realbasic-nug
[Top] [All Lists]

Re: Dealing with multi-processor or multi-core

To: REALbasic NUG <realbasic-nug at lists dot realsoftware dot com>
Subject: Re: Dealing with multi-processor or multi-core
From: Jean-Luc Arnaud <jean-luc at cdmultimedia dot fr>
Date: Fri, 30 May 2008 18:54:50 +0200
Authentication-results: mx.google.com; spf=pass (google.com: domain of realbasic-nug-bounces at lists dot realsoftware dot com designates 66.116.103.65 as permitted sender) smtp dot mail=realbasic-nug-bounces at lists dot realsoftware dot com
Delivered-to: listarchive at realsoftware dot com
Delivered-to: realbasic-nug at lists dot realsoftware dot com
References: <p06240802c4657bd7c318 at [62 dot 161 dot 36 dot 122]> <41B4C50C-AB85-4975-B2AE-C6A4DD49D787 at inspiringapps dot com>
Thanks for information, Joe.

Could you say a little bit more on helper app, or give us links to articles ?

Perharps, I'm doing so without knowing it. My main app runs as many 
apps as necessary for achieving parallel tasks. Of course, there is 
communication flow between main app and daughters apps.

Should I benefit of multitasking, doing that ? It seems like I don't 
(perharps badly coded !!).

TIA for details.

Jean-Luc

>On May 30, 2008, at 3:40 AM, Jean-Luc Arnaud wrote:
>
>>  My upgrade plan finishes soon and before paying for a new plan, I
>>  would like to get information about RB and multi-processor management.
>
>Use a helper app (perhaps via the Shell class).
>
>>  A lot of messages on this list claimed about the fact that RB does
>>  not use more than one processor (or one core) at a time. it is
>>  incredible and unacceptable !!
>
>Get over it.  Preemptive multitasking is, in general, very bad mojo.  
>If you access ANY functions that are not thread-safe (and this would 
>include pretty much the entire framework, as well as many of the OS 
>functions that you use even for simple things like drawing to a 
>window), then your program will have obscure bugs that will crop up in 
>odd circumstances after your software is released.  The Windows 
>platform layer used to use preemptive threads, and it was a nightmare 
>-- one that ended only when they coded a cooperative multitasking 
>system like the Mac already had.
>
>>  Would I be obliged to switch to C++ or other, in order to get high
>>  performances ?
>
>No, you would have the same limitations in any language; it's just 
>that those other languages would cheerfully let you shoot yourself in 
>the foot.  RB tries to prevent that when it can, and this in one of 
>those situations where it easily can.
>
>To make use of multiple processors safely (in ANY language), you need 
>to divide your work into separate processes.  In RB, the best way to 
>do this is via a helper app that you would access via the Shell.  It's 
>not that hard to do, and it's 100% safe.
>
>>  I hope no, because I like RB and its object-oriented language, its
>>  graphic interface, the ability to call dll or libraries, to compile
>>  for Mac, Windows and Linux (and a lot of amazing other features). But
>>  if RS does not plan to deal with multiproc/multicore, I will have to.
>
>Well, good luck with that.
>
>Best,
>- Joe
>
>--
>Joe Strout
>Inspiring Applications, Inc.
>http://www.InspiringApps.com
>
>
>
>_______________________________________________
>Unsubscribe or switch delivery mode:
><http://www.realsoftware.com/support/listmanager/>
>
>Search the archives:
><http://support.realsoftware.com/listarchives/lists.html>


_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives:
<http://support.realsoftware.com/listarchives/lists.html>


<Prev in Thread] Current Thread [Next in Thread>