Here is a list of stupid nonlinear scales
- f-numbers. Inverse square root of illuminance (for exposure calculations). I'd be okay with f-number squared if we measured shutter speeds in milliseconds. Then a shot that required 4 ms at f2/4 (1/250 at f/2) would require 8 ms at f2/8 (1/125 at f/2.8).
- Megapixels. Square of resolution. For photography, I think we should use equivalent 35mm dpi:
| dpi at 24x36mm | Resolution | Megapixels |
| 1000 | 1400 x 950 | 1.3 |
| 2000 | 2800 x 1900 | 5.4 |
| 3000 | 4200 x 2800 | 12 |
| 4000 | 5700 x 3800 | 21 |
Megapixels are useful for signal processing and file size considerations, though.
- Miles per gallon. Reciprocal of fuel consumption. Gets even more exaggerated when you look at differences between mpg ratings. We should use gallons/100 miles.
| Vehicle | mpg | gal/100mi |
| Old SUV | 13 | 7.7 |
| New SUV | 21 | 4.8 |
| TDI Jetta | 36 | 2.8 |
| Prius | 50 | 2.0 |
| Bicycle | ∞ | 0.0 |
These numbers are fabricated but not ludicrous. My point is that the mpg system makes high-efficiency vehicles look way better than they actually are.
- Going from an old to new SUV is a 60% increase in "efficiency", whereas going from a new SUV to a Prius is a whopping 140%. And yet they have the same impact on fuel consumption, which is what we actually care about.
- Same goes for old vs. new SUV compared to Jetta vs. bicycle (though more recent data suggests that we underestimated the TDI fuel economy)
Here's just a few to start - I'm sure there's more out there, and I'll update this post as I come across them.
No comments:
Post a Comment