Amen Jim! Thanks for bringing some sense to this frenzy :) new and delete (malloc and free) have to search for memory to allocate.
If you look at how some simple memory heaps are coded, its essentially a large array (the heap) with utility functions that scan it linearly looking for places to allocate/de-allocate. In theory (and in my hopes) I would state that the OS's memory management was optimized and thus significantly faster than code you write. However, there are still benefits to array-based implementations..... For instance, if you implement your particle array as a queue, insertion or deletion could be done in constant time. This would place a restriction on the particle system - if particles A, B, C are created, C must die out before B, which must die out before A....however, this is usually the way particles are used, so its not much of a restriction. Even an array of 2000 elements might suffice - all without the CPU overhead of doing list operations. david -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Jim Hunter Sent: Thursday, June 27, 2002 10:28 PM To: [EMAIL PROTECTED] Subject: Re: [hlcoders] Particle System > Sure, having the ability to create a *variable* number of particles is a > plus by using a linked list over an array. Why have 20,000 dormant > classes/structures taking up memory when you don't have to? How about > allocating new particles? Are you supposed to scan through the entire array > each time you wish to create a new particle to find that indexes 919 and > 9294-12300 are free? By using a linked list, you can simply store a pointer > to the last particle and then add on to it as necessary. Forgive my ignorance, but aren't you just shifting that burden (of finding empty space to store a new particle) from your own code to the heap manager? I have no idea which would be more efficient, but either way, *something* has got to look for storage space and determine that it's unused, and reclaim space that is no longer used. I would bet the performance would be close to equal; anyone know if a formal analysis of such a problem has been done? Just curious. Jim _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders
