Maximum speed of Harbour + FWH EXE

Maximum speed of Harbour + FWH EXE

Postby George » Tue Mar 13, 2012 4:52 pm

Hello forum,
Based in your experience what are the procedures to follow in order to get the maximum speed of generated EXE program?
By example I am thinking of:
1. Substitute Harbour code by C code (that is in functions that can be, or should be, optimized by using C Code).
2. Use Local variables instead of Global variables.
3. For database use ADS/ADT instead of DBF (I am not sure here).
4. Using HASH tables instead of Array
5. Using code blocks and eval() functions
6. using assembler code (It's is possible call assembler code from harbour?)

Is the Harbour P-Code a big limitation to get EXE running fast or we can override, this limitation, by calling C functions?

Regards,

George
George
 
Posts: 724
Joined: Tue Oct 18, 2005 6:49 pm

Re: Maximum speed of Harbour + FWH EXE

Postby Adolfo » Tue Mar 13, 2012 11:36 pm

George.

In my personal experience. the fastest exe is that one with clear, optimized code, avoiding loops, with no much branching.

Here you can find a lot of tips regarding optimizing code, just search a bit in the forum and that's all.

About db's, I prefer to use MYSQL, I used ADS some time ago, but it is expensive, and not so fast compared to MYSQL in a Linux server.
By the way, you have a complete 100% compatible Mysql db in MARIADB. (Also FREEEEEEEEEEE)

From Chile
Adolfo
;-) Ji,ji,ji... buena la cosa... "all you need is code"

http://www.xdata.cl - Desarrollo Inteligente
----------
Asus TUF F15, 32GB Ram, 1 TB NVME M.2, 1 TB SSD, GTX 1650
User avatar
Adolfo
 
Posts: 846
Joined: Tue Oct 11, 2005 11:57 am
Location: Chile


Return to FiveWin for Harbour/xHarbour

Who is online

Users browsing this forum: No registered users and 49 guests

cron