtually no noise until ISO 400. As ISO is increased, both sensor sizes start adding noise but APS-C cameras really manage to keep noise low much longer. Once in a while, I do a comparison just to see where the systems are at. Here is an interesting one - although slightly outdated - pitting the Fuji X-T1 (APS-C mirrorless) against a Panasonic GH4 (Micro Four-Thirds mirrorless).

Dynamic-range is always an issue. The top-of-the-line Micro Four-Thirds cameras are still about 1 1/2 stops behind in dynamic-range compared to even older APS-C cameras. This will be a problem when shooting highly contrasting scenes.

The point is that there is a difference and a clear advantage in terms of image-quality of APS-C digital cameras but there are also great advantages to a Micro Four-Thirds camera, most notably size and weight. For the ultimate in image-quality you will need a larger sensor (even Full-Frame) but if image quality from Micro Four-Thirds cameras is good enough for your needs, you might as well consider all the advantages you are getting.

Feel free to read my review of the Olympus OM-D E-M10 Mark II and do check out its gallery. It has full-resolution images at all ISO taken right from the camera. The images will speak for themselves.

\$\endgroup\$
7
  • \$\begingroup\$ Any idea why the DR difference would be 1 1/2 stops when the light gathering ability is only 3/4 stops different? Not that I doubt you, I'm just curious and seek enlightenment. \$\endgroup\$ Jan 22, 2018 at 21:17
  • \$\begingroup\$ A factor might be the smaller pixels get more read noise due to density. \$\endgroup\$
    – Itai
    Jan 22, 2018 at 21:29
  • \$\begingroup\$ @MichaelClark that's what I based my figures on. Taking log₂ of the ratio of the areas to give stops yields 1.17 for FF to APS-C and 0.77 for APS-C to m3/4. \$\endgroup\$ Jan 23, 2018 at 2:20
  • 1
    \$\begingroup\$ @MichaelClark, lg(0.586) = -0.77. Your calculations are in perfect agreement with Mark's. \$\endgroup\$ Jan 23, 2018 at 15:20
  • \$\begingroup\$ @MichaelClark converting a ratio to a number of stops requires taking the log₂ of the ratio. A sensor with twice the area will gather one stop more light, correct? log₂(2) = 1 stop. log₂(4) = 2 stops. log₂(384/225) = 0.77 stops. \$\endgroup\$ Feb 3, 2018 at 5:21
5
\$\begingroup\$

Since most of the answers here are kind of the general "trade-offs related to sensor area" type, I'm going to add something that I haven't seen mentioned: the difference in the aspect ratio might not be an insignificant consideration, depending on what you want to shoot. If you like the native 3:2 ratio of APS-C, because you shoot a lot of landscape or just because you like the look, you might be better served by that format, since cropping a 4:3 shot down increases the disparity in sensor size between the two systems.

On the other hand, if you like a native 4:3 ratio, maybe because you like to take a lot of portrait-oriented shots that are just too narrow in 3:2, or just like composing in something a little closer to square, it's worth noting that cropping an APS-C sensor to 4:3 decreases the difference. I've attempted to visualize this in the graphic below:

enter image description here Relative dimensions of FF, APS-C 1.5/1.6, and 4/3, with aspect cropping lines.

\$\endgroup\$
14
  • \$\begingroup\$ Cropped to a 4:3 ratio, a 1.5X APS-C sensor would be 21.33x16 mm, for an area of 341.33mm², compared to the 225mm² of a 17.3x13 mm µ4/3 sensor. That's still a significant difference. \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 0:45
  • \$\begingroup\$ @MichaelClark Stating the area that way exaggerates the case of the dimensional difference in the same way as MP numbers do, in my opinion. Anyhow, I've added a graphic that should clarify my primary point: that normalizing for aspect ratio from one direction amplifies the difference, while doing from the other direction diminishes it. \$\endgroup\$ Jan 23, 2018 at 3:42
  • 1
    \$\begingroup\$ @MarkRansom To be clear, I wasn't disparaging the existing answers, just pointing out an... ahem, aspect of the situation that hadn't really been addressed yet. Sorry about your troubles with the arbitrary available print sizes, but keep in mind that the importance of this can vary from "very much" to "not at all" for any given person. ;) \$\endgroup\$ Jan 23, 2018 at 3:56
  • 1
    \$\begingroup\$ @junkyardsparkle How is comparing the actual area one uses from one sensor to get a 4:3 image to the actual area of a 4:3 sensor exaggerating the case? A 1.5X APS-C sensor measures 24x16 mm. If we crop it to 4:3 we use 21.33x16 mm. The area of that part of the sensor we use is 341.33mm². The area of a µ4/3 sensor is 225mm². \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 4:44
  • 1
    \$\begingroup\$ @fkraiem The answer addresses one of the factors that directly affects final image quality. This is perfectly acceptable at stack exchange. Every answer does not have to completely and comprehensively answer every facet of the question. The standard is that the information presented in an answer should be helpful. \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 17:21
3
\$\begingroup\$

Yes, there is a difference in image quality. If I want the ultimate quality, I don't use my m43 kit - or my DX/APS-C either, I use full frame. The m43 has slightly less dynamic range, slightly worse low light performance, slightly harsher colour, slightly less sharpness (though, in fairness, my m43 glass mostly isn't as good quality as my SLR glass).

Note the slightlys that pepper that though. And note that, even though I've got bigger chip gear, I still keep and use the m43 as well.

The best camera is the one you actually have with you. My m43 kit is smaller and lighter, and goes with me almost everywhere because of that. My SLRs come out to play when I expect to need them and don't mind hauling the weight. APS-C mirrorless is slightly smaller than an SLR kit, but not that much with lenses.

You mention mountaineering. My sport's cycling, and I'm much happier riding with m43 kit than big chip - the weight and size difference is very noticeable.

You also mention 'entry level'. In my experience, the entry level kit from all brands is now quite good enough for the likely uses anyone getting started will put it to - there are differences, sure, but they're so small as not to be worth worrying about until you start pushing the system more than you're likely to for a little while.

If it were me, I'd buy the camera I'd happily sling in a bag without thinking about it, which is the m43 (and look at the tiny Panasonic GX800 as well as the E-M10 III - it's got less physical controls so slower to use, but it's significantly smaller and lighter) - it'll get you decent results and you'll take it with you more. If you find it's a limiting factor later, you can always upgrade then with more knowledge of what you actually need from the experience.

More examples of what you can get from (now quite old) m43 and a kit lens - https://www.flickr.com/photos/gpwebb/16421348901/in/album-72157629207953363/ https://www.flickr.com/photos/gpwebb/8964546310/in/album-72157629207953363/

\$\endgroup\$
2
\$\begingroup\$

I have the Olympus E-M10 Mark II, and a Canon Rebel T5 (probably not exactly fair, the T5 is a bit older)

Image quality with the indoor lighting is definitely comparable.

I've tried an outdoor night shot, side by side. Olympus with my Panasonic 25mm f/1.4, Canon with the 50mm f/1.8. I tried keeping everything the same, ISO, aperture (f/1.8) and shutter speed (at around 1/50s). The Olympus turned out better with the IBIS, where as Canon didn't have IS, and it became tricky with my overcaffinated hands.

I've had other shots where I did better holding the camera steady, and there wasn't that much difference.

For certain kinds of low light shots, you could use image stacking to average out the noise. Overall, at least with the two cameras that I have, I don't find that the Canon APS-C blows the Olympus out of the water at all.

\$\endgroup\$
2
  • \$\begingroup\$ For outdoor night shots, a tripod (or other stable platform) should probably be used with either one. In that situation, you're not comparing sensors, you're comparing stabilization. \$\endgroup\$
    – Michael C
    Jan 23, 2018 at 0:28
  • \$\begingroup\$ That's a fair comment. I was trying to do continuously shots for image averaging, and the Rebel didn't have mirror up, which even if I break out the tripod, it won't be perfect. \$\endgroup\$
    – Calyth
    Jan 23, 2018 at 0:33
0
\$\begingroup\$

The previous answers are all good, but I'd like to add another information.

If you are interested in the finest details, always remember that the smallest detail in the sensor is limited by diffraction and the spot has the same size, in microns, of the aperture value. So if you shoot at f/2.8, the smallest detail is 2.8 microns on the sensor. This is valid with perfect lenses.

Now, smaller sensors have typically smaller pixels: my APS-C D7100 has 3.9 microns per pixel, a D750 has 6 microns and a micro 4/3 OM-DEM-10 has 3.9 microns. Medium format may have bigger pixels.

In your case both with APS-C and micro 4/3 you can "fully exploit" (theoretically!) the resolution of the sensor when shooting at f/4 or brighter. Full frames can freely shoot at f/6 without loss of details.

Of course, producing small lenses with shorter focal lengths is more difficult, so the lens could be the limiting factor on micro 4/3 compared to APS-C.

\$\endgroup\$
-1
\$\begingroup\$

I don't think the difference would be that great, but, there are few things to consider:

Olympus has only its m4/3 system, so there's no moving from there if you ever feel the need to move to FF for instance. You'd have to switch brands. Also, Olympus has a smaller selection of accessories and lenses available for their cameras to choose from.

In my opinion Nikon is still a more versatile option as long as you don't necessarily need the compactness of a mirrorless camera. You can naturally progress to higher end bodies and you can 100% rely on the tool if you ever go pro. Don't get me wrong, I think Olympus has some nice tech, but if you're aiming for the best image quality, versatility etc., traditional DSLRs are still the way to go. One day the story will be different, but right now mirrorless is really mainly for people who need a compact body and even that isn't completely true, because as you get closer to features of a DLSR, the size and weight grows considerably. When my friend came over with his mirrorless camera to my place last year, funnily enough, my 6D with an L-lens was lighter and smaller than his Olympus...

\$\endgroup\$
10
  • 1
    \$\begingroup\$ @user5227744, having a smaller sensor always sacrifices some image quality. Go through some online photo galleries to see the example photos, but I don't think you would notice any huge difference unless you know what to look for. If you need a compact camera and you're fine with Olympus' downsides, go for it :) (Mind you this is a very opinion based question/answer and as such isn't really fit for the format of this site.) \$\endgroup\$
    – walther
    Jan 21, 2018 at 22:17
  • 4
    \$\begingroup\$ I'm resisting the urge to downvote, but I'm really not a fan of the "if you go pro..." comments here. Walther, I notice from your profile that you're not a professional photographer, but, meanwhile.... \$\endgroup\$
    – mattdm
    Jan 21, 2018 at 23:16
  • 3
    \$\begingroup\$ I'm judging your words based on... your words. I don't think we should encourage people to make decisions based on "maybe I'll go pro". I didn't mean anything disparaging based on your profile, but simply that if you were a professional photographer that would lend the weight of experience to that part of your answer. \$\endgroup\$
    – mattdm
    Jan 22, 2018 at 0:13
  • 1
    \$\begingroup\$ I disagree with traditional DSLRs being the way to go. I'd pick a Fujifilm X-T2 over a DSLR in the same price range any day. They've got many advantages. You see exactly how your picture will turn out. How many times I've seen fellow photographers picking the wrong ISO on their DSLR for a fireworks show during a festival and only notice after it finished... Many more focus points. Faster autofocus. Smaller, lighter. And you're not restricted to APS-C. Leica M is full-frame. \$\endgroup\$
    – Belle
    Jan 22, 2018 at 7:30
  • 5
    \$\begingroup\$ Going with an Olympus Micro 4/3 doesn't make one stuck with only Olympus — it buys the person into the Micro 4/3 system. Panasonic's, or any other brand's, MFT lenses are completely compatible. Additionally, a selection of MFT lenses (of any brand) can be used on cameras such as BlackMagic's pocket cinema cameras, etc. MFT is a very versatile system. And boatloads of existing DSLR lenses can mount to MFT bodies with an adapter. Whereas with Nikon bodies, you can mount.... F-mount lenses, and that's it. (Disclosure: I'm a Nikon FF shooter). \$\endgroup\$
    – scottbb
    Jan 22, 2018 at 23:26
-1
\$\begingroup\$

The area of a Nikon APS-C sensor is 384 mm2, and the area of a Micro 4/3 sensor is 225 mm2. The ratio is 1.7, which you can convert to stops with log₂ yielding about 3/4 of a stop. Not a huge amount if the sensor technologies are equal, and for modern cameras I'd assume they're extremely close. This is less than the difference between full frame and APS-C.

If you can get a 1-stop faster lens on the micro 4/3 you should be able to use half the ISO of the APS-C and find there's no IQ penalty at all.

\$\endgroup\$
-2
\$\begingroup\$

The Nikon is a DSLR, Olympus is a mirrorless camera.

Mirrorless is far smaller and (I firmly believe) is the future. If cameras had been invented today, no-one would enineer a moving mirror.

While you focus on image quality, consider the number of shots not taken. Because you didn't carry your camera. Because is was too big.

Buy a camera that fits in your pocket, and you'll be much happier. If you want an APS-C, don't go for a DSLR but buy a Sony a6xxx. I have the Olympus myself, but the Sony is a very nice and good and popular camera - and even slightly smaller.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ This doesn't answer the OP's question, which asks about the technical merits of the MFT sensor vs. an APS-C sensor. The OP already understands the size issue (it was stated in their question). \$\endgroup\$
    – scottbb
    Jan 23, 2018 at 13:35

Not the answer you're looking for? Browse other questions tagged or ask your own question.