In fact, data rates (and CPU usage) are often LOWER during FF/REW, because it’s not reading the entire stream – it’s just skipping to frames much further beyond.
Instead of reading from the file like this:
A B C D E F G H I J K L …
it reads like this for 2x:
A C E G I K
or for 4X
A E I …
by reading only a few frames at a time and skipping all the intervening material.
Rewind is even more complex.
But, let’s assume it was as simple as just reading twice as fast…
That’d only get you up to maybe 4x FF/RW before the CPU is exhausted with decoding at 4x the rate.
