Quit Dissing My Megapixels–I love all 36 million of them!
First of all, I must say that I have resisted the urge to write this post for over a year because I know I am going to get dozens of comments saying nothing more than “Megapixels are unnecessary–I want ISO.” I often think that photographers simply repeat what they have heard other pro photographers teach online, and stick to it instead of investigating things on their own.
Before I get nerdy on you, allow me to provide some background if you’re newer to this debate. On a camera sensor, there are millions of light receptors (called photosites) that collect information about how much light is present, color, and other information. It used to be, in the early days of digital photography, that camera manufacturers could only fit 2 or 3 million pixels on these sensors. Now, a sensor of the same size can contain 30 or more megapixels. The problem with the proliferation of mexapixels on a sensor is that it reduces the space available for each photosite. The smaller the photosites are, the more difficult it is for camera manufacturers to create cameras that can take pictures with high ISO levels and still maintain a low amount of digital noise. So many photographers argue that they don’t want camera manufacturers to keep adding more megapixels on a sensor, and would prefer that the manufacturers instead focus on low light performance.
I COMPLETELY understand this sentiment. I have missed many pictures in low light environments because my camera simply couldn’t take a clean picture without adding more light to the scene with a flash. I understand that if all else is equal, the camera with fewer megapixels will produce images with less noise because each photosite has a larger area of light to gather from. I get it. Really–I do. This mirrors a conversation I had with Scott Bourne a few weeks ago at the Google+ Photographer’s Conference, where he argued that the new Nikon D800 (which has 36.3 megapixels) should not be used by portrait photographers because it has too many megapixels. I very respect Scott, but we disagree on this point.
UPDATE: Scott commented below and mentioned an article (which I hadn’t read yet) where he explained his position in more detail. You can read that here.
Basically, there are two reasons why I believe megapixel count should not be discounted: (1) Megapixels future-proof your images, and (2) Megapixels allow photographers to crop significantly.
Future-Proof Your Images with Higher Megapixel Cameras
My first digital camera had 2.3 megapixels. At the time, I remember other photographers telling me that it was plenty big to show the photos on a computer or to make small prints. Now, only 8 years later, photos from that camera are not even capable of filling the screen of my laptop.
This begs the question: Do you want your entire photo library to look outdated in 8 years?
Obviously, the argument against that point is that digital cameras have improved dramatically since my tiny 2.3 megapixel camera, and now that we commonly shoot cameras between 10 and 18 megapixels, we’ve hit a sweet spot where resolution does not need to improve. Photographers who believe this are quick to point out that even a 10 megapixel camera can print perfect photos at 11×14″.
I believe that this argument is short-sighted. First of all, I’m not talking about print size. Digital cameras have plenty of resolution to make large prints, and frankly, it’s rare that I feel the need to print. Most of my photos are viewed on computers.
Speaking of computers, computer screens are currently undergoing a major transformation. In January 2012, the mean screen resolution of a computer was 1366 x768; however, newer computers are coming out with incredibly improved screen resolution. The reason for the improved screen resolution is that screens are being held closer and closer to the viewer. My 55″ TV looks fantastic when viewed from across the room, even though it only has 1080 pixels on the short edge. But now that iPads, phones, and other mobile devices and laptops are the most common way to view the web, the screens are only a few inches from our eyes (centimeters), meaning that we are able to see sharp details better.
In fact, the new Macbook Pro Retina Display has a resolution of 2880 x 1800 at 220 pixels per inch. While that is a dramatic improvement, it still amounts to only 5 megapixels, which is fewer pixels than any modern DSLR. So why do I think 10 megapixels is too few? Because I want my images to look fantastic 10 years from now.
Increased Megapixels Allows for More Cropping
This argument for more megapixels was taught to me this last year while shooting photography in Yellowstone. I was equipped with a 600mm lens from BorrowLenses.com and a 1.4 teleconverter. That means that I was shooting at 840mm. The lens was like a bazooka!
Even though I was shooting with the highest-end gear and the longest lens I could find, I often found myself just barely out of reach of some of the wildlife in Yellowstone. In one particular instance, there was a beautiful red fox posed against the white snow that was just a bit too far for my lens to reach. My camera had 16 megapixels, but when I cropped in to where I wanted the composition to be, there simply weren’t enough pixels to have the quality and detail that I would like. Opportunity lost.
In fact, I find that this is frequently the case. When I’m out shooting, I sometimes frame the shot to the composition that I would like, but then return home and find that I cropped in too tight. With more megapixels, I could always shoot slightly wider than I imagine, and then crop in to the exact spot in Lightroom without worrying about losing detail in the photo.
A prime example of megapixels over focal length was mentioned by Juan Pons a few weeks ago on his excellent podcast. Many wildlife photographers continued shooting the Canon 7D rather than upgrading to the 5D Mark II because the 7D is a crop sensor camera and therefore adds to the focal length of the lens (if you’re new to this concept, read this article). For wildlife photographers, long focal lengths are essential. However, photographers discovered that the 5D Mark III (a full-frame camera and thus shorter effective focal lengths) is actually is better than the crop sensor 7D for focal length, because its increased megapixels allow photographers to crop in more with the same number of pixels than the 7D can crop in effectively with its crop multiplier due to the sensor size.
BUT WHAT ABOUT LOW LIGHT PERFORMANCE!?!??!
What photographers have discovered in the last few months since the D800 was released is that its low light performance is actually quite superior, despite the high megapixel count on the sensor. In fact, it performs better at high ISOs than its D700 predecessor even though there are nearly three times more pixels on the D800. Although the size-per-pixel argument makes perfect scientific sense, it ignores the reality that camera manufacturers have seemed to defeat the physics with advanced noise reduction in the camera.
Another interesting finding among photographers who are experimenting with high megapixel cameras is that, even if these cameras produce more noise, the fact that they capture such fine detail allows photographers to use aggressive noise reduction in Photoshop or Lightroom without losing a significant amount of sharpness.
You see, software tools such as Photoshop and Lightroom are capable of eliminating noise in photos, but the noise reduction always reduces the overall sharpness of the photo; however, if a high megapixel photo is used for the noise reduction, there is enough fine detail in the picture that the noise reduction does not affect the file as much, so more of it can be applied while maintaining sufficient sharpness.
For example, the D800 produces slightly more noise than the Nikon D3s, but if you apply noise reduction to photos from both cameras, the D800 file can take more noise reduction and still retain more detail. When noise reduction is applied to a file from the D3s, the picture falls apart because there are not enough pixels of information to apply much noise reduction. More on that here.
Therefore, if noise reduction is taken into account, high megapixel counts can produce files with less noise.
One More Thing… (Okay, maybe two)
I currently have a Nikon D800 on order (it has been on order FOREVER, so I’d appreciate it if Nikon would ship me one!). It is a $3,000 camera and wields an powerful 36.3 megapixels sensor. For me, this camera is nearly ideal. HOWEVER, I want to respond to a few drawbacks that many photographers have mentioned about shooting a camera with this many megapixels.
1. Shooting portraits with too many megapixels shows too many imperfections in the skin. HUH!!??! I have to admit that I was shocked when Scott Bourne cited this as a main reason for not liking the D800. Using a camera that is incapable of capturing fine detail is a horrible way to smooth skin. Skin smoothing is extremely easy and fast using Photoshop or Lightroom. You don’t have to take low-detail pictures just to have good looking skin. Further, this ignores the fact that there are parts of a person that you WANT incredible detail when shooting a portrait (like the eyes, the hair, the lips, etc).
2. Shooting high-resolution cameras makes for unwieldy file sizes. This is a very relevant argument for some photographers. In our studio, I use a BLAZING fast computer and we have 12 terabyte network attached storage devices, so file size is not an issue in the least. However, for photographers who do not have this type of equipment, I can see it being frustrating to work with such large files.
So yes, I DEFINITELY agree that if you’re not equipped with computer equipment to work with high megapixel cameras, then a high megapixel camera probably isn’t a great choice.
But then again, keep in mind that storage is much less expensive than a quality long lens. A wildlife or sports photographer who can’t spend $10,000 on a quality 400 or 600mm lens could get quite a bit of “focal length” by using a high megapixel camera and then cropping on the computer. Storage is cheaper than a long pro lens.
Conclusion
I apologize for this post being a bit argumentative and overly technical, but I just can’t stand seeing such a good debate passing by without taking the chance to toss in my two cents
Also, I hope that you’ll at least consider things from my point of view before following the photo lemmings on the web who simply repeat the photo advice that they hear others say, because the truth is that there are some very compelling reasons to embrace new technology with high megapixels.
Further Reading:
- Field of View in Full frame vs. Crop Sensor Cameras [Includes photo comparison]
- More than you wanted to know about noise
- The Case for Megapixels
- The Difference Between Full-Frame and Crop Sensor DSLR Cameras


July 24, 2012 

Holly smokes why so bitter man, Jim is only trying to day that there two sides to every story…in fact I agree with both of you. I shoot with D300(s) bodies and I am not planning a switch. I think that is plenty of MP for what I do. What i am saying is that you are basing your opinion on assumptions and and not facts. Did you ever offer your clients a choice between a 2 different MP count images?
Howie, did you read the entire article? Your comment makes me think that you didn’t…
HA! You weren’t kidding when you predicted a heated debate… Love my D800 but I don’t shoot portraits.
I read with interest the points made on the article and the somewhat heated comments below (Wow, folks sure do get defensive where open mindedness might be more productive).
It seems that for some genres or output requirements, higher megapixels (provided not packed into a camera phone sized sensor) offer some advantages.
So if low light performance is equally good and purchase price isn’t prohibitive – why not get a higher megapixel camera.
Portrait photographers not wanting the extra detail or the desire to soften skin in post process simply need to set the camera to one of the lower mega pixel settings. Normally we are always advised to “shoot at the highest quality setting” but this doesn’t have to be the case if for a particular genre yields better results at a lower setting.
I personally frequently shoot windsurf/kitesurf sports – primarily for web display. I used to shoot on 8mp camera with a medium quality 500mm zoom – now I use 18mp camera and a very high quality 200mm zoom and the crop images to get the same range. I find the cropped images superior in detail and sharpness.
neil@digital-photography-bureau.com
I have a question that seems very timely given this post although it is a bit off topic. I have a 12.2 MP camera and am trying to figure out how large an image I can print with it. In looking at web sites I get answers that are all over the map. I can’t really afford to just keep printing large images to try and figure out that. Any suggestions or “rules of thumb”?
Following photo is a good example of a cropped(!) portrait captured with D800 at ISO6400 where careful noise reduction applied to http://www.fotokritik.com/2853534/euro-2012-gunlukleri-4. I prefer spending a little more time for skin smoothing and noise reduction to keep such detail in the eyes and eyelashes. Maybe at the end we downsize the image for web resolution but for sure 36MP gives you a lot of detail for intensive noise reduction. People argue that instead of cropping an image from a FX camera we should buy a cheaper DX camera. I would have never taken such picture with my D5100. The only thing I dont like about my D800 is the large file sizes. However, the results are definitely worth it. I believe that we should also mention the beautiful gradations (maybe not like 16bit medium format cams but still great). Also amazing dynamic range made 3 breaketing HDR almost unnecessary where you can create with 1 single image already.
A lot depends on the quality your looking for, but you should get excellent results from a file that is set to 300 ppi. So if you divide the number of pixels for the horizontal dimension by 300, that will give you the maximum horizontal size in inches for your final print. Do the same for the numer if pixels in the vertical dimension. Hope this helps.
Excellent point. Why not just set the camera to capture a smaller image? Then you have the megapixels when you want them, but not when you don’t. Good point.
I usually keep the site pretty friendly and uncontroversial, but I got suckered into this debate. That’s okay. Sometimes I enjoy a little nerd debate
What do you think of this article where it talks about megapixels in relation to the sensor size? And about investing in higher-end lenses?
http://www.digitalreview.ca/content/Canon-Rebel-T3i-EOS-600D-Compared-to-Nikon-D5100.shtml
I’m glad you found my point insightful – plenty more where that came from! I responded to your request for volunteers to help with improvephotography.com but as yet received no reply…..
Whoa Jim, taking him to school in Latin? Thats cold. I think the saying “To each his on” would apply nicely here. I think sometimes as photographers we get too caught up in brand loyalty and the old tired repeating of “truths” from our predecessors (Like you mentioned) that skews our objectivity. Which is why we need someone to spark a little controversial debate every now and again to keep us on our toes!
Interesting article, thanks for writing it. And best wishes on finally getting your D800 from Amazon – I picked mine up from a local camera store in April, and it has changed my life. Well, the photography part of it, at least.
I agree with the future-proofing aspect – people now are expecting to be able to zoom in on images with their iThings and see more detail, and that’s only going to continue. It remains to be seen if photographers will allow those massive images to escape into the wild, and even though I won’t release anything in high resolution, it’s still useful to have for my own portfolio device.
I don’t follow the cropping argument. If you’re cropping from the centre of the image then a D7000 will put more pixels on the subject, and we know that Nikon has another generation of cameras just around the corner. If that’s the need, then I’d wait a few more months.
If cropping from different parts of the frame is the goal then that introduces all kinds of other problems. The D800 has poor AF point coverage on the periphery of the frame (compared to a DX camera) and cropping off-centre invites problems with asymmetrical lens performance. Coma, softness, mechanical vignetting and falloff, geometric and volumetric distortion, chromatic aberration: everything gets worse as an off-centre crop gets more pronounced.
Finally, can you link to any of the sites of the many wildlife photographers you know who have switched to the 5D3 over the 7D because of the pixel count? I’d love to read more about their reasoning and experience. After all, there are plenty of ways a wildlife photographer would be better off with a 7D instead of a 5D2 – autofocus and build quality come to mind – that are non-issues with the 5D3. But saying that 21Mpx on a full-frame sensor isn’t enough to overcome the greater pixel density of an 18Mpx APS-C sensor, while a 22.3Mpx sensor is, seems rather extraordinary.
It’s nothing new. In the past most photographers were quite happy with 35mm film. Some preferred medium format. A few shot with large format. Plenty of success stories among them all.
Just chalk it up to sour grapes. We are just jealous. Many of us can’t afford the latest and greatest gizmos, and we have to wait a year or two for the sweet new features to trickle down. In the meantime I guess we’ll just have to concentrate on on other aspects such as the light and story telling.
“I’m always amused by the idea that certain people have about technique, which translate into an immoderate taste for the sharpness of the image. It is a passion for detail, for perfection, or do they hope to get closer to reality with this trompe I’oeil? They are, by the way, as far away from the real issues as other generations of photographers were when they obscured their subject in soft-focus effects.” -HCB
I’m leaning your way on this one Jim. I would much rather have the ability to have more detail when needed and smooth out what needs smoothed than to have a smoother all-over photo. And because I nearly always shoot in low-light situations, I struggle with noise regularly. But that doesn’t mean I would want to give up the option of detail. Now I just have to find a happy medium.
“Future-Proof Your Images with Higher Megapixel Cameras” … interesting thought. I think its one reason I have kept using film and scanning for my landscape photography. The detail i’ve been able to keep getting over these past few years has been great, and i’ve not wanted to give up on that for now. But MFD is getting cheaper and the quality is very good. And the new Nikon D800 is excellent for DSLR, so times are changing.
A small point forgotten-Not all of us have an interest or need to work in low light. High megapixel is a big plus for studio photography, or any circumstance with adequate light. Or when your subject makes the light for you, case in point-Fire, fireworks, neon signage. That having been said, I trust my camera makers R&D on the newer models. They want to improve all the features but must in the end choose.