Sdhc speed classes?



I’m just curious if anyone knows anything about the sdhc card speed classes. The public specification only says minimum write speed of 2, 4 and 6MBps for class 2, 4 and 6. The reason I am wondering, is that I have a sandisk ultra 2 4GB that is rated at class 4 (why not class 6), and a sandisk ultra 2 8GB that is rated at class 2, even though benchmark results from the 2 cards are almost identical? Further, I have a transcend 4GB card that is class 6, and is slightly faster with smaller files, but slightly slower with large files (still in the same ball park overall). I’m just curious why the different ratings on similar cards not to mention why sandisk would rate a faster card class 2?

#2 ?


Yea I already read that and it give the same as the sd association.

SDHC cards have SD Speed Class Ratings defined by the SD Association. The SD Speed Class Ratings specify the following minimum write speeds based on “the best fragmented state where no memory unit is occupied”:[17]

* Class 2: 2 MB/s
* Class 4: 4 MB/s
* Class 6: 6 MB/s

That doesn’t seem to explain why 2 sandisk cards with nearly identical performance, and rated at the same maximum speed by sandisk are rated at different speeds (class 4 and class 2), or why a transcend card that has similar performance is class 6? Now that I think about it, a while back I got a pqi card that was class 6 but the performance was terrible across the board (much worse than the other 3 mentioned). That would make me think that they are not paying much attention to actually following the class ratings (and still doesn’t explain the two different rated sandisk cards). according to the sd association though, they must be licensed and use standardized testing equipment like this.
There must be some parameter that is not taken into account in your average benchmark? You have sigh a nda to find out the full specification though? I’m just curious if anyone had any idea what it might be? It makes purchasing a little frustrating if the ratings don’t have anything to do with real world speed. Fyi I know that benchmarks are not really real world speed either but I have done testing with file transfers as well as timed performance in a dslr and compared others reports on dpreview. I actually have plenty of cards now and know some good ones if I need more (returned several that didn’t perform well but got stuck with a couple of cosairs that were a little slow a while back). Its more just curiosity right now.