Several times I have expressed reservations about the review-preview-rumor culture and consumerism in photography (most recently in a fairy tale), but I do of course not completely ignore the technological side of photography. My main criticism is that most reviews are pseudo-objective, theoretical analyses of specifications or performances that have little practical relevance. Buying and using a camera is not objective, but a very subjective action and decision.
Although it has been stated over and over it must be repeated: the best camera is the one that you have with you most often and that you use most regularly. Therefore, it may be relevant how much (theoretical) image quality you obtain with each gram of camera that you carry around. Furthermore, I think it is important to question how much image quality you obtain for the money you spend on a camera. I have been wondering if the existing camera ratings, in particular the DxOmark overall scores, can be used to answer these questions. Here, I show you two graphs that I think reveal interesting aspects about camera performances, weight and prices.
1. IQPW & IQPM
I have arbitrarily selected 31 current cameras that were tested by DxOmark and for which I found current prices on www.toppreise.ch (for discontinued cameras I estimated the price of a used camera). Then, I calculated the IQPW (DxOmark overall score divided by the camera weight) and the IQPM (DxOmark overall score divided by the camera price). The IQPW tells you how much DxOmark score you receive for each gram of camera that you are carrying around and IQPM indicates how much DxOmark score each swiss franc (or whatever currency) buys you. The results of this analysis are shown below (the cameras have been sorted based on IQPW in descending order).
Image quality per camera weight (IQPW) and per money (IQPM) for a few mirrorless, DSLR and point-and shoot cameras. Image quality is defined by the DxOmark overall score and the current lowest price (in Swiss francs) was looked up at www.toppreise.ch.
Three noteworthy observations:
2. DxOmark score vs. price and weight
Next, I have selected all cameras with DxO mark scores, wrote down the prices (in US dollars) that are listed by DxOmark (I corrected one price that seemed completely wrong) and looked up the weight of all the cameras. I ended up with 189 cameras. Also in this case, compact cameras had a much better IQPW (the first 35 cameras belong to this category) than classical DSLRs. For the comparison depicted in the scatter plot below I cut the x-axis at 10'000 USD.
Two noteworthy observations:
I could continue to discuss and argue over these numbers at length, but I rather leave this to you. For me, there are two important take-home messages:
In order to finish off with photography rather than with tool talk, here is a photograph that was taken with a tiny point-and-shoot camera; most likely with an abysmal DxOmark score. However, at that moment it was the best camera of all because I had it with me!
Although it has been stated over and over it must be repeated: the best camera is the one that you have with you most often and that you use most regularly. Therefore, it may be relevant how much (theoretical) image quality you obtain with each gram of camera that you carry around. Furthermore, I think it is important to question how much image quality you obtain for the money you spend on a camera. I have been wondering if the existing camera ratings, in particular the DxOmark overall scores, can be used to answer these questions. Here, I show you two graphs that I think reveal interesting aspects about camera performances, weight and prices.
1. IQPW & IQPM
I have arbitrarily selected 31 current cameras that were tested by DxOmark and for which I found current prices on www.toppreise.ch (for discontinued cameras I estimated the price of a used camera). Then, I calculated the IQPW (DxOmark overall score divided by the camera weight) and the IQPM (DxOmark overall score divided by the camera price). The IQPW tells you how much DxOmark score you receive for each gram of camera that you are carrying around and IQPM indicates how much DxOmark score each swiss franc (or whatever currency) buys you. The results of this analysis are shown below (the cameras have been sorted based on IQPW in descending order).
Three noteworthy observations:
- With the Sony DSC-RX100 you obtain 4.6-times more DxOmark score for each gram than with the Canon EOS 1Dx (and the Sony already has a lens, while the Canon is only the body).
- You obtain twice as much DxOmark score for each Swiss franc spent on the Nikon D3100 as compared to the Canon EOS 1Dx.
- All compact cameras (compact system and point-and-shoot cameras, area shaded in blue) have a better IQPW than conventional DSLRs (area shaded in red).
2. DxOmark score vs. price and weight
Next, I have selected all cameras with DxO mark scores, wrote down the prices (in US dollars) that are listed by DxOmark (I corrected one price that seemed completely wrong) and looked up the weight of all the cameras. I ended up with 189 cameras. Also in this case, compact cameras had a much better IQPW (the first 35 cameras belong to this category) than classical DSLRs. For the comparison depicted in the scatter plot below I cut the x-axis at 10'000 USD.
Camera weight (in gram, left y-axis, red triangles) and DxOmark overall score (right y-axis, blue dots) plotted agains the camera price (in US dollars). There are much more expensive cameras in the database, but the x-axis was cut at 10'000 USD.
Two noteworthy observations:
- The DxOmark score reaches a plateau as soon as you spend about 1000 USD - beyond that point more money only marginally increases the DxOmark score (blue dots). For example, you can buy a camera with a DxOmark score of 81 or 82 for less than 700 USD (Nikon 3200) or almost 7000 USD (Canon EOS 1Dx)!
- Although there are outliers (for example the expensive but tiny Sony DSC-RX1, Leica M9 and M8), camera weight correlates better with price than the DxOmark score. This means that we do not pay for the quality of the sensor in a particular camera, but rather for the amount of material that was used to build it! I think this is an important observation: In many instances, you pay for more solid build quality, which results in a heavier camera, NOT for a higher quality image sensor! This is very different from many other industries, where the lightest items are often the most expensive ones.
I could continue to discuss and argue over these numbers at length, but I rather leave this to you. For me, there are two important take-home messages:
- Do not spend more than 1000 USD on a camera (unless you need incredibly high resolution or a camera that is built for eternity).
- Try to ignore DxOmark ratings and similar metrics because they are not relevant for real-world photography and fail to reflect important properties of a camera (e.g., weight, price or sensor size; after all, a medium format digital back is not comparable to a point-and-shoot camera).
In order to finish off with photography rather than with tool talk, here is a photograph that was taken with a tiny point-and-shoot camera; most likely with an abysmal DxOmark score. However, at that moment it was the best camera of all because I had it with me!
