Stripping down PC for audio use

[RobbieL]

I have a custom-built pc running windows 7 x86 with SONAR 8.5.3 and I've had a lot of problems with sampling issues (i.e. pops & clicks in recorded tracks). I've done the usual "increase of the buffer size" but the problem still seems to linger. I've flipped between ASIO and WDM with no change. My questions to you guys are this:

Which is a better driver model, WDM/KS or ASIO?

Is is a good idea to completly strip down (i.e. turn off unnecessary programs, resources, services, network, etc.) the computer since it's only use is for home recording and how?

What are the ideal settings that I should use in SONAR to prevent this sort of problem?

PC and audio Interface Specs (machine built in July of 2007):
Intel Core 2 Duo E6600 2.4 GHZ
4GB DDR2 RAM
Chipset: NVIDIA nForce 680i SLI
GPU: (2) NVIDIA GeForce 8800 GTS KO ACS3
Audio Interface: MOTU HD192

If there's anything I'm leaving out, let me know.

 

[Shadow_7]

ASIO is the better IMO. But it's the only option for my hardware.

It might be better to just have two bootable disks. Or a multi-boot system. With your audio version being a fresh and minimal install of whatever OS. Although with windows it seems that I spend 5x's longer than the installation time going back and turning OFF everything that is turned ON by default. Not that I've done that in a while. Netbeui (sp?) should be turned off, as that's one network hog. Networking should be turned off / disconnected as that's something that comes and goes in surges and can/will affect the audio chain. In theory it shouldn't, and with a special configuration it doesn't on my system (linux + real time priority). But just the intermittent system resources thing can cause burps in the system if used as the active recording device.

If you're having resource issues, double your RAM if you can. 4GB or less on 32 bit systems or 32 bit OSes. Since you listed x86, and not x86-64, that's probably you. Although spec wise that should be a 64 bit machine / CPU. Albeit dated by todays specs. You have 4GB of RAM which is good. And should be (with the right software) capable of a most audio tasks. Within limits, 100+ tracks all with realtime echo effects, probably not going to happen.

 

[TimOBrien]

do you have separate dedicated drives for your:

1. OS and apps/vsts
2. audio projects/data
3. sample libraries

(you should)

 

[phriq]

ASIO is the better IMO.

Agreed.

If you're having resource issues, double your RAM if you can.

If he is only on a 32bit Windows 7 OS, His system will not be able to recognize or utilize more than 3 or 4 gigs of RAm. So doubling his RAM will not do anything. He would have to just to a 64bit OS to utilize more RAM.

Also, you may want to look at this:

http://www.sweetwater.com/sweetcare/downloads/Windows_7_Optimization_Guide_revB.pdf

This is Sweetwaters optimization guide for Windows 7 recording systems.