The Canon 1D X MK2 is the new flagship of Canon. So why does this top camera only have a 20.2 MP sensor?

I mean its 2016, even mobile phones have a higher resolution. I knew its not all about the resolution but only 20.2MP?

What reason does Canon have to only use this limited resolution? Which technical limitations lead to a decision like this?

up vote 45 down vote accepted
  1. All pixels are not equal Larger pixel wells, such as those found on a 20MP full frame sensor, are able to capture more photons than smaller pixels like those on a high resolution phone sensor. The pixel pitch for the EOS 1D X Mark II is 6.6µm. The pixel pitch for the Samsung Galaxy S5 is 1.12µm. That means that in terms of surface area the pixels in the 1D X II are 35X the size of the pixels in the Galaxy S5. This gives each pixel well the ability to collect 35X as much light before reaching full well capacity. This results in much better dynamic range, signal-to-noise ratio, low light performance, etc.
  2. Data rates The more pixels you have, the more information you have that must be processed and stored per image. Given the same limits in processing technology, cameras with the highest resolution take longer to process and store images than cameras with lower resolution. The flagship cameras from both Canon and Nikon are built as much for speed as they are for anything else. Try shooting sports at 12 frames-per-second with any camera with a 30+ MP sensor (Hint: with current technology, at anywhere above frame grabs of video that tops out at 4-8MP you can't without spending tens or even hundreds of thousands of dollars).
  3. Power consumption The more data that has to be processed, the more power it takes to process it. The buyers of such cameras such as the 1D X Mark II expect their batteries to last for thousands of images. Even with the large form factor of flagship DSLRs the batteries can only contain so much energy and that energy must be shared with everything else the camera does. Moving focus elements on lenses in which just the moving element weighs more than several smartphones takes more of the available energy than focusing smaller, lighter lenses does.
  4. End use of the images produced The primary buyers of flagship models have always been photojournalists and sports photographers. That application has never particularly demanded the highest available resolution. The images those folks produce are normally distributed at fairly low resolution. Newsprint is a very low resolution medium. Web distribution is also relatively low-res. Most web images posted on news sites are well less than 1/10th the 20MP size of the 1D X II's output.
  5. Pixels aren't the only distinguishing features of top end cameras Flagship cameras are as much about their durability and ability to withstand abuse in the field and still just work as they are about anything else. It doesn't matter how great the sensor in your camera is if one hard bump or drop renders it useless when you are in the middle of a jungle, desert, war zone, etc. and the nearest repair center is several days or even weeks away. Not only must they be able to survive in such conditions, they must also be able to perform under environmental conditions that would destroy lesser cameras. There are many other features and capabilities that are packed into flagship models that allow their users to capture the images they desire under a wide variety of conditions faster and more easily than they could without those features and controls.
  • 1
    Comments are not for extended discussion; this conversation has been moved to chat. – jrista Oct 5 '16 at 17:19

Compromises are made in everything. Look a little harder and you can find "top" DSLRs with many, many more pixels. Hasselblad H6D-100c has 100 MP, for example.

But looking at only megapixels gives you a very incomplete picture of what the camera can do. Directly tied to resolution is throughput: shooting a 20 megapixel image at 16 frames per second = 320 MP of data per second. (The 1D X mark II shoots that fast! Does your phone?)

Look at other cameras, too: just a few years ago (2013) Nikon's D800 had the crown for "full frame" megapixel count, at 36 MP. Awesome, but limited to 4 frames per second, or 144 MP of data per second. So, wow, the 1D X is handling roughly 2.5 times more data!

Also, review the math: to double resolution you need four times as many megapixels. That means the Hasselblad (with an additional 80 MP over the Canon) has just a little more than double the resolution of the 1D X. And despite an additional 16 MP over the Canon, the D800 has just 1.3x more pixels than than the 1D X. Not much of a difference, except when you need it!

  • Thanks, I didnt thought about the frames per second. The 320MP/s is pretty awesome. – rockZ Oct 3 '16 at 14:01
  • 1
    the D800 has just 1.3x more pixels than the 1D X => That's not true, the surface is 1.3 wider, but the ratio from 20mpix to 36mpix is not 1.3... – Thomas Ayoub Oct 4 '16 at 11:48
  • 2
    @ThomasAyoub: Must have meant linear pixel count or resolution. sqrt(36/20) = 1.3 -> so the D800 has 1.3x the resolution of the 1D X. – Dietrich Epp Oct 4 '16 at 18:57
  • Also there is the Sony A7 series where the S variants only have 12 MP for (supposedly) superior low light and video performance – SztupY Oct 5 '16 at 8:20
  • 1
    @ThomasAyoub The surface of the D800 sensor and the 1D X Mark II sensor are the same size: 36x24mm. The last Canon 1-series camera to have an APS-H sensor was the 1D Mark IV released in 2009. The 1Ds series and the 1D X series all have had FF sensors since 2002. – Michael C Oct 5 '16 at 15:10

Quoting from DPReview's spec sheet for the 1D X Mark II:

Continuous drive 16.0 fps

That means that the 1D X Mark II is pushing 20 MP × 16 = 320 MP through its pipeline every second, which is a bigger number than you'll find on any of Canon's other models; for example, the 5DS R has only 5 FPS. 20 MP is good enough for the intended market of the 1D X series (namely, sports and photojournalists) and they'd rather have more FPS than more MP.

  • Thanks for your answer. A bigger sensor with a lower speed should be possible then, why not build a 40MP sensor which still can handle 16fps at 20MP? – rockZ Oct 3 '16 at 13:27
  • 2
    There's a lot more to it than just frame rate. You can make the pixels in sensors smaller, but you can't make the physics of light waves/particles smaller. – Michael C Oct 3 '16 at 13:30
  • 4
    @iBlink Because such trade-off might not be as easily achieved in semiconductor design as you think it is. – null Oct 3 '16 at 13:32
  • 1
    @iBlink because a bigger sensor does no better than a smaller one if it is only using half of its surface area to collect the same amount of light as the smaller sensor and discarding the light that falls on the other half of the sensor's surface area. If you build a 40MP FF sensor and then only collect the data from half the pixel wells you are collecting no more light than if you had made a smaller 20MP sensor with pixels the same size. – Michael C Oct 3 '16 at 14:13
  • 3
    @iBlink because your only options to do that are to either turn off part of the sensor effectively turning the camera messing with all the expected scale/depth of field/etc behavior, or to read the entire sensor and downsample it. The latter runs into problems both in that unless you went up to an 80MP sensor and could bin 4:1 you're going to have to resize the image lossily degrading the quality of your raws significantly; and that you'd still be reading all the pixels off the sensor meaning you'd still be bottlenecked there unless you doubled your readout rate. – Dan Neely Oct 3 '16 at 15:12

Imagine yourself at the side line of a major sports event. Yeah it's bloody dark despite the stadium lights and you cannot send in your assistant to chase the athletes with a flash. You'd rather have only 20MP but decent low light performance, because more pixels means smaller pixels means less photons per pixel.

  • 3
    Is there a significant difference between a sensor with N megapixels and a sensor with 2N megapixels if you scale the second one down to the same number of pixels? (Assume sensors of the same generation). – Philip Kendall Oct 3 '16 at 13:31
  • In the case of smartphones vs. 20MP FF the area ratio is about 1:35. – Michael C Oct 3 '16 at 14:06
  • @PhilipKendall It depends on the architecture of the sensor. Sometimes it is extremely significant. – Michael C Oct 3 '16 at 14:07
  • 7
    There's a reason the top sports photographers are willing to spend massive amounts on larger lenses to use with FF cameras when they could get the same reach with smaller, cheaper, shorter lenses and smaller sensors. – Michael C Oct 3 '16 at 14:15
  • 1
    The lights at most major sports events are pretty good. They're generally bright and flicker free. They have to be for the television cameras. It's the minor sports events where the lighting stinks. – Michael C Oct 4 '16 at 23:13

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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