[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 5.2 slower the 5.1???
- Subject: Re: 5.2 slower the 5.1???
- From: Phillip David <phillip_david(at)xontech.com>
- Date: Tue, 20 Apr 1999 16:55:21 -0700
- Newsgroups: comp.lang.idl-pvwave
- References: <email@example.com> <MPG.firstname.lastname@example.org> <email@example.com> <MPG.firstname.lastname@example.org> <email@example.com>
- Xref: news.doit.wisc.edu comp.lang.idl-pvwave:14433
It actually depends on whether your directories are on your local hard
drive or on a remote machine (such as a file server), and on whether the
search path is the same between the two versions. This whole problem
showed its ugly head for me in the initial post when I was trying to
create some simple objects. I defined them without an init method, and
tried creating an array of 10000 of them. When my search path included
network drives, the time consumed was HUGE compared with the time
required when I provided the init method or when the search path stayed
on my local drive. The moral of the story I gave at the time was to
make sure that IDL can find an actual file each time it searches for
one. Otherwise, you may spend a LONG time searching. NOTE: IDL wasn't
smart enough (at least then) to figure out that when it couldn't find
the init method for the first object, it wasn't going to be there the
next time I tried to find it either... This appeared to be the real
problem I had.
By the way, I was searching a (local) NT server with a 100Mb/s line, and
was still spending 3-4 seconds waiting for the init method to not be
> David et al.,
> Isn't a delta of 7 seconds an awful lot to pin on directory search time?
> - Bill B.
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own