Rui Caldas

i'm building a patch in two computers and the cpu usage is very diferent in the two computers.

So i'm asking you to download the patch and tell me the cpu usage and the computer description:

The file is a *.mxf created in a PC Computer. i don't know if it works on mac! i hope so!

Desktop: 
PC Computer
CPU Usage = 19%
CPU: AMD Athlon FX53 @ 2.4 GHz
RAM: 1GB (dual channel 512)
soundcard: MOTU 896

Notebook: Fujitsu Siemens Amilo M1437g
PC Computer
CPU Usage = 50%
CPU: Intel Pentium M @ 1.86 GHz
RAM: 1GB
soundcard: MOTU 896

cpu-usage-need-help

on a desktop computer, you will always get better performances and stability

yes... it's true but in my girlfriend's computer: notebook asus a6tc: 1gbram and cpu amd x2 tl50 @ 1.8 GHz the cpu usage is 20%.

when i strat building the patch i think in [poly~] to save cpu usage. i'm not using poly 'cause the computer has to run all players at the same time. (if some one whant to use all players).

The patch is not completed, and the parts missing will add more stuff to the cpu.

well... my problem in cpu usage it's not the player (download the player.pat file to see)... (and if you want i can send you all the patch... unlocked)

the cpu usage with one player only patch it is < 1%, so 10 players will use +- 10%... not 20.

i think the problem is in the jit.cellbox object, because it is allways sending info in the left outlet.

when i strat building the patch i think in [poly~] to save cpu usage.  i'm not using poly 'cause the computer has to run all players at the same time. (if some one whant to use all players).

Hi again!all the patch here... unlocked!


> Hi again!all the patch here... unlocked!

I can't tell much about the CPU usage, but looking at your patch, it 

seems all your message passing could be done much much easier.

It seems you don't know how to pass arguments to subpatchers, that way 

you could send commands to the subpatchers directly...

If you have a [r #1] in the player patch and your main patch has an 

object [player p1] you can send messages to that specific player directly...

That way you can get rid of 90% of your patch chords (and in/outlets) 

If you put a list like "p2 f 4 l 1 c play go 0" into the patch below it 

will send the list "f 4 l 1 c play go 0" directly to a receive element 

in the player with the argument p2 if there is a [r #1]. then you 

unpack/iter that list in the player... (no need for your 

comandbus_receivers subpatcher anymore for example)

I agree, that in your case there is no need for using poly~ btw...

If you need more than 9 inputs in a subpatcher, its time to think about 

Rui Fernando Caldas wrote:
> Hi again!all the patch here... unlocked!

I can't tell much about the CPU usage, but looking at your patch, it 
seems all your message passing could be done much much easier.
It seems you don't know how to pass arguments to subpatchers, that way 
you could send commands to the subpatchers directly...
If you have a [r #1] in the player patch and your main patch has an 
object [player p1] you can send messages to that specific player directly...
That way you can get rid of 90% of your patch chords (and in/outlets) 
and probably 60% of your subpatchers.

If you put a list like "p2 f 4 l 1 c play go 0" into the patch below it 
will send the list "f 4 l 1 c play go 0" directly to a receive element 
in the player with the argument p2 if there is a [r #1]. then you 
unpack/iter that list in the player... (no need for your 
comandbus_receivers subpatcher anymore for example)

If you need more than 9 inputs in a subpatcher, its time to think about 
better message passing... ;-)

-- 
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com


CPU Usage. NEED HELP