"Rational" Shopping and Data Visualization

Although today's entry is not about educational technology narrowly, it is about choosing tools and the limits on our ability to do that. It's also an exercise in data visualization--and its limits as well.

When I shop for cameras, I try to be rational about it. But rationality has its limits.

This image shows my spreadsheet comparing camera image quality as rated by DXOMark.com, along with sensor size--compared to price. (The spreadsheet is too wide to capture nicely on my laptop in a single image.)

Each row's cells are color-coded from the minimum in that row (red) to the maximum (green).

Price is colored in reverse in each of the three tables: so the lowest price is green, the highest red. This reversal is underlined by a box around these row. (Where the camera can no longer be bought new, I left out the price.)

This gives a visual representation of the relation of price to image quality/sensor size.

To quantify further, I represented each number as a percentage of the maximum, and I also show rank from high (1) to low (18): these are the two lower tables.

Columns in bold represent cameras I own--God help me.

Not every similar camera is here: just those that interest me most. (I also made this thinking of a friend who needs a camera, and thus the choices also tilt towards those she might like.) Recent models, like the Canon GX7 and GX1 Mark II, aren't represented.

At only $300, the Olympus XZ2 represents an excellent bargain: very good image quality, a bigger sensor than most quality point-and-shoots, and a low price.

Similarly, the Olympus E-PM2 ("Pen Mini") is very high in image quality and costs under $500.

The Finepix X100 is notable for getting a high-quality image, due in part to its very large sensor. But then again it commands a premium price.

But the chart can't tell you everything.

Even though the Lumix FZ200 has a small sensor and relatively low image quality, it has a long zoom that stays at f/2.8 throughout its range, and it takes nice pictures. So the chart doesn't tell you everything.

Likewise (but in the other direction), the Canon EOS M is a tremendous bargain, but I will tell you: its autofocus is slow, and its touchscreen and interface clunky.

At this point, the Sony A3000 is also an amazing bargain, though it's actually a DSLR (like the Canon t4i).

Although the Nikon 1 J3 is in the middle of the pack, the camera overheats when shooting video and just plain shuts off. It's thus useless to me, despite taking good, sharp images. (I've blown them up to 18x24 with no noticeable artifacts.)

Finally, the columns are sorted by the average of all the percentages: the relation of each camera to the maximum in the cart. But of course we don't buy things "on average": we buy them for the things we want. And I always find that I'm shopping based on getting the best quality for a good price--hence the existence of this chart.

Sony A3000, here I come?

Not really, because to get a lens that achieves all that quality would add another $800....Eeek.