Okay, now for a real exciting post, CF (Compact Flash) card test results!
I'm taking a trip soon, and wanted some extra CF capacity, as newer cards are out with higher speeds and capacities and lower prices.
I wanted to know if there was a significant difference in performance in the cards I already owned.
As the Kingston 133x cards are very inexpensive now, I wanted to know if it made sense to buy more.
I devised a very simple test, which reflected to some degree real world conditions. All tests were with my Nikon D300 camera with a battery pack, with newly charged batteries. (If you have a different camera your results may vary!)
I formatted each card before each test. I then shot the same nearby target (something across the room) until the buffer filled. (ten shots in raw at 8 frames/second, in each test). I then used a stopwatch to record how long it took for the buffer to clear (for the green light to go out), so high-speed shooting could resume. I repeated the test three times and averaged the results.
Here are the results:
Kingston 133x 4gb card: 21 seconds.
SanDisk Extreme III 8 gb card: 7.7 seconds.
Lexar 300x 8gb card: 7.7 seconds.
I then formatted the cards, and simply held the shutter down for thirty seconds, to see how many shots I could shoot. (raw) The results:
Kingston 133x 4 gb card: 32 frames.
SanDisk Extreme III 8gb card: 56 frames.
Lexar 300x 8gb card: 56 frames.
Interestingly, I tried the same test using jpeg-high, and got these results:
Kingston: 55 frames
SanDisk Extreme III: 56 frames
Lexar 300x: 56 frames
The buffer, though, took 28 seconds to clear with the Kingston, and approximately 7 seconds to clear with the Lexar and SanDisk.
I then tested download speeds onto my computer using a Lexar FireWire 800 card reader, timing how fast 56 raw images (733 mb) were downloaded.
Kingston 133x 4 gb: 38 seconds (19.3 mb/sec)
SanDisk Extreme III 8 gb: 33.3 seconds (22 mb/sec)
Lexar 300x 8 gb: 28.5 seconds (26 mb/sec)
These are the only cards I currently use, so these are the only cards I tested.
What do the results mean?
If you shoot sports, and frequently fill your buffer, then you want a fast card. The Extreme III and the Lexar 300x were essentially identical in in-camera performance, in my D300. The Kingston 133x was much, much slower.
If you do not shoot sports, the camera performance of any of these cards is more than adequate and would make no difference whatsoever.
Downloading speeds were not as different as I suspected they might be. The 300x was only 35% faster than the 133x, and only 18% faster than the Extreme III. These results assume a firewire 800 connection. When downloading 8 or 16 gb of data, though, that will mean minutes of extra time waiting with the slower cards.
These tests do not in any way test card reliability!
What do I recommend? If your livelihood depends on getting the shot in sports, then get the fastest card you can afford.
I plan to try the new Lexar 233X , since Ken Rockwell reports that their performance is nearly indistinguishable from the 300x cards, but the cost is much less. Since my livelihood does not depend on these cards, I will wait for the price to come down before buying SanDisk Extreme IV or Lexar 600x cards.
If you shoot landscapes and never fill your buffer, get the best price for whatever card you think is reliable enough for you. (you rarely hear of a Lexar or SanDisk card failing). But buy from a reputable dealer, counterfeit cards exist, and it would be a shame to lose valuable pictures because you bought a counterfeit card from someone on ebay!
I just discovered that Adorama (see link at left) is offering the Sandisk cards with an increasing rebate; the more cards you buy (up to three), the bigger the rebate. So, the Extreme III and IV cards become a much better buy.
For more cf card testing, see Rob Galbraith's site.