Path: news.mathworks.com!not-for-mail
From: <HIDDEN>
Newsgroups: comp.soft-sys.matlab
Subject: Is FEVAL slow?
Date: Wed, 10 Aug 2011 16:10:27 +0000 (UTC)
Organization: Rincon Research Corp
Lines: 1
Message-ID: <j1uahj$cfl$1@newscl01ah.mathworks.com>
Reply-To: <HIDDEN>
NNTP-Posting-Host: www-04-blr.mathworks.com
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: newscl01ah.mathworks.com 1312992627 12789 172.30.248.35 (10 Aug 2011 16:10:27 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Wed, 10 Aug 2011 16:10:27 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 2235717
Xref: news.mathworks.com comp.soft-sys.matlab:739522

I'm working on something where using FEVAL seems like the most elegant way to do the implementation, but it is running very slowly and I'm wondering if FEVAL is the bottleneck.  I'm doing an optimization problem with different measurement types and it's really slick to be able to add new measurement types by simply including in the measurement structure the name of the function that can generate a predicted measurement as a function of the state.  The alternative is to modify the base code with case statements (or something like that) every time a new measurement type is used.  I'll do that if I have to.  Thanks.