realbasic-nug
[Top] [All Lists]

Re: High Performance Code (Was: Bitwise Shift operators)

To: "REALbasic NUG <realbasic-nug at lists dot realsoftware dot com>" <realbasic-nug at lists dot realsoftware dot com>
Subject: Re: High Performance Code (Was: Bitwise Shift operators)
From: Daniel Stenning <d0stenning at msn dot com>
Date: Sat, 29 Sep 2007 19:30:40 +0100
Delivered-to: listarchive at realsoftware dot com
Delivered-to: realbasic-nug at lists dot realsoftware dot com
Thread-index: AcgCxtYrFLrfDm66Edy8rgAbY5KfzA==
Thread-topic: High Performance Code (Was: Bitwise Shift operators)
Ideally yes , but if not , certainly at least as fast as achievable with C
and a good optimised compiler.  Most commonly used C compilers have
intrinsics which can help optimise certain kinds of code for vector based
processing. I think that there is a place for some new syntax that targets
vector based processing. What that syntax would be I don't know, but it
might help the compiler to better optimise and use the modern SIMD and
vector instruction sets we find in all modern CPUS these days.



On 29/9/07 19:06, "Norman Palardy" <npalardy at great-white-software dot com>
wrote:

> 
> On 29-Sep-07, at 11:46 AM, Marco Bambini wrote:
> 
>> What is fast for you?
>> I mean, how many Microseconds it should take the FastSqrt function in
>> order to be "fast enough"?
> 
> Since Dan, the original poster, seems to be using this sort of stuff
> for real time audio ... "fast enough" probably has to be "close to
> the speed of the routine if done in assembler"
> 
> _______________________________________________
> Unsubscribe or switch delivery mode:
> <http://www.realsoftware.com/support/listmanager/>
> 
> Search the archives:
> <http://support.realsoftware.com/listarchives/lists.html>
> 

Cheers,
Dan




_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives:
<http://support.realsoftware.com/listarchives/lists.html>


<Prev in Thread] Current Thread [Next in Thread>