Re: [NTLK] Pauls future direction...

From: Eckhart Köppen (eck1000_at_gmx.net)
Date: Tue Oct 29 2002 - 02:10:46 EST


On Tuesday, October 29, 2002, at 08:00 , Paul Guyot wrote:

> The reason why SimpleMail and Newtscape are so slow is that they ask
> NIE to send them raw data as arrays of integers.
> So here we have the NewtonScript heap that suffers from a big array
> plus the fact that each byte (8 bits) is represented by a
> NewtonScript integer (32 bits in memory). Plus all the type check
> when you access it.
> It would be more convenient to have binaries there.
> Plus it would be better if the C++ code handled more of the transfer
> (like sending HTTP requests, checking the result, etc).
>

the protoMP3Streamer in MAD Newton actually uses a binary to store the
incoming data. When doing streaming, I can fix the amount of data to be
read. If I have a truncated frame at the end of the decoding buffer, the
MP3 decoder simply leaves it there and waits for the next frame to
complete it. It is still not ideal but the data comes in fast enough.
However, streaming suffers badly from the small (or nonexisting) network
buffers, so whenever the network stalls, playback stutters.

The problem with MAD Max in general is the scheduling of the sound
frames to the sound channel. There is too much NS object creation and
shuffling going on for my taste. Rather, I'd like to feed the channel
directly from the C++ layer without creating NS objects first.

Eckhart

-- 
Read the List FAQ/Etiquette: http://www.newtontalk.net/faq.html
Read the Newton FAQ: http://www.chuma.org/newton/faq/
This is the NewtonTalk mailing list - http://www.newtontalk.net/



This archive was generated by hypermail 2.1.2 : Thu Oct 31 2002 - 12:04:35 EST