Help - USB DAC drops out after resuming from sleep

[SpittinFax]

Hi all,

I've been trying to resolve a USB connectivity issue for a long time now and have reached the end of my troubleshooting abilities. I need some forum help because I've run out of ideas.

WHAT THE ISSUE IS - My Ifi Zen V2 external USB DAC works great, until I put my computer into sleep mode. There's at least a 30% chance my DAC will be disconnected and I have no sound when I resume. Device is completely gone in the Sound options and needs to be re-plugged to work again. I use sleep mode a LOT when I'm AFK so this is a very annoying issue.

TROUBLESHOOTING STEPS SO FAR (with no success):

1. Different USB ports on the motherboard.
2. Audio drivers (installed and uninstalled)
3. DAC firmware update
4. Using an external USB hub (powered and unpowered).
5. Using an internal USB expansion card.
6. Different cables.
7. Auxiliary power supply versus USB power only.
8. Changing Windows power option profiles and settings (including "USB selective suspend").
9. Changing Windows sound options and enhancements.
10. Disabling other audio devices.
11. Checking Event Viewer logs for errors (couldn't find anything)
12. Changing BIOS options (including USB legacy mode).
13. Disabling unused devices in BIOS.
14. Using PCI Gen 3 instead of Gen 4
15. Updating to a newer BIOS version

SYSTEM SPECS: Ryzen 5 5600X, Asus B550-F Gaming (WiFi) on BIOS version 2604, 64GB G.Skill Ripjaws V 3600 CL18, 1TB Samsung 970 Evo, Nvidia GTX 1060 6GB, Windows 10 Pro (latest updates).

IMPORTANT NOTE: THIS ISSUE DOES NOT OCCUR ON INTEL MACHINES RUNNING THE SAME WINDOWS VERSION. I have also used an identical Ifi Zen V2 DAC (yes, I own two of them) and it displays the same behavior on my computer. The root cause seems to be specific to my Ryzen hardware setup.

The only solution I've found so far is to re-plug the device when it has dropped out. Or to just avoid using sleep mode, but avoiding the issue doesn't solve anything....

Open to thoughts on this.

 

[tabascosauz]

Not sure as I don't use an Asus board anymore, but there should be options related to providing USB power during sleep, somewhere under the Advanced tab or wherever the other peripheral settings are (eg. LAN port, Wifi, etc). Might be called S3 sleep or something, or maybe USB power when power off. Could be worth a try.

FWIW, it doesn't make a difference on any AM4 board I've had whether I've got that setting enabled - I have a E10K and Scarlett Solo and both always fire up after sleep.

 

[Mussels]

Are the latest ryzen chipset drivers installed?

There are some on vendor websites like MSI and gigabyte newer than whats on the AMD website directly

 

[SpittinFax]

Not sure as I don't use an Asus board anymore, but there should be options related to providing USB power during sleep, somewhere under the Advanced tab or wherever the other peripheral settings are (eg. LAN port, Wifi, etc). Might be called S3 sleep or something, or maybe USB power when power off. Could be worth a try.

FWIW, it doesn't make a difference on any AM4 board I've had whether I've got that setting enabled - I have a E10K and Scarlett Solo and both always fire up after sleep.

There's a setting in the BIOS in the onboard devices configuration section called something like "USB power delivery in soft off state (S5)". I tried both Enabled and Disabled with no change. I thought that surely there would be more USB or sleep settings in there somewhere, but apparently not.

Are the latest ryzen chipset drivers installed?

There are some on vendor websites like MSI and gigabyte newer than whats on the AMD website directly

Good idea, I tried though and no luck. I installed the complete chipset software package.

I sent through a ticket to Ifi requesting assistance, might need to do one through Asus as well. But it takes a while to get past the initial "did you try turning it off and on again" questions.

UPDATE:

So I tried uninstalling Asus Armory Crate to see if that was the issue. Didn't fix it. While in the Apps section I saw that a few Realtek programs were installed, so I plucked those out of there because it's not the first time Realtek drivers have caused conflicts. I had a long run of no drop-outs, but still it eventually dropped out.

Looks like when I do get a drop-out, I need to re-plug both the power supply and the USB cable to the DAC in order to bring it back. Re-plugging just the USB cable alone doesn't bring it back.

I hate these type of issues.

UPDATE 2:

I finally found out how to get the Hibernate option. Typed "powercfg /hibernate on" into Command Prompt and then I could enable it via power settings. I might try using Hibernate instead of Sleep, since hopefully that gives the USB devices a full power-on cycle each time and avoids this issue.

 

[Assimilator]

Almost certainly the DAC itself is the problem - seems like it isn't properly responding to the system resume event, most likely because its firmware and/or driver is/are buggy. So contacting anyone except the DAC's manufacturer is going to be a waste of time. (It could even be a potential hardware fault with your DAC unit itself.)

It could also be an incompatibility between the DAC and your PC - you could verify this by connecting the DAC to a different PC and seeing if the same problem occurs.

 

[SpittinFax]

Almost certainly the DAC itself is the problem - seems like it isn't properly responding to the system resume event, most likely because its firmware and/or driver is/are buggy. So contacting anyone except the DAC's manufacturer is going to be a waste of time. (It could even be a potential hardware fault with your DAC unit itself.)

It could also be an incompatibility between the DAC and your PC - you could verify this by connecting the DAC to a different PC and seeing if the same problem occurs.

I conveniently have an identical DAC (albiet different date of manufacture) that exhibits the same issue on my Ryzen system. When I test them on my Xeon X99 system however, no problems at all. So the good news is that it's not a hardware fault with the DAC. These DACs work all day on chinese Huananzhi motherboards and laptops, which (I'm speculating....but) would suggest a nuance with this particular system. I don't have the ability to drill down far enough to find out more though.

Fortunately I'm having success with using Hibernate instead. FINALLY, a workaround that doesn't involve re-plugging. A small mercy indeed.

Not sure how relevant this is, but Ifi state this on their website:

But then my device is already getting a constant 5V of power from the wall. So......a dead end as far as I'm concerned.

 

[Deleted member 24505]

The handshake thing must be the problem somehow. I'm using a cambridge USB dac and pretty sure it does not have this problem

 

[Bill_Bright]

I was going to suggest trying a different computer, but...

IMPORTANT NOTE: THIS ISSUE DOES NOT OCCUR ON INTEL MACHINES RUNNING THE SAME WINDOWS VERSION. I have also used an identical Ifi Zen V2 DAC (yes, I own two of them) and it displays the same behavior on my computer. The root cause seems to be specific to my Ryzen hardware setup.

Since it works fine on Intel based motherboards, I have to assume this is a BIOS Setting issue with your AMD based boards. USB is USB and therefore, should not matter if the board supports Intel or AMD. And it would not be a Ryzen thing either.

I think you need to dig deep into the BIOS Setup Menu.

You say you changed the selective suspend settings - but not to what. See https://www.startech.com/en-us/faq/usb-general-windows-sleep-mode

Almost certainly the DAC itself is the problem

I don't see how can it be the DAC itself if it works fine and as expected on Intel based motherboards.

 

[Mussels]

This could be related to the PSU and its 5V SB rail?

 

[robot zombie]

Almost certainly the DAC itself is the problem - seems like it isn't properly responding to the system resume event, most likely because its firmware and/or driver is/are buggy. So contacting anyone except the DAC's manufacturer is going to be a waste of time. (It could even be a potential hardware fault with your DAC unit itself.)

It could also be an incompatibility between the DAC and your PC - you could verify this by connecting the DAC to a different PC and seeing if the same problem occurs.

Hmmm... I'm remembering a time way back when Schiit Audio used to have a lot of problems with their USB DACs and retaining connectivity. One big one was definitely a driver issue that if I recall did affect only certain machines and configurations, loopy as that may sound. But a lot of that also had to do with their earlier DACs being hardware-limited to older protocols which actually made them spotty with the newest operating systems.. Some of their more expensive models featured modular slots for I/O and later functionality upgrades - pretty cool idea, all things considered. Those would get new USB cards that didn't have problems with dropouts and sleep/wake, and also worked plug and play.

I wish I remembered more about it. I only caught it at a glance, and that was years back now, when they were a spanking new company. Honestly, I don't understand how USB works nearly well enough to explain what was going on with them. But what you said made me think of that.

 

[SpittinFax]

The handshake thing must be the problem somehow. I'm using a cambridge USB dac and pretty sure it does not have this problem

The odd thing is that I don't recall having any issues in my first few months of ownership. So it could be that something has changed that I don't know about. Definitely not BIOS, could be possibly a rogue app or Windows update that's thrown it off the tracks at some point. I might force update Windows on a secondary system and test that.

I was going to suggest trying a different computer, but...


Since it works fine on Intel based motherboards, I have to assume this is a BIOS Setting issue with your AMD based boards. USB is USB and therefore, should not matter if the board supports Intel or AMD. And it would not be a Ryzen thing either.

I think you need to dig deep into the BIOS Setup Menu.

You say you changed the selective suspend settings - but not to what. See https://www.startech.com/en-us/faq/usb-general-windows-sleep-mode


I don't see how can it be the DAC itself if it works fine and as expected on Intel based motherboards.

Certainly I've combed through all the connectivity-related settings in the BIOS and had no luck with changing anything remotely related. Haven't dug deep into the CPU overclocking settings though. My setup has been running a PBO negative offset for probably a year so far.

This could be related to the PSU and its 5V SB rail?

Yeah I forgot to mention the power supply. The one in my system is a Be Quiet! Straight Power 11 650W Platinum, but I also use a 5VDC wall socket adapter (not sure the brand) for the DAC. Using either source doesn't make a difference.

Hmmm... I'm remembering a time way back when Schiit Audio used to have a lot of problems with their USB DACs and retaining connectivity. One big one was definitely a driver issue that if I recall did affect only certain machines and configurations, loopy as that may sound. But a lot of that also had to do with their earlier DACs being hardware-limited to older protocols which actually made them spotty with the newest operating systems.. Some of their more expensive models featured modular slots for I/O and later functionality upgrades - pretty cool idea, all things considered. Those would get new USB cards that didn't have problems with dropouts and sleep/wake, and also worked plug and play.

I wish I remembered more about it. I only caught it at a glance, and that was years back now, when they were a spanking new company. Honestly, I don't understand how USB works nearly well enough to explain what was going on with them. But what you said made me think of that.

That's a pretty neat idea to have upgradeable USB I/O. I looked at Schiit but didn't buy one, the import tax here effectively kills that brand as an option.

 

[Mussels]

Ok so it uses the USB 3.0 type B (printer cable) at the back

Have you tried USB 2.0 cables?

USB C cables?

USB 5V from a PC instead of the power brick?
(run the power from an intel machine, with the data to the AMD. rule out the power source as the cause?)

 

[Deleted member 24505]

I am actually using a C to USB B on my Cambridge DAC

 

[Ferrum Master]

All USB are not equal. An have special feature sets, with some being depreciated for security reasons and yet some older controller drivers like NEC/Renesas have ignored Microsoft regulation and still incorporate it. They are very controller/driver specific, and the older boards actually have more legacy features than the new ones.

The worst case, some old designs rely on those features, like additional headers/descriptors. If i hook up, some older phone in bootloader, I can trigger an instant BSOD on any PC. USB was responsible for considerable amount of BSOD's and Microsoft took measures for it years ago.

Basically, it ain't the blame of AMD, that it filled out the recommendation, ditching some USB features. Well you mouse and keyboard works fine after idling, so basically, it is the problem of the DAC really.

One of the reasons I am not fond of using USB for audio.

 

[SpittinFax]

So I've been using Hibernate for a day or so and have had zero issues. Definitely the full initialization of Hibernate mode helps get around the issue.

I was experimenting yesterday to see if the time between sleep and wakeup is a factor, and it isn't. So it's not like the connection is "timing out". There's something in the power-on procedure that's not working correctly.

Gonna do some testing on another Windows 10 laptop tomorrow to see if I can replicate the problem.

EDIT: Just got a response from Ifi. "This is a known issue that we are looking into, we have passed this information to our development team to look into for a possible future firmware update to fix this issue".

UPDATE: I tried with my Lenovo Thinkpad (10th Gen Intel) laptop and couldn't replicate the issue. Not a single drop-out. This is definitely machine (or platform) specific.

 

[Mussels]

So I've been using Hibernate for a day or so and have had zero issues. Definitely the full initialization of Hibernate mode helps get around the issue.

I was experimenting yesterday to see if the time between sleep and wakeup is a factor, and it isn't. So it's not like the connection is "timing out". There's something in the power-on procedure that's not working correctly.

Gonna do some testing on another Windows 10 laptop tomorrow to see if I can replicate the problem.

EDIT: Just got a response from Ifi. "This is a known issue that we are looking into, we have passed this information to our development team to look into for a possible future firmware update to fix this issue".

UPDATE: I tried with my Lenovo Thinkpad (10th Gen Intel) laptop and couldn't replicate the issue. Not a single drop-out. This is definitely machine (or platform) specific.

In other words, they designed and tested it on intel machines
(Remember that intel haven't changed their USB ports in a loooooooong time)

 

[eidairaman1]

Hmmm... I'm remembering a time way back when Schiit Audio used to have a lot of problems with their USB DACs and retaining connectivity. One big one was definitely a driver issue that if I recall did affect only certain machines and configurations, loopy as that may sound. But a lot of that also had to do with their earlier DACs being hardware-limited to older protocols which actually made them spotty with the newest operating systems.. Some of their more expensive models featured modular slots for I/O and later functionality upgrades - pretty cool idea, all things considered. Those would get new USB cards that didn't have problems with dropouts and sleep/wake, and also worked plug and play.

I wish I remembered more about it. I only caught it at a glance, and that was years back now, when they were a spanking new company. Honestly, I don't understand how USB works nearly well enough to explain what was going on with them. But what you said made me think of that.

A pretty bad name for sure

 

[robot zombie]

A pretty bad name for sure

You could say so, but it worked out for them, in terms of getting it out as a new company. People, erm... crapped on it for a brief period but ultimately I think people actually took to it It helps that the products themselves have generally gone over well for them over the years. Their budget stuff these days, I'd say is pretty good for the money and probably the most your average person needs to be spending on any of that stuff if they want discreet headphone output at their PC. They're Schiit, not the other word

I mean, you really would think it would limit them but the audiophile world is basically a big intermixed boy's club for rich weirdos in a lot of ways. They embrace it. When COVID first came and there were TP shortages, I remember seeing them stock rolls of toilet paper with their logo printed on each sheet. They're basically a cult brand now, sort of embedded in the internet lore. Dude... when it comes to the way internet communities and commerce mix, I feel like I am continuously catching up to the things that pop out of those things mingling reallllyyy closely.

 

[Mussels]

The weird name worked for them, because it turned them into an 'in' crowd
You explain that no schiit audio is actually not shit, and they feel like they learned something and discovered some secret to these cool foreign people with their fancy audio stuffs


I'm curious how this turns out long term, hopefully it's a firmware fix

 

[SpittinFax]

You could say so, but it worked out for them, in terms of getting it out as a new company. People, erm... crapped on it for a brief period but ultimately I think people actually took to it It helps that the products themselves have generally gone over well for them over the years. Their budget stuff these days, I'd say is pretty good for the money and probably the most your average person needs to be spending on any of that stuff if they want discreet headphone output at their PC. They're Schiit, not the other word

I mean, you really would think it would limit them but the audiophile world is basically a big intermixed boy's club for rich weirdos in a lot of ways. They embrace it. When COVID first came and there were TP shortages, I remember seeing them stock rolls of toilet paper with their logo printed on each sheet. They're basically a cult brand now, sort of embedded in the internet lore. Dude... when it comes to the way internet communities and commerce mix, I feel like I am continuously catching up to the things that pop out of those things mingling reallllyyy closely.

The weird name worked for them, because it turned them into an 'in' crowd
You explain that no schiit audio is actually not shit, and they feel like they learned something and discovered some secret to these cool foreign people with their fancy audio stuffs


I'm curious how this turns out long term, hopefully it's a firmware fix

The sticking points for me with Schiit Audio is that the only combo DAC + headphone amp product they have (that I know of) is the Schiit Fulla 3. Here it costs the same as an Ifi Zen V2, but the reviews seemed to suggest it wasn't a standout product (not amazing audio quality, no RCA out, no balanced circuits). Oh and of course they're usually out of stock anyway. They don't really have the opportunity to stand out in the non-US markets due to pricing and availability.

Aside from this drop-out issue though, the Ifi Zen actually doesn't leave anything to be desired for myself personally. It'll probably be my first and last DAC. And now they've released the budget Zen Air DAC version with cut-down features for US$100 and similar audio quality to the standard US$160 version. Wouldn't mind getting one when they're in stock.

 

[Ferrum Master]

The sticking points for me with Schiit Audio is that the only combo DAC + headphone amp product they have (that I know of) is the Schiit Fulla 3. Here it costs the same as an Ifi Zen V2, but the reviews seemed to suggest it wasn't a standout product (not amazing audio quality, no RCA out, no balanced circuits). Oh and of course they're usually out of stock anyway. They don't really have the opportunity to stand out in the non-US markets due to pricing and availability.

Aside from this drop-out issue though, the Ifi Zen actually doesn't leave anything to be desired for myself personally. It'll probably be my first and last DAC. And now they've released the budget Zen Air DAC version with cut-down features for US$100 and similar audio quality to the standard US$160 version. Wouldn't mind getting one when they're in stock.

Just get a Topping dude... my five cents...

 

[Deleted member 24505]

Or if you have the cash, a Cambridge Audio DacMagic Plus or Cambridge Audio DacMagic 200M

 

[SpittinFax]

Just get a Topping dude... my five cents...

I hear they're good but I would need an amp for it. The Zen DAC doesn't have any analogue inputs so I can't use it as a standalone amp. I've heard good things about the Zen CAN headphone amplifier (which has a budget "Air" variant coming mid-year) but it's powerful and apparently doesn't play nice with low-impedance headphones.

Or if you have the cash, a Cambridge Audio DacMagic Plus or Cambridge Audio DacMagic 200M

I'd benefit from a headphone upgrade before a DAC/amp upgrade. I've got Sennheiser HD599's running on the balanced 4.4mm output, but even then, there's no need. I'm happy with the audio quality at the moment. The over-arching principle of my setup is that my gear is good enough to enjoy music while avoiding the slippery audio upgrade addiction slope.

 

[Deleted member 24505]

I hear they're good but I would need an amp for it. The Zen DAC doesn't have any analogue inputs so I can't use it as a standalone amp. I've heard good things about the Zen CAN headphone amplifier (which has a budget "Air" variant coming mid-year) but it's powerful and apparently doesn't play nice with low-impedance headphones.



I'd benefit from a headphone upgrade before a DAC/amp upgrade. I've got Sennheiser HD599's running on the balanced 4.4mm output, but even then, there's no need. I'm happy with the audio quality at the moment. The over-arching principle of my setup is that my gear is good enough to enjoy music while avoiding the slippery audio upgrade addiction slope.

I'm just using the Cambridge DacMagic 100 which was only £80, then using the ana outs to my speakers. The difference sure is noticeable. I will get a headphone amp at some point. I did look at the ifi but this cambridge is still £160 new, and the ifi is £130 so thought i might as well just keep this.

 

[MarekRMac]

So I've been using Hibernate for a day or so and have had zero issues. Definitely the full initialization of Hibernate mode helps get around the issue.

I was experimenting yesterday to see if the time between sleep and wakeup is a factor, and it isn't. So it's not like the connection is "timing out". There's something in the power-on procedure that's not working correctly.

Gonna do some testing on another Windows 10 laptop tomorrow to see if I can replicate the problem.

EDIT: Just got a response from Ifi. "This is a known issue that we are looking into, we have passed this information to our development team to look into for a possible future firmware update to fix this issue".

UPDATE: I tried with my Lenovo Thinkpad (10th Gen Intel) laptop and couldn't replicate the issue. Not a single drop-out. This is definitely machine (or platform) specific.

Hi. I've go the same issue with my Zen Dac V2.
It's good to know that Ifi is aware of this problem.

My platform is AMD Ryzen 3800 on B450 chipset (Msi B450M Mortar Max). Problem was present on Windows 10pro, and i've recenty updated it to Win11pro.

I don't want to use hibernation, because I think it's no good for SSD (on each hibernatie all RAM is written to SSD).

Yesterday I've installed firmware 7.40 for Zen Dac, I'm curious if it will help.