Re: processing time & stack overflow
Re: processing time & stack overflow
- Subject: Re: processing time & stack overflow
- From: Timothy Bates <email@hidden>
- Date: Thu, 30 Aug 2001 11:55:09 +1000
On 8/30/01 1:44 AM, "Victor Yee" <email@hidden> wrote:
>
Did some time tests and this script seems to be a pretty decent mix of speed
>
and scalability. Here's one set of results OMM:
>
00:01:45 for about 1,000 words
>
00:04:00 for about 2,000 words
>
00:12:13 for about 5,000 words
>
It seems that the processing time scales in a predictable manner. Probably
>
about 25 to 30 minutes for 10,000 words.
I think this thread highlights a really important issue for AppleScript.
Compared to almost any other language it is HOPELESSLY slow. I often have
solve problems involving large text files or arrays, and, while I will often
play with a solution in AppleScript, I always end up implementing the
solution in LabVIEW. LabVIEW is usually 2-3 orders of magnitude faster. not
a bit faster, or even twice the speed but 100th to 1000th of the execution
time.
This pathetic performance from AppleScript is not all because of dynamic
binding or the slow scripting extension call mechanism - much of the
slowness is just in chugging through repeat loops. On an 800MHZ DP machine,
most scripts like this text processing example should execute in
milliseconds, not minutes.
I think that if AppleScript was integrated with Apache and had the speed
even of PERL, then it would make major inroads into web-serving and would
drive the sales of many many high-end boxes from people wanting an
easy-to-understand scripting platform. But as long as it is crippled by a
hopeless compiler, that will never happen.
tim