[crackles caused by Intel SATA raid]Renoise And Intel Cpu

I’ve finally built together my new Intel system, but i’m having problems with Renoise crackling at very low CPU usage, even 5% to 10%. Doesn’t matter if i use ASIO or DirectX and i’ve also chosen the same latency settings, which i had on my old AMD. I’m pretty sure it can be fixed with some BIOS setting. Any hints? That’s my first Intel in ages. :)

That’s pretty much my BIOS, except i have some extra settings in the P35-DS3r version of the board.


have you checked the audio cables from your soundcard to the mixer? Are they connected properly? :slight_smile: Can’t think of much else, not a bios-wizzard either, looks pretty normal to me.

two things i can think of:

  1. make sure your X-FI is using a PCI slot which is exclusively linked to a single IRQ. if the manual does not give any reference to IRQ allocation, just cycle through the remaining slots - won’t take too long either with a board only featuring three PCI slots altogether (which is normal nowadays).
  2. if all fails, even though this is pretty unlikely, you might wanna try to play with the PCI latency timer (see CMOS setup -> PNP/PCI Configuration) in multiples of 16 (up&down).

I don’t even have a mixer. :)

EIST is enabled here, CPU Enhanced Halt and the Virtualization thing too. My version of the board doesn’t support that Dynamic Energy Saver feature, only the EP versions.

The card indeed shares a IRQ with other stuff, worth a try, will report back.

No luck, the card just moved between IRQ 18 and 20. I’ve set the PNP configuration to manual and assigned a free one, but it seems this get’s ignored.

If you have an Intel chipset, you may want to update the Intel Matrix Storage manager to a new/latest version.
Your symptom is not exactly the same but when I used ASIO in my previous machine the polling against the harddrives caused crackles in the audio. After the update it was gone.

Disabled EIST, CPU Enhanced Halt (this seems like CnQ from AMD) and the Virtualization, set PCI latency to 32, 64 and 128 and also updated the Intel matrix Storage Manager directly from the Intel site. Guess i’m stuck… :unsure:

The crackles happen with sample only songs too, even very basic ones not depending on the number of cores enabled. I did test some movies, listened to music and did even a gaming session and the sound was fine there. The problems described on the website linked above seem to include crackles in all of these scenarios it seems. I do also have the latest beta driver for my X-Fi card installed. Will test a bit more tomorrow.

Some people also have trouble with smooth playback due to (onboard) wi-fi. With some wi-fi chipsets the polling causes trouble for audio playback…

Good to know. My Board doesn’t have WiFi on it though, but i have found a solution meanwhile: I’ve mirrored my whole RAID0 Array to a single external eSATA drive and Renoise is running perfectly fine now. Looks like the RAID is producing to much traffic/spikes or whatever, which influences Renoise somehow. Another funny thing i had was that my new DVD burner automatically closed the tray, when i’ve opened it which was really weird. This problem is gone now too, already thought it might’ve been defect. :)

I could still try using the Gigabyte RAID chip instead of the Intel one, but i’ll delete the RAID array now and will mirror my system back to a single drive. Tried it out, because i was curious and i think i can live without it.

The really final solution was to put my eSATA drive and the SATA DVD burner to the Gigabyte controller and my two hard drives to the SATA port 4 and 5 of the board. If i put them to port 0 and 1 i get the crackles. The damn Intel controller behaves weird. I could’ve saved some money buying the normal DS3 version of the board instead of the DS3R RAID version.

So if somebody is interested in buying a Gigabyte board of that type, should be aware, that problems might arise.

Finally i can Renoise again normally. :)

It may be an issue on the gigabyte boards in that case, I’m running a RAID0 using Intel chipset (ASUS P5W DH) and I don’t have those issues.

My old machine which also used an Intel chipset had the issues I described above even when I didn’t run the drives in a raid setup.
Not sure though if that was the implementation or just a driver issue.

Well, i think now, that not the actual RAID0 configuration was causing this, but maybe various issues coming together.

I can run my two Seagate drives on the Intel controller just fine, but as soon as i connect the SATA burner to it, i’m getting the crackles and the burner closes the tray automatically. Same happens with my eSATA drive.

I’m now forced to connect both to the Gigabyte controller, where the tray from the burner works fine. Meanwhile i’ve noticed another annoying thing. As soon as a program tries to access the burner in some way, even WinAMP when starting up, i’m getting heavy mouse lags and the hardware interrupts process takes 25% CPU, which can’t be quite right, because all 4 cores are working like crazy in that moment. It takes a while until WinAMP is finally open and works normal. But in VideoStudio 9 for example, if i try to import an audio track it’s constantly freaking my CPU with the hardware interrupts until i manage to close it somehow in that mess.

I’ll contact the Gigabyte support and hopefully get this sorted in some way. Never had that much problems before, even with my two DFI boards, which are kinda known to be little beasts.

There is a rule in RAID world that says that the slowest device in the chain brings everything down and the same rule applies for the device with the lowest storage sets the maximum possible configurable space
If you add a 200GB disk to a chain that contains 100GB disks, you have 100GB unused on the added device that is not configurable.

Never mix 5400RPM devices in a 7400rpm chain, never mix a 10Krpm device in a 15Kchain and certainly don’t ever mix optical devices like CD/DVD player/writers or tape backup devices on a RAID chain.
I have though never experienced other peripheral devices than the drive chain itself that got affected by an event that breaks the above rule.

Might be true, i simply don’t know. The drives used for the RAID0 were exactly the same at least and currently both drives run in AHCI mode as single drives, so i would expect connecting a burner should actually work. Contacted Gigabyte support meanwhile, hope it won’t take ages for them to respond.

Maybe the burner is the black sheep in some way? I mean it behaves weird at the Intel controller and now at the Gigabyte controller. Checked for a new firmware version, but nothing yet.

I’m finally free from those hardware interrupt bursts. I’ve set the mode for the Gigabyte controller from AHCI to IDE and replaced the Gigabyte JMicron driver with the standard IDE dual channel driver from Windows. Only downside is now, that when i power on my eSATA drive it doesn’t get recognized automatically, so i have to go to the device manager and refresh it manually, same goes for powering it off. That’s really the smaller one of the problems though.