Jump to content
IGNORED

Euphony OS w/Stylus player setup and issues thread


Recommended Posts

1 hour ago, davide256 said:

awful idea unless you have the engineering skills and material resources of Aurender or Innuos. Stock MOBO inputs for SDXC, SATA and M.2 NVMe

compromise audio quality vs buffering same over Ethernet from NAS or streaming service.  Optane drive works well but is cost prohibitive.

Connecting media by USB doesn't change this.

I compared Euphony's SQ using same copy of music stored on NAS (powered by LPS) vs same Music stored on external HDD (USB3 cable connected to HDD Docking Station using iFi iPurifier3, docking powered by LPS). I prefer the SQ of the latter. 

Link to comment
5 hours ago, davide256 said:

sounds like something in your audio chain is compromising the midrange... tone color density, dynamics and background instrument separation should be

better with NAS. There are devices that call themselves a NAS but aren't worthy of the name. I had a Buffalo NAS once, it was terrible sounding

compared to using Synology NAS, all streamed music sound squashed and lifeless. Since you don't post a system, hard to know whats coloring your sound.

 

It is a Synology NAS. The only noticeable difference is the HDD-attached music is a little bit more dynamic. 

Link to comment
3 hours ago, NanoSword said:

Finally, Very happy my wife always adds a small router to improve the wireless signal but this step makes me hate my system. This small router causes poor sound quality.I disconnect the network cable every time so that the sound is not affected, but now there is a solution without moving my lazy ass while I am drinking coffee and listening to the music .

 

Sometimes the coffee I drink improves the soundstage and the details I think caffeine works better than this little router.

Alcohol works the best! 

Link to comment
  • 1 month later...
  • 3 weeks later...
10 hours ago, jacky5555 said:

Hi, anyone who tried 2 box solution for stylus server + endpoint, and how it compare to 1 box, in term of SQ?

My current server is a fanless NUC8i7 which is nearly identical to the Summus server, and I'm thinking of another NUC8i3 for the endpoint, if the sound improvement's worth it. 

I have done extensive and extended (trial extension) testing regarding this 2 box Euphony config. The best I have so far is Euphony Server on NUC 8i3 to a Windows UPNP Endpoint on Intel NUC 7i3. The endpoint is runs on Windows Server 2016 core and running Jplay Femto. This 2 box setup is much better than a stand alone Euphony in my system. It may be due to the DAC driver which is developed specifically for Windows. The generic Linux DAC driver on Euphony could not even do Native DSD 256. Both the machines are powered by oversized linear power supplies. 

Link to comment
  • 1 month later...

421 is smoother sounding than 417 but soundstage of 417 is wider and deeper and imaging is better too with the previous release. 

 

While on it, I played around with the CPU Isolation and Interrupt assignments. Here's what I ended up with my NUC 8 logical cores CPU.

 

CPU Isolation:

0-1 stylus 2-3 gstp 4-6 

 

Interrupt CPU assignment: 

xhci_hcd = Core 7 (for USB port connected to DAC)

eth0 = Core 2 (for Ethernet port)

nvme* = distributed to either Core 0 or Core 1 (nvme0q0 to nvme0q8)  

 

I did not assign dedicated Interrupt cores to the following as interrupt counts for them is insignicant in my system:

   

timer
rtc0
acpi
INT34BB:00
i801_smbus
PCIe PME

 

The above gives the balance between resolution, attack/decay and fullness to the sound. Giving a lot of cores to gstp (e.g. 0-2 gstp 3-7) resulted in ultra resolution but losses on overall musicality. I will let the system settle in and will revisit the settings but for now I'm enjoying this new version of Euphony.  

Link to comment
  • 2 weeks later...
1 hour ago, edwardsean said:

Okay, okay, okay, I've converted to 421.

 

I kept playing with 421 and what finally did it was that I disabled Ramroot. That alleviated the tightness I was experiencing in 421's presentation. 421 has this incredible precision and definiteness where 102 can image overly soft and diffuse. However, with Ramroot engaged, the sound feels like the screws are overtightened and it feels stiff.  I don't like giving up Ramroot, but once I unloaded it, a natural sense of flow returned. 

 

I still long for more of 102's fullness, larger low end and staging. Also, I am uncomfortable losing Ramroot from a technical standpoint. I had to remind myself that the only that counts is the end result sound. Well, you can't have everything. This is still quite a good net win, and I'm glad to be going forward with Euphony!

 

 

I too felt the tightness or more mechanical presentation but since I use tubes, I just reduced the bias current of the tubes and voila!

Link to comment
  • 2 weeks later...
54 minutes ago, dc-audiogeek said:

I went back to Stylus version 108 with my Summus because I thought 421 sounded a bit dull in comparison. Streaming directly to my DAC via Ethernet sounded better than the Summus on version 421, more lively. Version 108 and direct Ethernet still sound different, with Ethernet having a little more liveliness but a little less focus and refinement.  I may flip back to 421 and compare again. I would love to have a version of Stylus that's a clear winner sound-wise. 

Hmm, what is your DAC? That may be the way to go moving forward. By removing Euphony/Summus means 1 less component to thinker on. 

Link to comment
On 5/14/2021 at 1:50 AM, davide256 said:

Mehh... I'm looking for lower electrical noise feeding back into the PS bus and screwing with USB out. Speed isn't relevant unless I use

HQPlayer and using HQPlayer in NUC USB out server creates its own noise issues.

For less digititis in my system, I set my CPU frequency to the I/O Bus clock frequency of my RAM. For my DDR4-2400 RAM, the clock speed is 1.2Ghz so I set my CPU speed to 1.2Ghz.

Link to comment
12 minutes ago, jbm_2021 said:

What is the default speed of your CPU?

My NUC 8i7 has a native CPU frequency of 2.7Ghz and Turbo up to 4.5Ghz. I also tried 2.4Ghz which is the double data rate of the RAM but I prefer the sound with with the 1.2Ghz single data transmission per clock cycle. 

Link to comment
22 minutes ago, Lukasluis said:

My NUC 8i7 has a native CPU frequency of 2.7Ghz and Turbo up to 4.5Ghz. I also tried 2.4Ghz which is the double data rate of the RAM but I prefer the sound with with the 1.2Ghz single data transmission per clock cycle. 

BTW, I'm using a pair of HyperX Impact low latency RAMS. I tried the Apacer slow RAMs but I felt a lot of inner details and air are lost plus PRAT is much better with the fast RAM. The Apacers made the music boring in comparison. 

Link to comment
9 minutes ago, jbm_2021 said:

CPU frequency and memory frequency matching does not add up to me. There are so many factors that kick in when it comes to bottlenecks. The bandwidth of the memory lanes have direct impact on latencies. Just by matching CPU frequency to memory frequency will not mean much unless it is purely by accident. Just my thoughts based on prior experience.

 

It mean much to my ears. Same findings on both Euphony and Windows servers that I have. 3 machines, 2 different Players, same findings. 

Link to comment
  • 2 weeks later...
On 5/23/2021 at 1:13 AM, davide256 said:

Done listening @Lukasluis the combination of 1.2Mhz CPU speed  + Apacer RAM has really helped. Was aware of added irritants before when PCM

sample rate went up, thats gone now and if I bump up CPU speed can detect more high frequency stridency. Thanks for the speed tip!

Glad it worked for you too. It was a frequency I used when I still use Euphony. 1.2 Ghz works well for my Windows Control PC too; I use 600Mhz for my Audio PC. 

Link to comment
  • 2 weeks later...
9 hours ago, genvirt said:

Hello!

Yesterday I moved to Tidal HiFi subscription , but not really understand if this make sense when using Euphony Stylus.

When playing M signed tracks , I see same 16/44 pcm.

How to understand if I’ve got really hires?

Thanks.

Euphony does not support Tidal Masters

Link to comment
  • 4 weeks later...
10 hours ago, [email protected] said:

Hi, 

I am intend to build an Euphony upon this PC

Key Feature

  • Intel® Celeron® J4105, Quad-Core 1.5-2.5GHZ
  • Dual Gigabit Ethernet
  • Raspberry Pi 40-Pin Compatible
  • 2 x M.2 PCIe (B Key and M Key)

X86-08-n.png

 

And I plan to use Raspberry Pi 40-Pin Compatible to output digital signal to DAC via this HAT

digione.jpg

SO I wonder whether Euphony support this configuration?

 

Thank you.

 

Interesting, let us know where to source it. I would like to try it myself. 

The best configuration I have with Euphony is using Stylus with UPNP output to my JPLAY Audio Pc. As more and more features are added to Euphony, I find it less and less musically-engaging. 

Link to comment

Been using Windows DUAL-PC booted in RAM for a while now using the NUC I used with Euphony thus I disabled all devices except for USB and I use USB-Ethernet adapter. Recently I tried Euphony on USB-NVME adapter with Optane drive in it and  booted Euphony from there and sending the music to Upnp JPLAY and that sounded very good. I got curious why is it so good so I decided to try Euphony alone with the same hardware config. Euphony never sounded this good!

As this is a new HW config I am using Trial version 20210421 and Root copied to RAM. 

Link to comment
23 minutes ago, c-w said:

 

I'll admit that I never looked at any network driver or NIC firmware implementation but I found all your premises suspect:
1) Disabling the network stack surely doesn't 'shutdown' the card (whatever "shutdown" should mean...)
2) If the network were completely shutdown why would it be constantly checking anything?
3) Why would the card check if the cable is there - what is there to check? I would rather assume that connecting the cable triggers the activity in the card's firmware.
4) Even if the NIC would constantly check if the cable is there, why would this create "network activity"? This would be entirely inside the card and since I cannot assume that disabling network interface in kernel somehow powers-down the card this would hardly influence anything since some kind of idle loop must be going on in the card firmware at all times (enabled/disabled, cable connected or not...)

 

I may be completely wrong (i.e., while disabled, NIC is in low power mode and removing the cable indeed wakes it up increasing the noise from card circuits...), but all this is moot anyway... Somebody ought to really check it. 

I disabled the internal NIC of my NUC and now using USB to Ethernet adapter. Also disabled the Sata and pci controllers for Optane and now using a USB to NVME adapter. 

 

Could this be one of the reasons why I'm liking Euphony now? BTW I disabled hyperthreading too. 

Link to comment
19 minutes ago, seeteeyou said:

16271840166335151718828974324184.jpg

4i3TQHXl.jpg

PBBEEjH.jpeg

 

(Since it might be somewhat tricky to figure out what's in the picture, I'm assuming that you're feeding 5V to the micro USB input of iDefender 3.0 or maybe the USB-C input of iDefender+)

 

Is your Orico M2PJM-C3 enclosure able to downregulate the voltage from 5V input to 3.3V output automatically? The lowest voltage of iPower is 5V and they never made anything with 3.3V so far.

 

Besides, did you get a chance to compare the differences between adding and taking away iSilencer? Thank you.

Yes, I'm feeding iFi 5volts to the iDefender and yes the Orico is made for USB input so I guess it has a built-in regulator. The iSilencer+ adds a touch of clarity to the setup. 

Link to comment
3 hours ago, Lukasluis said:

Yes, I'm feeding iFi 5volts to the iDefender and yes the Orico is made for USB input so I guess it has a built-in regulator. The iSilencer+ adds a touch of clarity to the setup. 

After all of these, I now find that USB thumb drives sound better still. And the slowest thumb drive (Sandisk Cruzer Blade) sounded the best as compared to the higher speed ones. 

Link to comment
  • 2 weeks later...
8 hours ago, beautiful music said:

I'm not sure if there is an alternative way to stream sound from Android Phones like AirPlay audio server for iOS.

 

This is off topic but mostly I use BubbleUpnp on my android devices stream to JplayFemto Renderer. Best combi I tried. 

Link to comment
49 minutes ago, anhton82 said:

Thanks for your help. So I'm still continue use Mac M1 with BigSur and HQPlayer. Can not try Euphony for now, just hope maybe in future.

BigSur and HQPlayer is best!   Especially that the M1 chip works well with the filters and upsampling algorithms. Plus library management is better with Hqp in my opinion. 

Link to comment
  • 1 month later...

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



×
×
  • Create New...