Audiolinux Server configurations, Software, Hardware, and Listening Impressions

[Bricki]

1 hour ago, lmitche said:

 

Later, I changed the NUC server power setting in the bios to low power just the I7 endpoint. This change delivered the best presentation I have ever heard of "Midnight Sugar" from the Three Blind Mice. Breathtaking.

 

So there you go! Any questions?

 

Larry

Does the low power setting reduce the TDP to 10w from 15w? I left the power as is in my sonicTransporter i5 but I might try having a fiddle with it now...the BIOS on the sonicTransporter is a bit utilitarian compared to the nuc but hopefully I can find a similar low power setting.

 

Can a similar low power tweak be done on the NUC7CJYH? Or is the only option there the slider bars I mentioned in the other thread that you said were good if you wanted to power from an LPS 1.2 but had no effect on SQ?

[rickca]

1 hour ago, lmitche said:

I changed the NUC server power setting in the bios to low power just the I7 endpoint

Can you please clarify this statement?  Are you talking about the server or the endpoint?

[lmitche]

18 minutes ago, rickca said:

Can you please clarify this statement?  Are you talking about the server or the endpoint?

Both are in low power mode.

[lmitche]

30 minutes ago, Bricki said:

Does the low power setting reduce the TDP to 10w from 15w? I left the power as is in my sonicTransporter i5 but I might try having a fiddle with it now...the BIOS on the sonicTransporter is a bit utilitarian compared to the nuc but hopefully I can find a similar low power setting.

 

Can a similar low power tweak be done on the NUC7CJYH? Or is the only option there the slider bars I mentioned in the other thread that you said were good if you wanted to power from an LPS 1.2 but had no effect on SQ?

Bricki,

 

Yes, low power mode moves the NUCi7DNXX to 10 watts.

 

And yes there are adjustable limits on the peak power consumption and processor modes on the Celeron and Pentium NUCs but I'd have to look at the bios to remember the details. The only model resident here at the moment is an i7 NUC.

 

Sorry I can't be of more help.

 

Larry

[Dev]

Did anyone compare Hdplex to other audiophile LPS ?

[juliocat]

I'm testing audiolinux version 0.5 for Raspberry using MPD with UpnP for MPD,  and i can't get sound at 44.1 khz, the only way to hear my 44.1 files is to upsample to 88.1 or more, 96 khz and up even DSD have no problems, i'm using Audirvana+, everything works with dietpi, but Audiolinux has better sound to my ears,  any suggestions???

 

Thanks

[greenleo]

Larry,

 

Just want to check if my understanding of your posts is correct.  Is the following table correct?

 

Time.             Mid Sep 2018                                                  Now

Config.           1.                     2.                      3.                      4.

Server.            7CJYH           i7-6700K          i7-6700K         i7DNBE (HDPlex200W)

Endpoint.       7CJYH           7CJYH              i7DNBE           i7DNBE (LPS-1.2)

SQ.                  Good/Great  Better               Even Better    Best

 

 

[lmitche]

20 minutes ago, greenleo said:

Larry,

 

Just want to check if my understanding of your posts is correct.  Is the following table correct?

 

Time.             Mid Sep 2018                                                  Now

Config.           1.                     2.                      3.                      4.

Server.            7CJYH           i7-6700K          i7-6700K         i7DNBE (HDPlex200W)

Endpoint.       7CJYH           7CJYH              i7DNBE           i7DNBE (LPS-1.2)

SQ.                  Good/Great  Better               Even Better    Best

 

 

I have no idea. Things change too often here and rankings make no sense to me.

 

Anyway, people need to listen and decide for themselves.

[Bricki]

I have selected this Max battery setting in the BIOS of my sonicTransporter i5 AL roon server. Previously I have moved it from turbo to non-turbo and heard an increase in SQ.

 

After Larry mentioned low power mode (reducing the TDP) on his server was beneficial I decided to try some settings to achieve the same result.. after selecting Max battery I have heard a quieter sound with a darker background that has better separation and more depth. I am hearing more textures on vocals and more color in piano... I also think it is more relaxed...I won't be changing it back off Max battery that's for sure 😁👍

[greenleo]

7 hours ago, lmitche said:

rankings make no sense to me.

Get it.

 

Is it correct that your current system is i7DNBE both as server and endpoint with endpoint powered by LPS-1.2 and server powered by the HDPlex200W?

[lmitche]

4 hours ago, greenleo said:

Get it.

 

Is it correct that your current system is i7DNBE both as server and endpoint with endpoint powered by LPS-1.2 and server powered by the HDPlex200W?

Hi Greenleo,

 

There are multiple systems here in various states that change all the time. Here is a snapshot of whats happened in the last few days and weeks.

• A wonderful Uptone Audio JS-2 power supply has come, and gone.This beautiful powerhouse sounds great powering NUC I7 based servers and endpoints.
• An audition of an RME ADI2 DAC has been underway powered by various LPSUs.
• The NUC I7 server referred to yesterday is now in a maximum power state to test the limits of temperature and upsampling resolution with Hqplayer embedded.
• In the i7-6700k system, the second LT3045 output from the HDPLEX is being used to externally power the Startech USB card.
• The I7-6700k machine will be dual powered once again to determine the benefits in a NUC world.
• In the next month a new fanless case, Intel CPU and motherboard will be ordered as the i7-6700k, and z270 based motherboard are now unobtainable.

Systems are chains of components in various hardware and software configurations with a ridiculously large number of combinations possible. Armed with various hypothesis derived from our listening experience, we can try to test those combinations most likely to yield a new level of SQ.

 

Which is best? I don't know. The good news is that every properly configured NUC based system tested here delivers a step function increase in SQ, so there are no bad systems. Remember that the NUC configurations used today by most of us has evolved over time from a stock NUC in a stock box with stock bios settings running a stock power supply. Even the first listen from that stock NUC with disk based LXQT AL configured as a single box Roonserver sounded much better then any other machine heard before. I have a friend running a Celeron NUC that way today and it beats the sound from any of his expensive CD players. He is delighted!

 

2018 was a banner year for CA innovations: Here is the list from my perspective. I am sure there are more.

• Dual powered motherboards
• JSSG360 DC cables
• JSSG360 USB cables
• NUCs generally
• AL headless ramboot
• LPS1.2 powered NUC endpoints

 

Can I rank the most important of these? Not really, and if I could, what is the point? Most of these innovations are additive and in various combinations we have managed to obtain a fantastic level of musical performance and for that I am grateful. Can we go further?  I don't know, but it wont be for a lack of trying.

 

Larry

 

 

 

[greenleo]

10 hours ago, lmitche said:

Hi Greenleo,

 

There are multiple systems here in various states that change all the time. Here is a snapshot of whats happened in the last few days and weeks.

• A wonderful Uptone Audio JS-2 power supply has come, and gone.This beautiful powerhouse sounds great powering NUC I7 based servers and endpoints.
• An audition of an RME ADI2 DAC has been underway powered by various LPSUs.
• The NUC I7 server referred to yesterday is now in a maximum power state to test the limits of temperature and upsampling resolution with Hqplayer embedded.
• In the i7-6700k system, the second LT3045 output from the HDPLEX is being used to externally power the Startech USB card.
• The I7-6700k machine will be dual powered once again to determine the benefits in a NUC world.
• In the next month a new fanless case, Intel CPU and motherboard will be ordered as the i7-6700k, and z270 based motherboard are now unobtainable.

Systems are chains of components in various hardware and software configurations with a ridiculously large number of combinations possible. Armed with various hypothesis derived from our listening experience, we can try to test those combinations most likely to yield a new level of SQ.

 

Which is best? I don't know. The good news is that every properly configured NUC based system tested here delivers a step function increase in SQ, so there are no bad systems. Remember that the NUC configurations used today by most of us has evolved over time from a stock NUC in a stock box with stock bios settings running a stock power supply. Even the first listen from that stock NUC with disk based LXQT AL configured as a single box Roonserver sounded much better then any other machine heard before. I have a friend running a Celeron NUC that way today and it beats the sound from any of his expensive CD players. He is delighted!

 

2018 was a banner year for CA innovations: Here is the list from my perspective. I am sure there are more.

• Dual powered motherboards
• JSSG360 DC cables
• JSSG360 USB cables
• NUCs generally
• AL headless ramboot
• LPS1.2 powered NUC endpoints

 

Can I rank the most important of these? Not really, and if I could, what is the point? Most of these innovations are additive and in various combinations we have managed to obtain a fantastic level of musical performance and for that I am grateful. Can we go further?  I don't know, but it wont be for a lack of trying.

 

Larry

 

 

 

Thank you very much Larry for the very detailed reply.  Anyway to dual power a NUC?

 

Leo

[lmitche]

9 minutes ago, greenleo said:

Thank you very much Larry for the very detailed reply.  Anyway to dual power a NUC?

 

Leo

Not that I am aware of for a single NUC. We do manage to split tasks in a two NUC configuration, but it's not the same as powering the CPU separately from the motherboard.

[bobfa]

20 hours ago, Bricki said:

I have selected this Max battery setting in the BIOS of my sonicTransporter i5 AL roon server. Previously I have moved it from turbo to non-turbo and heard an increase in SQ.

 

After Larry mentioned low power mode (reducing the TDP) on his server was beneficial I decided to try some settings to achieve the same result.. after selecting Max battery I have heard a quieter sound with a darker background that has better separation and more depth. I am hearing more textures on vocals and more color in piano... I also think it is more relaxed...I won't be changing it back off Max battery that's for sure 😁👍

I am thinking about how to follow in your footsteps a little bit.  I have a ST i7 DSP and I am going to try AL on it.  I will have to setup bridging but there appears to be some direction on how to do that.  

[bobfa]

I am thinking about different server systems for Roon and I am going to try AL on my Sonic Transporter i7 DSP first. I am thinking about buying/building a Xeon test system of some sort.  I am having a hard time picking a direction.  I am sort of holding until I can pick one of the new network switches.  

 

I am not sure where to point my energy on the server/network side of things.  Looking for a direction that will help. 

[bobfa]

So here is an interesting idea.  From: https://silentpc.com/fanless-pcs

 

There is a  Xeon Server chip model:  https://silentpc.com/fanless-pcs/fanless-xeon-server

 

This is just a starting place for an idea.  There are a lot more silent PC system ideas out there. 

[Bricki]

5 hours ago, bobfa said:

I am thinking about how to follow in your footsteps a little bit.  I have a ST i7 DSP and I am going to try AL on it.  I will have to setup bridging but there appears to be some direction on how to do that.  

Yeah definitely try it and tell us what you think...Audiolinux on my sonicTransporter has had a very nice impact...and the BIOS tweaking is still going on 😁

[lmitche]

Interesting server candidate  especially for a hqplayer upsampling machine . . .

 

https://www.anandtech.com/show/13757/asus-launches-mini-pc-proart-pa90-a-tiny-workstation-w-core-i9-quadro

[seeteeyou]

List price of ProArt PA90 is starting at 2,499 quids with i7-9700K and Quadro P2000

 

https://www.overclock3d.net/news/systems/asus_announces_their_mini_pc_proart_pa90_-_workstation_performance_in_a_compact_form_factor/1

 

£1,969.99  ex VAT
£2,363.99  inc VAT

https://www.uk.insight.com/en-gb/productinfo/desktop-pcs/0008555698-00000001

 

£2,362.99  ex VAT
£2,835.59  inc VAT

https://www.uk.insight.com/en-gb/productinfo/desktop-pcs/0008555645-00000001

 

£2,519.99  ex VAT
£3,023.99  inc VAT

https://www.uk.insight.com/en-gb/productinfo/desktop-pcs/0008555740-00000001

[lmitche]

Two followup points, the first a bit off topic:

 

1) An audition of an RME ADI2 DAC has been underway powered by various LPSUs. This DAC was tested powered by the Hdplex 200 watt lt3045 output at 15 volts, a Sigma 11 at 12 volts, and the UpTone JS-2 at 12 volts. The RME needs too much current to be powered by the LPS1.2. An I7 NUC/AL server endpoint pair running Roonserver/Roonbridge or Roonserver/Hqplayer embedded/NAA were used to test both bitperfect and dsd256 upsampled USB playback. The DACs single ended and balanced outputs were tested.

 

Operationally this DAC is really clean with no surprise pops or clicks in any mode or output. The JS-2 yielded the most musical results. Nevertheless the RME DAC never reached the dynamics and open sound of IFI microIDSD BL powered by an lps1.2. The DAC was returned to the vendor.

 

2) In the big i7-6700k server, the second Hdplex LT3045 output was moved from the wifi extender, set at 7.5 volts and used to feed an additional lt3045 chain to externally power the Startech USB 3 gen 2 card. The Hdplex fixed 12 volt power now feeds the wifi extender.  Attached to the Startech card is an external Startech USB hard disk enclosure and 4tb music drive, and a USB nic for network connectivity. Besides AC power, these are the only connections to this server.

 

The addition of external power driving the Startech USB card delivers a nice increase in image density and lowering of the noise floor just as it did in my pre-NUC one box setup. Using one supply to power three peripherals greatly reduces complexity.

 

I am looking forward to using the unused 19 volt Hdplex output to restore split power to the motherboard with an Hdplex 400 watt DC to DC atx power supply.

[ronfint]

Just a quick note to confirm that the nuc8i7hvk performs well with AudioLinux as a roon server upsampling to dsd512. I have it in low power mode, and it is running quietly (in my basement) and it's fairly cool. Core temperatures have been mostly in the high 50s to mid 60s (C). Unlike my MacBook Pro, there have been no dropouts. The roon processing speed when upsampling to dsd512 is 2.6x. On my MacBook Pro it was 1.6.

 

My only problem was learning the linux command to get my network interfaces running. (It's   ip link set interface up. Thanks, Piero.)

[lmitche]

16 minutes ago, ronfint said:

Just a quick note to confirm that the nuc8i7hvk performs well with AudioLinux as a roon server upsampling to dsd512. I have it in low power mode, and it is running quietly (in my basement) and it's fairly cool. Core temperatures have been mostly in the high 50s to mid 60s (C). Unlike my MacBook Pro, there have been no dropouts. The roon processing speed when upsampling to dsd512 is 2.6x. On my MacBook Pro it was 1.6.

 

My only problem was learning the linux command to get my network interfaces running. (It's   ip link set interface up. Thanks, Piero.)

Hi Rofint, this is great to know. Which Hqplayer settings are you using with the nuc8i7hvk? Do you have speedstep turned off in the bios?

[jbparrish]

1 hour ago, lmitche said:

Two followup points, the first a bit off topic:

 

Nevertheless the RME DAC never reached the dynamics and open sound of IFI microIDSD BL powered by an lps1.2.

 

 Hi Larry, 

 

this seems to be a significant complement for the IFI black. I recall you owned one for sometime before your current DAC. It sounds like a really nice alternative for a lower priced DAC that plays well in this environment and can be powered by the also less expensive LPS 1.2. Is that about right?

 

Thanks for all the intense research into this so very interesting subject.

 

John

[ronfint]

Sorry, I wasn't clear. I'm not using HQ Player. I'm only upsampling in roon. I may try HQ Player in the future, but I'd like to sit back and enjoy for a while.

 

In the bios I turned off everything that I don't use — which is almost everything. There was no speedstep setting that I could see. One feature in the bios that I might take advantage of in the future is the ability to undervolt vcore and uncore settings, but I don't want to bother with this unless core temperatures start to rise too much.

[lmitche]

9 minutes ago, ronfint said:

Sorry, I wasn't clear. I'm not using HQ Player. I'm only upsampling in roon. I may try HQ Player in the future, but I'd like to sit back and enjoy for a while.

 

In the bios I turned off everything that I don't use — which is almost everything. There was no speedstep setting that I could see. One feature in the bios that I might take advantage of in the future is the ability to undervolt vcore and uncore settings, but I don't want to bother with this unless core temperatures start to rise too much.

Got it! Many thanks.