1-2 audio clicks per minute despite full optimized Windows 10

An area for people to discuss Scope related problems, issues, etc.

Moderators: valis, garyb

User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

Well, now you know why I said the age of the system was a potential issue. As GaryB indicates, Core2Duo & Core2Quad systems are well supported under Win10, so one would think the stock chipset drivers for X58 are probably ok, however in almost all cases when I check the date(s) on such drivers I find they're dated 2006-2009, which is often before the chipset even was publicly available. That means any potential conflicts discovered from field testing (ie, users) are left out. Intel has some of the best hardware qualification in the business, but even they can't test every combination that users will experience and it's very doubtful they have Scope PCI cards on hand, even in that era.

And again, 3rd party chipsets and that GIGABYTE GBB36X Controller (which I think is a rebranded JMicron RAID chipset if memory serves) are the things that are the most concern with older boards and Win10. There's so much variability when it comes to interactions with components like that, it can be a pain to isolate if any are the issue. I tend to DISABLE stuff I don't use in the BIOS for this reason, reducing my system down to just the core features needed for operation. There's no reason to have parallel port & serial port drivers active for instance if you'll never use them.

Win 8.1 drivers are technically 'compatible' with Win10, you can try to find the Intel chipset drivers (and realtek) for that era, or see if Intel's INF Utility for chipset drivers finds anything. That's less likely as it draws on Intel's current pool of downloads, so 8.1 drivers are your best bet. I briefly looked at Intel's download site for you and the only 8.1 chipset drivers that came up were Intel's IME updates. Best bet there is probably to spend a bit of time searching and find out the exact last set of INF Utility/drivers for Win 8.1 & X58, and then find that version on a site like guru3d.com where archives are maintained that can be trusted: Guru3D.com » Download » Categories » Mainboard Chipset drivers » Mainboard drivers - Intel & Guru3D.com » Download » Categories » Generic Audio Drivers & Media Codecs

Also, it may be worth looking at what CPU Microcode updates have been put out by Windows to mitigate attack vectors like Spectre. These make your system more secure, but absolutely throttle cpu performance as the 'holes' they plug were lookahead buffers and other pipeline tricks to keep modern CPU cores working at full speed. I doubt this is going to fix your immediate issue, but on my dual Xeon from that era they have collectively cost me about 15% of the past CPU performance without any other changes to that machine.

Lastly, I keep archives of things I have downloaded in the past so I have on hand various older packages like:

Intel drivers
SetupIntelChipset-10.1.1.45.exe & IntelChipset-10.1.1.45-public-mup (the same I believe, 5/4/19 probably not for X58)
SetupChipset-10.0.26.exe
IntelChipset-Win8_10.0.24_Public.zip
SetupChipset-9.0.4.1015.exe

Realtek_HD_Audio
64bit_Win7_Win8_Win81_Win10_R282.exe
64bit_Win7_Win8_Win81_Win10_R281.exe
64bit_Win7_Win8_Win81_Win10_R280.exe
64bit_Win7_Win8_Win81_Win10_R279.exe
64bit_Win7_Win8_Win81_Win10_R277.exe
...etc.

If any of those appeal and you can't find newer ones with the methods above, let me know via PM and I'll sort something out.
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by garyb »

why would anyone need the raid driver for a DAW anyway? this isn't 1990...
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

valis wrote: Thu Jan 23, 2020 3:30 am Well, now you know why I said the age of the system was a potential issue. As GaryB indicates, Core2Duo & Core2Quad systems are well supported under Win10, so one would think the stock chipset drivers for X58 are probably ok, however in almost all cases when I check the date(s) on such drivers I find they're dated 2006-2009, which is often before the chipset even was publicly available. That means any potential conflicts discovered from field testing (ie, users) are left out. Intel has some of the best hardware qualification in the business, but even they can't test every combination that users will experience and it's very doubtful they have Scope PCI cards on hand, even in that era.

And again, 3rd party chipsets and that GIGABYTE GBB36X Controller (which I think is a rebranded JMicron RAID chipset if memory serves) are the things that are the most concern with older boards and Win10. There's so much variability when it comes to interactions with components like that, it can be a pain to isolate if any are the issue. I tend to DISABLE stuff I don't use in the BIOS for this reason, reducing my system down to just the core features needed for operation. There's no reason to have parallel port & serial port drivers active for instance if you'll never use them.

Win 8.1 drivers are technically 'compatible' with Win10, you can try to find the Intel chipset drivers (and realtek) for that era, or see if Intel's INF Utility for chipset drivers finds anything. That's less likely as it draws on Intel's current pool of downloads, so 8.1 drivers are your best bet. I briefly looked at Intel's download site for you and the only 8.1 chipset drivers that came up were Intel's IME updates. Best bet there is probably to spend a bit of time searching and find out the exact last set of INF Utility/drivers for Win 8.1 & X58, and then find that version on a site like guru3d.com where archives are maintained that can be trusted: Guru3D.com » Download » Categories » Mainboard Chipset drivers » Mainboard drivers - Intel & Guru3D.com » Download » Categories » Generic Audio Drivers & Media Codecs

Also, it may be worth looking at what CPU Microcode updates have been put out by Windows to mitigate attack vectors like Spectre. These make your system more secure, but absolutely throttle cpu performance as the 'holes' they plug were lookahead buffers and other pipeline tricks to keep modern CPU cores working at full speed. I doubt this is going to fix your immediate issue, but on my dual Xeon from that era they have collectively cost me about 15% of the past CPU performance without any other changes to that machine.

Lastly, I keep archives of things I have downloaded in the past so I have on hand various older packages like:

Intel drivers
SetupIntelChipset-10.1.1.45.exe & IntelChipset-10.1.1.45-public-mup (the same I believe, 5/4/19 probably not for X58)
SetupChipset-10.0.26.exe
IntelChipset-Win8_10.0.24_Public.zip
SetupChipset-9.0.4.1015.exe

Realtek_HD_Audio
64bit_Win7_Win8_Win81_Win10_R282.exe
64bit_Win7_Win8_Win81_Win10_R281.exe
64bit_Win7_Win8_Win81_Win10_R280.exe
64bit_Win7_Win8_Win81_Win10_R279.exe
64bit_Win7_Win8_Win81_Win10_R277.exe
...etc.

If any of those appeal and you can't find newer ones with the methods above, let me know via PM and I'll sort something out.
Thank you for this valuable information, this can be of use for a broad type of PC (t)error users for solving related issues.

What I will do is the following. First, I'll swap the GPU card and measure the results. Then, as I've already tested to disable various USB ports, there are not many options left. That solved my problems once upon a time but on Windows 7 though.

Gary suggested that I shall use 'Freeze Instruments tracks' etc. That would be an option, but I need the Scope resources for post-processing with external gear. So, things must happen 'live'. And for that, of course, the audio signal must be 'pure'.

Nevertheless, I'm in the middle of an important project and for now will be using an external USB sound card, until I get some time to re-activate this issue and this thread.

Thank you all for your support.
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

Just for the record, I love scope in an older secondary machine connected to an RME interface on my primary rig. Zero issues, has literally run for 18 years with only 1 PSU replacement, and nothing stops me from keeping my DAW box on the latest & greatest if something tickles my fancy. Scope is as irreplaceable in that role as it always has been.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

valis wrote: Thu Jan 23, 2020 5:19 pm Just for the record, I love scope in an older secondary machine connected to an RME interface on my primary rig. Zero issues, has literally run for 18 years with only 1 PSU replacement, and nothing stops me from keeping my DAW box on the latest & greatest if something tickles my fancy. Scope is as irreplaceable in that role as it always has been.
Agree! Same situation here, almost 20 years with Scope platform, and I do not want to part with it just because of OS upgrade. :o
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

Well you know of another option now. Win7 would still run Scope fine, my system is on XP still. It doesn't connect via network to anything besides my other computers, never accesses the outside internet.
al_bot
Posts: 36
Joined: Sun Dec 25, 2016 2:14 pm
Location: USA

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by al_bot »

Interesting about the GPU. I'm running into a similar situation with audio clicks using ASIO2 in Scope 7 - Pulsar 2 with Windows 10 and with Cakewalk Pro Audio or with the Bandlab version. My GPU was EVGA Geforce 8400 GS, which was a budget video card / no win 10 drivers. Oddly, I get no clicks if I switch to using the Wave Driver with Cakewalk. Also don't get any audio clicks when using ASIO2 in Scope 7 + Win 10 and Reason 10. Was thinking of switching to MSI Gaming GeForce GT 710. But if you're having issues with that model maybe I'll look at something else as well.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

al_bot wrote: Thu Jan 23, 2020 9:05 pm Interesting about the GPU. I'm running into a similar situation with audio clicks using ASIO2 in Scope 7 - Pulsar 2 with Windows 10 and with Cakewalk Pro Audio or with the Bandlab version. My GPU was EVGA Geforce 8400 GS, which was a budget video card / no win 10 drivers. Oddly, I get no clicks if I switch to using the Wave Driver with Cakewalk. Also don't get any audio clicks when using ASIO2 in Scope 7 + Win 10 and Reason 10. Was thinking of switching to MSI Gaming GeForce GT 710. But if you're having issues with that model maybe I'll look at something else as well.
The audio interference I experiencing is regardless of DAW sequencer. Even with Wave Driver. They appear maybe 1-2 per minute, sometimes it can pass several minutes before they appear again.

To be fair, at this point, I'm not sure if the GPU is to blame, as I haven't tested with another GPU under Windows 10. But specific hardware combined with a fact that my mobo is quite old may give such a result?

When I use an external USB connected audio card, everything works fine. That tells me that GPU may not be malfunctioning, rather some very very deep settings or/and hardware combination. Tricky, but I'll try with another GPU.
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

Again, a USB sound card is not utilizing resources from the system in the same way as your DSP cards. There is MUCH more going on with scope than passing audio streams from the DAW to I/o via drivers (WAV, ASIO, etc). Plus USB != PCI.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

Update.

Under Windows 10 and Luna 3DSP Scope v7, the audio clicks interference proportionally appears depending on the ULLI buffer settings. Highest ULLI none, lowest, 1-2 per minute. Regardless if Wave or ASIO source. No change if ASIO and CPU are on load. For latest tests, ASIO meter, and CPU just dancing around 5-10%.

For now, I can use it with higher buffer settings as I need to use the Scope system for post-processing.

So my conclusion is that Scope is interfering with something only at lower latencies, despite ASIO and CPU meter dancing only between 5-10%.

Maybe it is just 'fighting' with GPU then?
Last edited by CreamWare4Ever on Fri Jan 24, 2020 3:14 pm, edited 1 time in total.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

valis wrote: Fri Jan 24, 2020 2:43 am Again, a USB sound card is not utilizing resources from the system in the same way as your DSP cards. There is MUCH more going on with scope than passing audio streams from the DAW to I/o via drivers (WAV, ASIO, etc). Plus USB != PCI.
Yes, I know that. I was mentioning that just to illustrate that my setup CAN deliver real-time audio. Just as the elimination process phase as a part of a full diagnostic.
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

That's good. Also please don't mistake that I'm trying to talk down to you. When we're having technical discussions I'm the type to err on the side of being more complete as I don't know what you know, and others are generally reading along. GaryB tells it like it is.. :D

As for capable of real-time audio, Scope is more than the ASIO channels.

There's considerable traffic going across the PCI bus even when there's no ASIO driver present, and it will differ based on what device(s) are loaded. Reverbs, delays and such are the most obvious culprits for load the PCI bus but they're also the easiest for us audio guys to understand why when we hear 'they use delay lines' it makes it obvious they're sending numerous channels of audio across the PCI bus to achieve thick/rich reverbs and saucy dubby delays. That's all SYNC data in your DSP meter, meaning audiorate/realtime. Not all SYNC data goes across the PCI bus, in fact most doesn't. ASYNC data is midi and control signal information (fraction of audiorate), and I believe (GaryB can correct me here) UI code hooks. All of that communicates with the host PC as well...

My 'backup' backup soundcard in my gig bag is a Komplete Audio 6, it doesn't even have a control panel *at all*. Nothing except a class compliant USB port that exposes a few i/o's and a big knob on top. There's almost nothing to fail there as long as the unit hasn't taken a knock. The Babyface Pro that sits alongside it has never really failed me either, but it has a lot more functionality and software control that can go wrong, so I keep the former in my bag as a last ditch way to get audio running if need be. It's also a handy thing if someone needs a loaner (in the past I've been at festivals all weekend running stages, doing visuals, performing and many other tasks). That comparison is really apples:bananas where scope is a bushel of throwing knives, but still illustrates the point I hope.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

valis wrote: Fri Jan 24, 2020 9:25 am That's good. Also please don't mistake that I'm trying to talk down to you. When we're having technical discussions I'm the type to err on the side of being more complete as I don't know what you know, and others are generally reading along. GaryB tells it like it is.. :D

As for capable of real-time audio, Scope is more than the ASIO channels.

There's considerable traffic going across the PCI bus even when there's no ASIO driver present, and it will differ based on what device(s) are loaded. Reverbs, delays and such are the most obvious culprits for load the PCI bus but they're also the easiest for us audio guys to understand why when we hear 'they use delay lines' it makes it obvious they're sending numerous channels of audio across the PCI bus to achieve thick/rich reverbs and saucy dubby delays. That's all SYNC data in your DSP meter, meaning audiorate/realtime. Not all SYNC data goes across the PCI bus, in fact most doesn't. ASYNC data is midi and control signal information (fraction of audiorate), and I believe (GaryB can correct me here) UI code hooks. All of that communicates with the host PC as well...

My 'backup' backup soundcard in my gig bag is a Komplete Audio 6, it doesn't even have a control panel *at all*. Nothing except a class compliant USB port that exposes a few i/o's and a big knob on top. There's almost nothing to fail there as long as the unit hasn't taken a knock. The Babyface Pro that sits alongside it has never really failed me either, but it has a lot more functionality and software control that can go wrong, so I keep the former in my bag as a last ditch way to get audio running if need be. It's also a handy thing if someone needs a loaner (in the past I've been at festivals all weekend running stages, doing visuals, performing and many other tasks). That comparison is really apples:bananas where scope is a bushel of throwing knives, but still illustrates the point I hope.
Well, doing live things are always mission-critical tasks, going into such projects requires reliable backup. As for now, I cant remember that Scope let me down once in front of clients in the studio. Of course, theoretically, that could have happened, but that's 2 different things: To 'fail' in front of one artist vs. in a front of a pulsating crowd on a festival

Nevertheless, soon we will know more, after swapping GPU.
Last edited by CreamWare4Ever on Fri Jan 24, 2020 7:14 pm, edited 1 time in total.
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

The difference imho is that at a live event things get dropped, knocked around, drives crash and computers may be squashed or drowned in everything from alchohol to rain (I've seen all that and far worse). The only time Scope ever failed me, was when a PSU was failing.

As for the GPU, wishing you the best of luck. You may want to go to guru3d.com and search the forums for various DPC latency discussions. Things have gotten better in the last few years, but results used to vary wildly from driver version to driver version.
CreamWare4Ever
Posts: 204
Joined: Sun Apr 22, 2007 5:07 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by CreamWare4Ever »

valis wrote: Fri Jan 24, 2020 7:03 pm As for the GPU, wishing you the best of luck. You may want to go to guru3d.com and search the forums for various DPC latency discussions. Things have gotten better in the last few years, but results used to vary wildly from driver version to driver version.
Well, that is the only thing left to do, as we have pretty much-tried everything else. I will keep you posted.
guppy
Posts: 388
Joined: Thu Aug 16, 2012 6:42 pm

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by guppy »

[edit]redundant post[edit by Valis]
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

no need to overshare guppy, most users read the few new active threads when visiting (and I can see this in my analytics view). Thanks!
ron3113
Posts: 6
Joined: Sat Mar 28, 2020 9:25 am

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by ron3113 »

CreamWare4Ever wrote: Fri Jan 24, 2020 7:16 pm
valis wrote: Fri Jan 24, 2020 7:03 pm As for the GPU, wishing you the best of luck. You may want to go to guru3d.com and search the forums for various DPC latency discussions. Things have gotten better in the last few years, but results used to vary wildly from driver version to driver version.
Well, that is the only thing left to do, as we have pretty much-tried everything else. I will keep you posted.
As I seem to be having the same problem with two Pulsar cards running on an older Xeon quad-core (the problem, in short, being 2-3 audible clicks per minute despite implementing a vast variety of audio optimization tips and tricks), I'm on the edge of my seat. Did you finally get any progress on your issue, Creamware4Ever?
User avatar
valis
Posts: 7306
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by valis »

Use the troubleshooting tips in this thread and refer to the post at the top of this forum please.
ron3113
Posts: 6
Joined: Sat Mar 28, 2020 9:25 am

Re: 1-2 audio clicks per minute despite full optimized Windows 10

Post by ron3113 »

Dear Valis,

Thanks for your extensive guidance in this and other threads. It took me quite a long loop of learning-by-doing to get back to some of your original advice which - hurray! - turned out to work for me.

In case anyone else follows this thread to the finish, let me just list a few lessons I learned on my way from 1-2 clicks per minute to less than one click per 5 minutes.

1. Once the 'ordinary' audio optimization tips and tricks have been implemented (kill pagefile, eliminate useless services, stop windows defender, unpark cpus, check disk and memory, etc.) then what is left in terms of clicks (e.g. 1-2 per minute even at low CPU loads) may very well be IRQ conflicts.

2. From Windows 7 x64 and onwards, there is no longer any way to control IRQ assignment (at least not from within Windows - you BIOS may have options that my Dell bios does not). You CANNOT reset IRQ settings in the 'properties' of each component without disabling automatic IRQ handling first, and - alas - you cannot disable automatic handling of IRQs in the properties of the PCI Bus.

3. However, IRQ conflicts are MORE COMMON if all drivers are not updated, so this is your only hope at this point.

4. The good news is that if you think you have updated all drivers, you may not have. Clicking 'update driver' in Device Manager DOES NOT dependably retrieve the latest drivers for system components. You need to retrieve the latest driver versions from the manufacturers.

5. There are free programs that do this for you, so that you don't have to rummage around Intel's website and many others. I used Driver Booster 7, which updated 21 drivers that Device Manager told me were fine. Among these were the PCI Bus driver and others, which are essential to how Windows handles IRQ (se Valis' excellent explanation, stickied to the top of this subforum).

6. Once all drivers were ACTUALLY updated, I was down to one click every 1:30 mins. I was then able to make use of the suggestion earlier in this thread to disable components that share the same IRQ as my Pulsar cards. In this case, I disabled six different USB driver (my box still has 2 left) and voila - in my latest test I went 5 mins before I heard one (faint) click. Hurray!

Thanks to Creamware4Ever, Valis, and GaryB for excellent guidance! I'm quite sure that for GaryB and Valis there is a lot of "Duh, that's what we were saying all along" here, but I guess sometimes we have to learn for ourselves before we can actually take in good advice. I hope this story might be helpful for someone in a similar situation.
Post Reply