The Compact Flash Cards We tested the below memory cards, the Lexar Professional 300x, and SanDisks Ultra II, Extreme III, Extreme IV and Extreme Ducati, in a Nikon D200 to see which card performs best in the camera. See our Memory Card Performance Tests article for the reasoning behind the test, and for data for other cameras.
The test was simple: fire continuously for 10 seconds to fill the buffer, and to get an idea of how fast the camera can still shoot once the buffer is full, and how long it takes to clear the buffer.
The results surprised us a little - as expected, the SanDisk Ducati (rated at 45MB/sec) was the fastest, but the Extreme IV was right there with it. The Lexar Professional 300x (also rated at 45MB/sec like the Ducati card) under performed - not only was it noticeably slower than the Extreme IV, it also holds almost 200Mb less images than the same sized SanDisk cards in the 8 Gb size. On to the results....
RAW (NEF) Results The following table highlights the results of these cards with the Nikon D200 shooting RAW (NEF files), based on a 10 second burst at its maximum frames per second rate:
||Frames Captured (10 Secs)
||Full Buffer fps
||Data Transfer Rate
||Time to Clear Buffer
|SanDisk Extreme IV
|Lexar Professional 300x
|SanDisk Extreme III
|SanDisk Ultra II
- The Actual Size column is the available space on the card after it has been formatted.
- The Shots Available column is the number of remaining shots indicated by the Nikon D200 for a newly formatted card. This number is an estimate by the camera, and is usually conservative.
- The Card Speed column is the speed claimed by the manufacturer.
- The Frames Captured column is the total number of pictures taken during the 10 second burst. All the cards allow the camera to shoot at full speed until the buffer is full, then they all slow down significantly.
- The Full Buffer fps it the speed (frames per second) the Nikon D200 drops to once the buffer is full and it can't take another picture until a frame from the buffer is written to the card.
- The Data Transfer Rate is the average rate at which the camera transfers data to the memory card during the test.
- The Time to Clear Buffer column is how long after we stopped shooting that the CF card light remained on.
- Tests were repeated 3 times and averages taken.
- The camera was set on AF-C, release priority, shutter priority set to 1/500th, all internal processing and auto-ISO was turned off, to try to maximize the frame rate.
- NEF files vary in size based on the actual image taken - repeating the same test taking pictures of different subject may yield slightly different results - what is more important is the relative performance of the cards given the same controlled conditions.
As you can see, none of the cards could even get to 1 fps with a full buffer. The Ducati and Extreme IV are very closely matched, both managing to take 26 pictures in the 10 seconds, and maintaining very similar frame rates once the buffer fills. The Lexar comes in a respectable third, noticeably slower than the two UDMA SanDisk cards, but faster than the Extreme III. The Ultra II came in a distant last place, taking nearly a full minute to empty the buffer, and only managing about one frame every three seconds once the buffer is full.
The graph below illustrates the same data. As soon as the Camera takes the first picture, it starts writing to the card. All cards run at the same rate until about the 4 second mark when the buffer fills, and then the Ducati and Extreme IV are just about identical.
Summary/Recommendations Bottom line, if you shoot RAW at a high frame rate, we have to recommend the SanDisk Extreme IV cards. The Ducati's are only marginally faster, but at the time of writing cost significantly more. The Lexar card does perform, but we were expecting more from it.
The Nikon D200 does not support UDMA, so it can't use any of the three UDMA cards in this test to their fullest potential - the biggest limiting factor in performance here is the Nikon D200. If you are constantly filling the buffer with even the fastest of cards, upgrading to something like the Nikon D300 may be a good move.