>>>>> "m" == mikecheng <[EMAIL PROTECTED]> writes:
me> in l3psy.c, it uses optimized algorithm only when 44.1kHz,
me> long block data to calculate the spread function(like sprdngf2
me> routine). my version will use all situation(all sampling
me> rate, long/short block). Basically, the initial routine will
me> find "none zero" region of the spread matrix and use this
me> imformation to calculate the spread function.
>> I added this. It should speedup encoding non 44.1kHz input.
m> Excellent! Sounds like a very neat solution. The spreading
m> function hack was done in Oct97, and I didn't use anything
m> other than 44.1khz, so I didn't take the time to optimize for
m> 32 or 48khz.
I add some.
This algorithm can be applied to the short block spread function
calculation, so we can get little speed up even when 44.1kHz.
On my machine(Celeron 103*4.5 = 464MHz), I got a 10% speed up for 44.1kHz,
from this algorithm.
---
Takehiro TOMINAGA // may the source be with you!
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
