Jump to content
austinpop

A novel way to massively improve the SQ of computer audio streaming

Rate this topic

Recommended Posts

I already put no turn off monitor or hard disk when plug in power.

 

Perhaps you should check the property setting specifically for the wireless controller. Sometimes you have the option of allowing windows to shut down the device to save power according to power scheme

Share this post


Link to post
Share on other sites
OK I have some findings. Very interesting results! Thanks for pointing me in this direction.

 

I compared the following:

 

  1. switch > FMCs > Aries Mini
  2. switch > Aries Mini
  3. bridge > Aries Mini
  4. bridge > FMCs > Aries Mini

 

1 is my current configuration. My previous results were based on comparing 4 with 1.

 

Comparisons

  1. 2 vs 1 (remove FMCs): yup, expected drop in SQ.
  2. 3 vs 2 (switch to bridge, no FMCs): wow, big improvement!
  3. 3 vs 1: this is a wash, to my ears. About the same. See conclusions below
  4. 4 vs 3: (add FMCs between bridge and Aries Mini): minimal impact, if any.

Conclusions

 

In my setup, the bridged connection provides about the same improvement as putting FMCs between the switch and my Aries Mini. Refer to my diagrams in my previous post. The two FMCs are powered by Teralink X1/X2 (upstream FMC) and the Uptone LPS-1 (downstream FMC).

 

So on the one hand, by going bridged, I can eliminate the FMCs, and repurpose my LPS-1 for something else. But on the other hand, I abhor the presence of the PC in my listening room. And - I still have some lingering network issues with the bridge. In addition to the SMB woes I mentioned in my previous post, I also found today that my control point (lightning DS app on my iPhone) would lose contact with the Aries Mini in the bridged configuration. I know this relies on UPnP and multicast support in all routers and switches, so it's possible the bridge messes with that.

 

The right answer for me, at this point, is to close the chapter on bridging and go back to my switched environment. At least I won't be wondering if I was missing out on some unclaimed SQ improvements.

 

I may revisit this if/when I decide to get an SMS-200, and/or try Roon.

Austinpop,

 

I think your weak link here is the Auralic device. Indeed any dedicated device is going to be tough to modify, and a DIY solution will give you the most flexibility, best value for money, and highest sound quality.

 

I resisted the move to a DIY device for a long time, but am now delighted with the results. To my knowledge, the various technologies I was able to integrate do not exist in a commercial product, and it is the unique combination of these that had led to the best sounding solution.

 

You seem to have the both the curiosity and hardware and software technology skills, which suggests you are likely to succeed.

 

Let me know if I can help in anyway.

Share this post


Link to post
Share on other sites
While my client naa is Intel nuc but running audiolinux. After music stopped, I opened settings and saw music device still highlighted and I would press OK to pick it up again naa connection. Then I could play another 1 hour plus before it would stop again. As you said very predictable behavior. If I don't click settings I got to continue clicking the song in Playlist until it plays again

 

Sent from my Redmi Note 3 using Computer Audiophile mobile app

 

Hmmm, have you checked if you have any power / energy saving modes active for the OS or wireless controller?

I had a very similar issue when I was using a USB to ethernet adapter to act as my second nic when I was using Jplay last year. This nic was attached to a wireless AC bridge. The adapter would essentially power down after some period of time. The only way to get it to operate correctly again is to unplug and plug it into the USB port again. The PC using the USB adapter was running Server 2012 R2 Essentials, AO, and Fidelizer. I messed around with all kinds power settings and even bought a second ethernet adapter but could never resolve this issue. I never had a problem with the internal nic. I know not the answer you were looking for.

 

BTW, off topic but I will be trying out Audiolinux as my NAA as well soon to compare it to Server R2 +AO + Fidelizer. This will be running on an i3-5015u mobile cpu.

Share this post


Link to post
Share on other sites
While my client naa is Intel nuc but running audiolinux. After music stopped, I opened settings and saw music device still highlighted and I would press OK to pick it up again naa connection. Then I could play another 1 hour plus before it would stop again. As you said very predictable behavior. If I don't click settings I got to continue clicking the song in Playlist until it plays again

 

Sent from my Redmi Note 3 using Computer Audiophile mobile app

 

Do you have wireless enabled on your modem and have a wireless router? Sometimes the 2 wireless networks fight each other and changes ip settings. Also try and increase the dhcp lease times on your modem. It might be set to 1 hour. I changed mine to 365 days.

 

 

Sent from my iPhone using Computer Audiophile

Share this post


Link to post
Share on other sites

Finally I found a way to make a network-bridge that does not lead to stuttering in my NUC-I3 :

 

Lateboomer mentioned earlier a bridge between Wifi and LAN. This worked in my setup, but I didn't like the sound so much.

 

It might be dynamic with strong bass, but I perceived a slight, rosy veil to the sound. Like I heard earlier when I connected my USB-Wifi adapter directly to the SOTM200.

 

So I tried a 3-way bridge, consisting of the NUC's LAN (connected to my SOTM with a short cable), the USB-LAN adapter (connected by cable directly to my router-modem) and my USB-Wifi adapter

 

I did not give static IP-addresses to these 3. After the bridge was formed I gave it a static IP-address. This setup (with Jriver MC20 and Jremote on Ipad) finally did not stutter and sounded as I remembered from my first try with a 2-way LAN bridge (why I can't repeat that remains unclear). The sound was very transparent and clean.

 

It stayed that way even when I broke the connection of the Wifi-dongle to my modem. But I could not inactivate the Wifi or take it out of the bridge. Then the stuttering came back.

 

I can't say I understand why this worked for me and this is probably not a final solution : I might try this setup with a cheap internal Wifi card for my NUC or discard the Wifi and try it with an extra NIC on a PCIe-card. (Attached by a riser cable to the NUC's mini-PCI-bus and dangling outside my NUC-case. not a very elegant solution)

Share this post


Link to post
Share on other sites
Do you have wireless enabled on your modem and have a wireless router? Sometimes the 2 wireless networks fight each other and changes ip settings. Also try and increase the dhcp lease times on your modem. It might be set to 1 hour. I changed mine to 365 days.

 

 

Sent from my iPhone using Computer Audiophile

The wireless router is Tplink N wireless router. I just connected the music pc laptop to this router via wifi connection with internal built-in wifi of the laptop.

 

I relied on the dhcp of this router and the router was not connected to modem or to internet.

 

Then I setup a bridge between this wifi connection and the direct Lan Ethernet connection between Hqplayer and NAA. So maybe there is settings at the router to improve connection stability which I don't know yet. But reading at latest development, things are getting weirder and weider. Hopefully good solution can be found.

Share this post


Link to post
Share on other sites
Finally I found a way to make a network-bridge that does not lead to stuttering in my NUC-I3 :

 

Lateboomer mentioned earlier a bridge between Wifi and LAN. This worked in my setup, but I didn't like the sound so much.

 

It might be dynamic with strong bass, but I perceived a slight, rosy veil to the sound. Like I heard earlier when I connected my USB-Wifi adapter directly to the SOTM200.

 

So I tried a 3-way bridge, consisting of the NUC's LAN (connected to my SOTM with a short cable), the USB-LAN adapter (connected by cable directly to my router-modem) and my USB-Wifi adapter

 

I did not give static IP-addresses to these 3. After the bridge was formed I gave it a static IP-address. This setup (with Jriver MC20 and Jremote on Ipad) finally did not stutter and sounded as I remembered from my first try with a 2-way LAN bridge (why I can't repeat that remains unclear). The sound was very transparent and clean.

 

It stayed that way even when I broke the connection of the Wifi-dongle to my modem. But I could not inactivate the Wifi or take it out of the bridge. Then the stuttering came back.

 

I can't say I understand why this worked for me and this is probably not a final solution : I might try this setup with a cheap internal Wifi card for my NUC or discard the Wifi and try it with an extra NIC on a PCIe-card. (Attached by a riser cable to the NUC's mini-PCI-bus and dangling outside my NUC-case. not a very elegant solution)

Setup your machine with a 2 port bridge. Goto the command line and run tracert followed by the address of your sotm-200. This will show you the network path taken between your PC and the sotm-200.

 

Just because both machines are directly connected doesn't mean that traffic moves over the same path. You may need to add a route to enable the most efficient path. I have seen traffic flow from the bridged machine, to the router, back to the bridged machine and then to a directed connected device. Stuttering was the result.

 

Also be careful with USB/Ethernet adapters. You don't want to share the interrupts with a USB DAC connection. In Windows Device Manager view by connections will show you two devices on the same bus if it exists. Obviously for the Sotm-200 this isn't an issue.

Share this post


Link to post
Share on other sites

 

Where are your FMCs? Mine used to sit between my computer and network player. My equipment, once NICs are bridged that allow for a direct connection between computer and network player, definitely sounds better when I remove the FMCs.

 

Several network players ago (Auralic Aries, Aurender N100H, microRendu) I installed TP Link FMCs and 10 m of fiber between the Apple AirPort Extreme used as a hub between NAS, internet feed from downstairs and Mac Mini, and my network streamer. I thought it was a worthwhile improvement in sound quality. More so when I replaced the FMCs' OEM SMPS wall-warts with Jameco linear regulated power supplies.

 

However, when I bridged the Mini's installed Ethernet port with an Apple Tthunderbolt-to -Ethernet adapter and fed my microRendu directly from the Mac, whether from the installed Ethernet port or the adapter, I was underwhelmed by the change. I thought there was a very small improvement but nothing worth giving up optical isolation for.

 

When I replaced the FMCs and fiber with a straight run of decent but nothing special CAT6 copper Ethernet cable I heard a much larger improvement with the direct Ethernet feed, more along the magnitude of what Romaz and others have reported.

 

To ensure I wasn't hearing an improvement simply from pulling the FMCs' Jameco power supplies I left the power supplies plugged in and energized and only unplugged both ends of the fiber run's jumpers from Mini and microRendu. With or without the Jameco power supplies energized the sound improvement remained the same -- that is, significantly better with the all-copper direct Ethernet feed between Mini and microRendu than with the optically isolated direct feed.

 

I also tried coiling up the fiber and using it in place of the short copper CAT6 jumpers between the Mini and AirPort Extreme and also between NAS and AirPort Extreme. In either location the sound quality dropped with fiber and FMCs installed.

 

Without fiber and FMCs in the network music sounds more vivid and tonally more saturated as well as having more a more vibrant and life-like presentation, particularly noticeable with voices. With fiber anywhere in the audio network music sounds more recessed and grayer, as if a thin veil has been placed between me and the music.

 

Other details of my system not evident from my signature -- the Mini is a late-2012 i7 with 16GB RAM, 480GB SSD running Roon and HQPlayer. I'm still using the stock internal power supply in the Mini and haven't done any specific optimization other than turning off Spotlight and only using the Mini for music. I have all the "computer stuff" (NAS, Mini, AirPort Extreme) on a CyberPower pure sinewave UPS, plugged into a Shunyata PS8 with Venom Defender in an adjoining room on a different electrical circuit from the main stereo system.

 

So go figure -- where I once thought fiber optic isolation was a significant improvement over all-copper Ethernet and that any sort of direct connection -- ethernet or USB -- between a noisy computer and an audio device was less than optimum, I have now come full circle back to an all-copper direct connection between computer and streamer.

 

Steve Z

Share this post


Link to post
Share on other sites

When I replaced the FMCs and fiber with a straight run of decent but nothing special CAT6 copper Ethernet cable...

Was this still bridged? Straight run from where to where? Do you mind elaborating as to the complete set up that you felt provided the best SQ?

Share this post


Link to post
Share on other sites

Goto the command line and run tracert followed by the address of your sotm-200. This will show you the network path taken between your PC and the sotm-200.

 

 

Question regarding tracert...i just used this command from my HQP server PC to the NAA PC. These PCs are connected by a fiber switch. When I run the tracert command, the switch does not show up in the results. The results table just shows the IP address of the NAA PC. When I try this command from/to other PCs in my house, I get the same result. I know for a fact I have a wireless router/switch in my network.

 

As I understand it, some of my switches and routers is not decrementing the TTL fields of an IP datagram. I need to look into how to activate this feature.

Share this post


Link to post
Share on other sites

In reading more about tracert, it seems if you are running this command for devices on your local network, the path through switches wont show up. Tracert will only show hops through a router. Therefore, unless I am totally wrong (highly likely) the tracert command wont be helpful for us to verify that this bridging method is working.

Share this post


Link to post
Share on other sites

This is as expected, switches shouldn't show up, routers do. If you have a one hop response between the HQplayer server and NAA that is good to know.

 

If you can get to a command line on your NAA you can try the same.

Share this post


Link to post
Share on other sites
This is as expected, switches shouldn't show up, routers do. If you have a one hop response between the HQplayer server and NAA that is good to know.

 

If you can get to a command line on your NAA you can try the same.

Thank you. So what would be a surefire way to test if the bridging approach is working as intended, i.e. data being sent directly to the downstream device (NAA, microrendu, etc) versus through the switch? I guess MACs could test by removing the connection to the switch since MACs have a DHCP server. However, how would a Windows person test?

Share this post


Link to post
Share on other sites
So go figure -- where I once thought fiber optic isolation was a significant improvement over all-copper Ethernet and that any sort of direct connection -- ethernet or USB -- between a noisy computer and an audio device was less than optimum, I have now come full circle back to an all-copper direct connection between computer and streamer.

 

This is where I've landed too. Copper is used on both spans of a bridged connection; from Mac Pro to microRendu, and from Mac Pro to wall jack leading to the rest of my network.

 

 

Sent from my iPhone using Computer Audiophile

Share this post


Link to post
Share on other sites
Setup your machine with a 2 port bridge. Goto the command line and run tracert followed by the address of your sotm-200. This will show you the network path taken between your PC and the sotm-200.

 

I have done that recently and nothing special was found : all connections showed only 1 hop with tracert. (2-LAN bridge)

 

Last weekend a network specialist (friend of my son) did more thorough testing with iperf (from iperf.fr) We even replaced the SOTM with a notebook to be able to test that connection to my NUC also from the other end. He noted nothing special : good bandwidth with normal data, but as soon as a music-player (Jriver or window's mediaplayer) started to send data to the SOTM a severe drop in bandwidth was noted and the stuttering started (2-LAN bridge). So he concluded that my problem could possibly lie with the software (DLNA server?)

 

Also be careful with USB/Ethernet adapters. You don't want to share the interrupts with a USB DAC connection. In Windows Device Manager view by connections will show you two devices on the same bus if it exists. Obviously for the Sotm-200 this isn't an issue.

 

I have no DAC connected to my NUC. In the device manager I have 3 USB root-hubs, but I can't see from that view if my USB-LAN is connected to the same hub as my Logitech remote keyboard sensor (the only other USB device). All my USB hubs have their energy saving feature turned off. (2-LAN bridge)

Share this post


Link to post
Share on other sites
Austinpop,

 

I think your weak link here is the Auralic device. Indeed any dedicated device is going to be tough to modify, and a DIY solution will give you the most flexibility, best value for money, and highest sound quality.

 

I resisted the move to a DIY device for a long time, but am now delighted with the results. To my knowledge, the various technologies I was able to integrate do not exist in a commercial product, and it is the unique combination of these that had led to the best sounding solution.

 

You seem to have the both the curiosity and hardware and software technology skills, which suggests you are likely to succeed.

 

Let me know if I can help in anyway.

Thanks, Larry, I'll keep that in mind.

 

I do have the skills, and I've spent a long career on computer system performance and scalability, both as a hands-on engineer, and then building and running teams. This may be why my posts can read like lab reports. Experimental rigor is in my DNA, as is an open mind! This has served me very well here.

 

Despite possessing the skills, my objective on this forum is primarily to maximize my music enjoyment experience. It may sound ironic, but this is why I don't want to tinker with computers in my listening room - I do that as part of my day job!

 

So yes - the tweaking here comes naturally to me, and I probably possess deeper insights than many on what is going on in the OS, but building DIY components? Nah - not my cup of tea.

 

Bringing this back to the topic at hand - I would challenge your assertion that the Aries Mini is a weak link. There are probably others, but at least @Cornan and myself have optimized the snot out of systems around our Minis and it has scaled incredibly well.

 

I am still very curious to see if devices like the mR and the SMS-200 can take me another step further, but for now, I am content. I suspect Uptone, Sonore, and others have impending products that I want to wait for before pulling the trigger on a new purchase.

 

Of course, if anyone wants to lend me an SMS-200 or mR for a couple of weeks in the interest of science, I would love to compare how it does!

Share this post


Link to post
Share on other sites
No one can fault you for your effort, that's for sure. I have been getting the sense for some time now that this "direct" connection tweak benefits mainly the mR and sMS-200, with or without HQP. This finding adds further to the mystique of what is actually happening with this direct connection. A real head scratcher...

 

Thanks, Roy.

 

My results in no way diminish the value of your bridging optimization. In fact, the more people who try different variants in different systems, the better we get a sense for the bounds and constraints, which can only add to our collective knowledge.

 

I always treat every optimization reported here on CA, no matter how many people have replicated it, as a potential improvement, until validated by my own observation. Or, stated another way, digital audio system optimization is like exploring a multidimensional space. It's easy to get lost. Treat the optimizations reported here as hints and beacons to a guided exploration, and often you will find benefits. But you still have to explore!

 

I see so many people here who helicopter in to these forums and just want "answers," or "cookbook recipes," and then get upset if they don't get the same results. ?

Share this post


Link to post
Share on other sites
While my client naa is Intel nuc but running audiolinux. After music stopped, I opened settings and saw music device still highlighted and I would press OK to pick it up again naa connection. Then I could play another 1 hour plus before it would stop again. As you said very predictable behavior. If I don't click settings I got to continue clicking the song in Playlist until it plays again

 

Sent from my Redmi Note 3 using Computer Audiophile mobile app

 

Windows has an "advanced energy or power settings" where every hardware can be set to a "sleep mode" after a specific time. Make sure they're all set to "never" or something like that. Save that profile, restart and see if the problem is gone. Ecaxt terms may be different, this is by head.

Share this post


Link to post
Share on other sites
Was this still bridged? Straight run from where to where? Do you mind elaborating as to the complete set up that you felt provided the best SQ?

 

Yes, still bridged. I removed a 10 meter fiber optic isolated run of ethernet, consisting of:

 

Bridged Mac Mini ethernet port > copper jumper > FMC > fiber optic > FMC > copper jumper > microRendu

 

and replaced it with:

 

Bridged Mac Mini ethernet port > copper ethernet cable > microRendu

 

The latter configuration sounded much better.

 

The entire "audio network" configuration is:

 

NAS and a copper ethernet cable from downstairs carrying internet > copper jumpers > Apple AirPort Extreme > copper jumper > Mac Mini bridged Thunderbolt-to-Ethernet adapter, Mac Mini onboard bridged ethernet port > copper ethernet > microRendu.

 

Running Roon and HQPlayer on the Mac Mini

 

Hope this is more clear.

 

Steve Z

Share this post


Link to post
Share on other sites

Reading that some people trying this tweak on Windows machines,

being able to set up the bridge, but unable to get HQPlayer/Roon unable to play through the mRendu in NAA mode. Are people using any optimisation software on their windows machines, like for example Fidelizer?

 

Don't know if this impacts anything, but I removed Fidelizer from my windows machine, I.e. Took the machine back to its baseline build, before changing/adding any hardware and setting up the network bridge. I got the network bridge working playing music first and then reapplied the optimiser software......just a thought....!

Share this post


Link to post
Share on other sites
Reading that some people trying this tweak on Windows machines,

being able to set up the bridge, but unable to get HQPlayer/Roon unable to play through the mRendu in NAA mode. Are people using any optimisation software on their windows machines, like for example Fidelizer?

 

Don't know if this impacts anything, but I removed Fidelizer from my windows machine, I.e. Took the machine back to its baseline build, before changing/adding any hardware and setting up the network bridge. I got the network bridge working playing music first and then reapplied the optimiser software......just a thought....!

 

I'm using Fidelizer and Audiophile Optimizer on a Windows 10 machine.

 

On my machine the bridge was created easily even with that software running.

Share this post


Link to post
Share on other sites
USB to Ethernet Adapter -

 

Supposedly this one works in promiscuous mode: Network Adapter, Anker USB 3.0 to RJ45 Gigabit.

 

I have one coming later today. Will report back soon...

 

I feel weird for quoting myself...

 

Anyway, that USB to Ethernet Adapter works really well (to make up one of the NICs in the bridge).

 

Network Adapter, Anker USB 3.0 to RJ45 Gigabit

 

$16 from Amazon.

Share this post


Link to post
Share on other sites

Tried all combinations of on-board port, usb-ethernet adpater, and 2nd ethernet card on Win8 Audio PC that has upnp server and files. Still stuttering in 3 seconds. Promiscuous mode didn't help either. Can anyone educate me how devices communicate? Bridge is 192.168.1.6 and sms-200 is 192.168.1.3. How can sound data go directly to sms-200 without help of router? Have to upgrade to Win10?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×