Originally Posted by
pwnage101
If that is so, blazing fire, how are the "number of shots" definitively measured if they are subject to a changing variable: memory card speed? Perhaps they are not definitive?
I'm not exactly sure, for I don't work in such an industry, but it's logical to assume that they take the memory size of the buffer and divide it by the average size of a photo. In other words, they do not factor in the rate at which data is written into the card.
so when I read 6 raw frames in a Nikon d80 dpreview, that can change with a faster/slower card?
Maybe, maybe not. Depends whether it is the official specs, in which case it should almost always be higher in reality, >6. If you take the data from their "performance" section, it would most likely represent the number of shots per buffer in acutal situations.
In the case of a d700, the actual number of shots taken with a slow card is 17. A fast card increases this value by 2. Refer to the water tank analogy for the explaination.
http://www.dpreview.com/reviews/nikond700/page13.asp
Code:
Burst of RAW images (12-bit NEF)
TimingTime, secs
(4GB SanDisk Ducati)
Time, secs
(2 GB Lexar 133x)
Frame rate 5.0 fps 5.0 fpsNumber of frames 1917Buffer full rate 2.2 fps1.1fpsNext burst11.0 secs21.0 secs