SD card benchmarking
Posted: Tue Oct 02, 2012 3:06 pm
Hi,
When loading games from SD card, in what size chunks does Swiss read the data? I was thinking that the latency of each read request might vary for different makes of SD card. (By latency I mean the amount of time between when Swiss sends a command to read data, and when data starts being returned from the card.)
Particularly when doing many small reads, a card with higher latency might perform noticeably worse than a card with lower latency, even if the transfer rates are the same. And the latency might not correlate exactly with a card's published class/speed rating; maybe certain Class 4 cards have lower latency than some Class 10 cards for example.
Would it be worth implementing some kind of SD read benchmark in Swiss, to give the user some idea of the latency of their various cards?
When loading games from SD card, in what size chunks does Swiss read the data? I was thinking that the latency of each read request might vary for different makes of SD card. (By latency I mean the amount of time between when Swiss sends a command to read data, and when data starts being returned from the card.)
Particularly when doing many small reads, a card with higher latency might perform noticeably worse than a card with lower latency, even if the transfer rates are the same. And the latency might not correlate exactly with a card's published class/speed rating; maybe certain Class 4 cards have lower latency than some Class 10 cards for example.
Would it be worth implementing some kind of SD read benchmark in Swiss, to give the user some idea of the latency of their various cards?