<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type"/>
</head><body style="">
<div>
Hi Fons
</div>
<div>
<br/>Fons Adriaensen <fons@linuxaudio.org> hat am 1. März 2014 um 13:27 geschrieben:
<br/>> Are you supposed to optimize by
<br/>>
<br/>> 1. understanding the algorithm and the context in which it is used,
<br/>> and eventually modifying it within given limits on performance,
<br/>> precision, etc., or
<br/>>
<br/>> 2. only by changing the way it is coded (which seems to be suggested
<br/>> by your message) ?
<br/>>
</div>
<div>
</div>
<div>
<div>
I'm not sure I get the difference here. As I understand, optimization includes (1) choice of suitable algorithm for the problem, (2) smart implementation, i.e. writing code such that the compiler can apply smart optimization (e.g. aliasing) and (3) optimizing for a particular microarchitecure (leveraging instruction level paralellism for a fixed CPU model).
</div>
<div>
If choosing a error bound + approximation algorithm yields acceptable result, I think that is fine. Is that your point?
</div>
</div>
<div>
<br/>> I've got something you work on but it would have to be treated as
<br/>> confidential.
</div>
<div>
</div>
<div>
We will have to submit code to a university-internal repository and it will run through some software plagiarism system. Would that be a problem?
</div>
<div>
</div>
<div>
Thanks,
</div>
<div>
Jeremia
</div>
</body></html>