Subject: Re: [linux-audio-dev] Converting float to int
From: Erik de Castro Lopo (erikd_AT_mega-nerd.com)
Date: Mon Oct 29 2001 - 22:22:57 EET
It really blew my mind when "Daniel V." <dvenkita_AT_ecel.uwa.edu.au> said:
> Hi all. I know this has been answered before but I can't find the post :(
> Anyway, the problem is I have a buffer of audio represented as floats [-1,+1]
> that I want to convert to signed shorts for writing to my card. I recall
> there is an efficient way of doing this, and an inefficient way. Can someone
> remind me with a snippet of code?
I am currently writing a paper on exactly this topic. It should be
up on my web site by the end of the week:
I have benchmarked this and achieved speed improvements on Pentium II, P III
and AMD Athlon processors of 6-10 times.
The benchmarking code will be made available for independant verification. Also
included is a cross platform header file so that similar speed improvements can
be achieved on Win32. If I had access to a Mac system I would like to add
support for that as well.
Erik
-- +-----------------------------------------------------------+ Erik de Castro Lopo nospam_AT_mega-nerd.com (Yes it's valid) +-----------------------------------------------------------+ Beware the Lollipop of Mediocrity. Lick it once, and you suck forever.
This archive was generated by hypermail 2b28 : Mon Oct 29 2001 - 22:19:16 EET