The Unofficial Access Virus & Virus TI Forum - since 2002

The Unofficial Access Virus & Virus TI Forum - since 2002 (http://www.infekted.org/virus/forum.php)
-   General discussion about Access Virus (http://www.infekted.org/virus/forumdisplay.php?f=105)
-   -   Virus TI cpu usage (http://www.infekted.org/virus/showthread.php?t=25887)

F5D 07.10.2005 01:15 PM

Virus TI cpu usage
 
I want to know how much the TI really needs computer's own cpu to run when using the usb transfer and virus control plugin. If possible, post your results here in percentage (for example 20%), host sequencer, processor speed, soundcard model and asio/core audio latency. Also tell how many virus ti tracks you used. If possible, use as many as you can. Don't use any other vsti effects or synths.

I guess everybody wants to know how it really is...

ten 07.10.2005 06:37 PM

Its minimal on a decent spec machine.

On mine with 2x stereo streams via USB and 3x stereo via analog at 64 samples (2ms) it uses 7% host cpu while playing.

Cubase sx 3.1
Dual dualcore opteron 275s (quad cpu)
RME fireface

Oh this was using all 16 parts spread across the 5 stereo outputs. I also tried just using the USB outputs only with 8 sounds each, still it showed 7% host cpu.....very cool :)

ten

F5D 08.10.2005 01:56 PM

Thanks for the results Ten.

Still seems to need alot more cpu than the virus powercore. Other results?

ten 08.10.2005 02:44 PM

Believe me, the TI SLAYS the virus poco in every asepct.....then comes back again and kicks in while its down. There is no comparison.

The tiny cpu overhead is worth the sound seriously (just hope they fix the buggy vcontrol :)

ten

Midi_Glider 08.10.2005 09:53 PM

Quote:

The tiny cpu overhead is worth the sound seriously
Still, the actual sound quality & complexity (i.e: oscillator types, effects etc) is irelative to CPU usage on the host machine, as you must know... :wink:

I am also surprised in regard to the USB overhead witch seems rather high (especially in regard to a monster machine yours that can makes my own Dual Xeon 3.4 rig look rather slow). though, I must also suspect that like most plug ins, the VC is probably utilizing only a single thread in a multicore mechine.

could it also be the very low latency? how is the CPU measure with a 512KB buffer?


Best,
midi.

ten 08.10.2005 11:09 PM

Yes, as the latency drops the cpu % increases.

At 512 its like 2% and doesnt move.

ten


All times are GMT. The time now is 11:52 PM.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2002-2022, Infekted.org