Accuracy Of The Calculations

This page will try to explain a few problems that can arise when comparing digital cameras. If you want to learn about the limitations and accuracy of the calculations that you encounter throughout this website, read on.

Comparing digital cameras can present you with quite a few problems. The underlying technology can vary in such a way that comparisons may not be very accurate or even useful, especially if you're comparing wide range of products of different generations.

So why bother? Can you really assess the overall quality of a digital camera by looking at some numbers? Well, maybe not accurately, but approximately - yes.

What counts in the end is the end result, namely the quality of the image. There are many factors that contribute to this end result and some can in fact be decently predicted by looking at certain numbers.

The numbers this site is focused on are in some way or another connected to sensor and pixel dimensions. In digital photography, size does matter.

All calculations on this site are based on two camera characteristics - sensor size and/or effective megapixels. Sensor diagonal, surface area, sensor resolution, crop factor, pixel pitch, pixel area, and pixel density are all derived from sensor size and/or effective megapixels.

In order to compare all cameras I had to use numbers that are available for all of them. And all cameras come with sensor size or type designation and number of megapixels.

Unfortunately, these numbers aren't completely accurate. Let's see why.

Sensor dimensions are pretty accurate. Sensor sizes are given either in mm or for smaller cameras in 1/N inch format. If given in mm, then we can say the numbers are accurate (although still rounded to 1 decimal).

However, if given in 1/N inch format, then the width and the height are estimated and not 100% accurate. They're close, but not 100% accurate. The dimensions for the same type of sensor (e.g. 1/2.3") can vary a bit from sensor to sensor. So bear in mind that when we're dealing with 1/N inch format, we have approximate dimensions. Blame the manufacturers for not disclosing the real sensor sizes.

For a list of all sensor sizes with corresponding dimensions see Sensor Sizes.

Megapixels are always an approximation given by the manufacturers, sometimes quite generously rounded up. Not all camera manufacturers provide total and effective megapixels for all models. In some cases they specify only total and in some only effective.

I will always use effective megapixels for my calculations, if they are available. After running some comparisons against SEM (scanning electron microscope) measured pixel sizes I have come to the conclusion that sensor sizes specified by the camera manufacturers correspond to the active imaging area of the sensor (effective megapixels). That's why I'll use effective megapixels in my calculations. If total or effective megapixels aren't explicitly specified, I will assume that given megapixels are effective.


As you can see, the bases for calculations are mostly approximations, so you can expect that the results will also be approximations. Being as it may, I still think it's better to have an approximation than no data at all. Just remember that this is not exact science.

Sensor pixels are usually square and aligned in a grid, but not always. All calculations on this site assume square pixels with no gaps between them. Therefore, pixel calculations yield the maximum potential pixel size, not the true effective size of a pixel or the light gathering area of a pixel. In reality pixels have gaps between them, they have different shapes (rectangular, octagonal...), even variable sizes, and different arrangements.

It's impossible to get precise numbers from data available from manufacturers. The only way to get exact numbers would be to disassemble each and every camera and put it under a microscope. Or force manufacturers to disclose the data. That's why it's important to understand that pixel related calculations are merely theoretical.

Sensor size numbers are quite accurate, so you can rely on them. Pixel related numbers have more limitations, so take them as approximations (pixel pitch is usually within 2% accurate).

I try to do the best I can with the data that's available to me.

The bottom line: Camera specifications provided by manufacturers are not going to tell you the whole story. Numbers found on this site will definitely fill in some missing information, but you should probably take them with a grain of salt.

When comparing two cameras also be aware of camera generation. Digital technology is advancing very rapidly so modern cameras with the same numbers generally produce better images than older cameras.