Delbert Franz wrote:
> This E-mail is a bit long, but I wanted to report on what I found,
> since nothing has appeared on these topics yet.
Good stuff ! It would also be nice to have some performance data on
NAND vs. uSD.
This isn't entirely trivial. While I think uSD read and write speed
should be relatively constant for a given card (although different
from each other), NAND is more complex because it is used with
compression.
So a large transfer from NAND should take something in the order of
max(R_n/S_c, R_r/S_c, R_w/S_u, R_u/S_c)
where
S_c is the size of the compressed data
S_u is the size of the uncompressed data
R_n is the "raw" throughput of the NAND
R_r is the speed at which the memory can be read, with ongoing
decompression
R_w is the speed at which the memory can be written, with
ongoing decompression
R_u is the rate at which the CPU core can uncompress data (assuming
that this is a function of the compressed size)
Note that, if R_r+R_w > R_m (even if R_r < R_m and R_w < R_m),
with R_m being the memory bus bandwidth, then R_r and R_w affect
each other.
So this would depend on the compressibility of the data, the
compression algorithm, and also any possible background load.
- Werner
_______________________________________________
Qi Hardware Discussion List
Mail to list (members only): [email protected]
Subscribe or Unsubscribe:
http://lists.en.qi-hardware.com/mailman/listinfo/discussion
