Re: [NTLK] Mad Max problems

From: Eckhart Koeppen (eck1000_at_gmx.net)
Date: Tue Mar 12 2002 - 18:29:18 EST


Thomas Hart wrote:
> While all the info about bit rates is interesting, wasn't the original
> question about how to get around the choppy play back?

The bit rate defines how much raw samples are decoded from an MP3 frame.
The more samples are decoded the better, resulting in a low memory and
CPU load.

> I've had the same
> problem and would really like to know how to get this going properly.
> So, what are the minimum requirements of heap etc.?

The heap requirements are acutally quite low. Not having the NTK here, I
think that heap memory consumption is about 15-40k. Two things however:
First, the less free heap you have, the more frequent garbage collection
cycles are, resulting in choppy playback. Heap fragmentation can also be
an issue. Second, MAD Max allocates three VBO playback buffers along
with some the MAD Newton data structure on the C++ heap and uses about
200k for that. You can check the different heaps with HeapShow (part of
the NTK).

Eckhart

-- 
                   -- eck_at_40hz.org -- www.40hz.org --

-- Read the List FAQ/Etiquette: http://www.newtontalk.net/faq.html Read the Newton FAQ: http://www.guns-media.com/mirrors/newton/faq/ This is the NewtonTalk mailing list - http://www.newtontalk.net



This archive was generated by hypermail 2.1.2 : Tue Apr 02 2002 - 14:02:40 EST