So I’ve been complaining for some time about the speed of Adobe’s Lightroom photo processing software.
I finally got around to doing some comparisons of import times, using files from different cameras. My initial thought was that the auto-correction used by micro 4/3 lenses was slowing things down on my recently acquired E-M5, along with larger files. To test that theory, I took 100 RAW files from a number of different cameras and lenses, and measured how long it took to import them and generate 1:1 previews.
| Camera | Size | Format | Lens Correction | Running time (s) |
|---|---|---|---|---|
| Nikon D70 | 6.1 MP | NEF | No | 172 |
| Nikon D700 | 12.1 MP | NEF | No | 318 |
| Olympus E-410 | 10.1 MP | ORF | No | 392 |
| Olympus E-620 | 12.3 MP | ORF | No | 446 |
| Olympus E-PM1 | 12.3 MP | ORF | Yes | 493 |
| Olympus E-M5 | 16.2 MP | ORF | No | 580 |
| Olympus E-M5 | 16.2 MP | ORF | Yes | 600 |
As is clear in the table, it turns out that lens correction actually doesn’t have a huge impact – enabling it (or using a lens that enables it automatically) causes less than a 5% performance hitch. On the flip side, increasing the number of pixels seems to linearly increase runtime.
What is also obvious is that the Olympus ORF’s file format takes a good bit more time and energy to process than Nikon’s NEF files. Processing 10MP ORFs takes roughly 25% more time than processing 12MP NEFs! I suppose I’ll have to add that to my already lengthy list of ‘things Olympus could afford to borrow from competitors’.