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

[asdf1000]

4 minutes ago, lmitche said:

What power supplies are you using for your two NUCs?

 

LPS-1.2 was ok for the NUC (and Up Board Gateway) as NAA... still a tad brighter than the Rendu + LPS-1.2 for my ears (over longer listening sessions) 

 

I used a battery to power the NUC running HQPe + RoonServer... No difference to my ears, to using it's stock SMPS though when it's running as a networked server (through NetGear GS108 with JS's grounding trick).

 

 

[asdf1000]

9 minutes ago, lmitche said:

Did you try bitperfect playback with Roonbridge? You have an ifi DAC right?

 

Yeh, tried bitperfect but much prefer HQPe... iFi Pro iDSD.

 

Same difference anyway - at the server end, no difference (to my ears) between AL and stock Debian and no difference between bitperfect (low CPU loading) and HQPe up-sampling (very high CPU loading).

 

At the NAA end, a slight difference but prefer the Rendu over longer listening sessions (slightly less fatiguing).

 

[asdf1000]

@Superdad now that you have the LarryNUC, can we get eye pattern measurements of the LarryNUC/AL compared with your microRendu?

 

If you haven't already (quietly in the background) I'm sure you and John S would be keen to see what may be happening with these observations... 

 

 

 

[asdf1000]

15 minutes ago, austinpop said:

 

That is most interesting. But hey, no one can (or should) argue with what you hear.

 

Kudos for at least trying it!

 

Not everyone thinks up-sampling to very high sample rates with something like HQPlayer or Audirvana, necessarily results in 'better' sound quality in their system (system dependant).. and that's more than OK of course.. each to their own.

 

but specifically for those that do... something like HQPlayer blows the lid off the theory that you need the networked server to have the absolutely lowest CPU loading for 'better' SQ...

 

The Roon COO said something similar - the context here is people using 'under powered' servers. My ears agree with him too:

 

"Show me any device that claims that good SQ requires low CPU performance, and I’ll counter that with exhibit #2: HQ Player"

 

https://community.roonlabs.com/t/nucleus-or-antipodes-cx/45916/55

 

I think many of us will agree we want the endpoint directly connected to the DAC to have ultra low CPU loading.. but my ears preferring the Rendu seem to think the hardware itself also makes a difference..

 

My ears think there needs to be network isolation between the server and endpoint and this is Roon's recommended best practise too. I think a lot of us here practise that already (for a long time - nothing ground breaking here for most of us). I expect the John Swenson EtherREGEN with even better network isolation may help here too..

 

[asdf1000]

2 minutes ago, austinpop said:

 

I was just referring to NUCs. Clearly you prefer your microRendu to the NUC running AL/RAM. That's cool.

 

I've now compared both. I know Larry has said he's never had an ultraRendu in his system (although he prefers his ISO REGEN + LPS-1.2 on the output of his NUC..).

 

How about you - have you compared with an ultraRendu?

 

[asdf1000]

6 minutes ago, austinpop said:

By this chain, I think I know how my comparison of NUC with the UR would go!

 

Not so sure about this since the ultraRendu and SMS-200 Ultra did sound different to my ears in my system... both compared at the same time over a couple of weeks..

 

I preferred the former but many prefer the latter but I don’t think you can guess how the ultraRendu would sound based on the SMS-200.. not at all.

 

Highly recommend you try an ultraRendu + LPS1-.2 if you get the chance..

 

[asdf1000]

4 minutes ago, austinpop said:

 

I think you misunderstood. I compared an actual UltraRendu vs. the sMS-200ultra. 

 

Hang on ... (refers to index) .. here is the report: https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?do=findComment&comment=742958

 

 

Noted, so that’s what you meant by indirectly - ie not comparing at the same time with NUC/AL. Relying on memory isn’t ideal but better than nothing obviously.

 

 

[asdf1000]

13 minutes ago, austinpop said:

Based on the comparison history above, do I feel motivated to?

 

No.

 

Hehe noted. The same feeling I struggled with, when buying and setting up AL and moving NUC from server to endpoint.

 

Quite the pain in the a$$ indeed but good to say I tried it and move on.

 

[asdf1000]

17 minutes ago, lmitche said:

How was the SQ jump from your microidsd to idsd pro? Did you document that anywhere?

 

No, never wrote about it anywhere. It was the jump in SQ I expected for the price difference for sure. 

 

It's one of the least sensitive USB inputs (sensitivity to upstream 'stuff') I've come across, comparing to some Chord DACs, DirectStream DAC and a few others... Before that gets mis-interpreted, a couple very important notes:

 

Important note #1: That's purely a comment about USB input sensitivity - not a comment on overall SQ/performance comparisons to those DACs...

 

Important note #2: Least sensitive does not mean immune to upstream changes. I think we're all still waiting for that immune DAC. Pro iDSD is very good in that area though.

 

 

[asdf1000]

25 minutes ago, lmitche said:

Thanks, you saw where I was going.

 

Did you happen to compare the microidsd bl vs. the idsd pro with the NUC/Al combo?

 

No. I could say since Pro iDSD is better than micro iDSD BL (to my ears) with Rendu source, and since Rendu is better than NUC/AL, I could connect the dots...

 

But I never had everything at the same time to compare, so my comments wouldn't be valid...

[asdf1000]

15 minutes ago, MrUnderhill said:

I have much simplified my USB chain, resulting in simplification AND better sound quality - result!

 

I agree... I wrote in the 'other' less moderated and similar titled thread:

 

"I'm of the belief that there is no free lunch with anything in this hobby - you can't keep adding components (and their PSU's...) to the chain without some trade-off... You're adding more noise/RF sources.

 

Of course, part of the fun of this hobby of ours is finding that perfect balance between lowest box count and sound quality that makes us happy. Finding that optimal ratio that satisfies our ear/brain system..."

 

 

[asdf1000]

20 minutes ago, lmitche said:

Sorry, I just saw that you documented your use of the Swenson ground shunt technique above. 

 

All good.

[asdf1000]

21 minutes ago, RickyV said:

Hi em2016,

I am in process of choosing a dual nuc setup. When choosing your nuc did you consider the nuc7i7beh? The nuc7i7beh is 100 bugs cheaper, better CPU but a tdp 28w. 

The NUC7i7DNHE has tdp of 15w and the tdp configurable. Less tdp  power is better?

Also I am not sure if I should get the same nuc as a streamer.

 

Hi RickyV

 

No I didn't consider another NUC.

 

The NUC7i7DNHE can be easily put into an Akasa fanless case so that was an easy decision for me.

 

I'm not sure how easy/hard it is to put the NUC7i7BEH in a fanless case - never looked into it.

 

For the endpoint / NAA end, USB cable connected to the DAC, the below is recommended by Jussi of HQPlayer and comes already in a fanless case. It is ultra low in power consumption... Mine is happily powered by 5Vdc 900mA...

 

https://up-shop.org/home/81-up-gws01w4g-memory32g-emmc-boardwo-vesa-plate.html

 

You can run whatever you want on it - DietPi, Volumio, AudioLinux, HQPlayer bootable images, whatever

 

[asdf1000]

32 minutes ago, lmitche said:

Sorry, I just saw that you documented your use of the Swenson ground shunt technique above. 

 

Actually no manual shunting was needed by me...

 

The NetGear GS108 is happily powered by 7.5Vdc.. so this comes in handy (it's already internally grounded):

 

https://uptoneaudio.com/products/uptone-branded-7-5v-4-8a-36w-smps-with-internal-ground-shunt

 

[asdf1000]

35 minutes ago, auricgoldfinger said:

 

Adrian told me he has shipped a few in the past week or two.  There may also be special introductory price for early adopters, but you'd have to confirm with him directly.  Here is the price list:  http://thelinearsolution.com/assets/files/2018%20TLS%20Price%20List-12-1.pdf

 

 

Do you know which NUC board he's used?

 

And apart from the mention of clocks, any idea if switching regulators have been replaced with ultra low noise linear regulators? For example.

 

Or apart from the mentioned clocks, it's just the stock NUC board?

[asdf1000]

6 minutes ago, austinpop said:

So is the DNKE powerful enough to be the server? I can't say.

 

 

[asdf1000]

2 minutes ago, austinpop said:

Obviously any of these machines can run basic Roon server functionality without even breaking a sweat.

 

Just in case others were wondering, the NUC7i7DN** can do quite a bit more than running Roon Server... including HQPlayer DSD512 up-sampling for quite a number of -2s filters (all but xtr-2s in my testing).

 

 

[asdf1000]

18 minutes ago, austinpop said:

 

No one is stating otherwise. You are missing my point.

 

I get your point.

 

In the context of the NUC7i7D*** specifically, "Obviously any of these machines can run basic Roon server functionality without even breaking a sweat" doesn't give the full picture.

 

A few people have messaged me surprised that this NUC can do what I've said/shown, so others may not know.

 

This is to help others that may have missed this. I was surprised too...

 

For information only...

[asdf1000]

On 11/17/2018 at 4:56 PM, Johnseye said:

 

Thanks! I didn't realize. My NUC has no hard drive. I wonder if it will boot from USB. 

 

Hi John

 

Did you end up comparing the NAA ramroot bootable image @Miska made, against Audiolinux ramroot NAA?

 

Any differences to your ears?

 

No sound quality differences to my ears...

 

But Jussi's image wins on performance for me - when playing new tracks there are pops just at the beginning of tracks with Audiolinux (and nearly every Linux distro I've had running NAA before).

 

With Jussi's bootable image, it's dead silence changing tracks, so Jussi's image easily wins for me. These pops between tracks (between, never during) are way more noticeable on headphones (which I use a lot) than speakers but they are still noticeable on speakers, again between tracks (not during for me).

 

Unexpected pops on headphones are never fun. Jussi's bootable image is flawless in this regard (as one would expect since he has full control of the bootable OS). And of course Jussi's bootable image comes free versus Audiolinux cost (assumption is someone already has a HQP license of course).

 

Interested to hear your impressions about sound quality though.

 

[asdf1000]

41 minutes ago, Johnseye said:

 

Yes, I tried both.  If there is any difference it's so minimal that bias comes into question.  I'd say both are equally as good for an HQPlayer NAA.  If you need a Roon Bridge or other as NAA, obviously Jussi's OS won't work.

 

 

 

This is very interesting because up until last weekend I had not experienced any pops, except when I swap between DSD and PCM.  Now for whatever reason I'm hearing a pop between every track.  Because this wasn't happening before, I question whether it's a software setting or other change I made.  I will put the HQPlayer NAA back on the NUC and see what happens.  I was also using an iso Regen over the weekend.

 

With my Audio Alchemy DAC I used to get these pops all the time, and some very loud.  It actually blew my B&W tweeters.  It was one of the reasons I bought a new DAC.  When I got the Holo Spring they stopped, until now.  They are very small and quiet pops now however.

 

This is what Peter Madnick described as the cause of the pop.  This is different than what we're talking about, but I question why it's happening, how it's impacted by the hardware and if any mute relay is involved.

 

"When you power a product on and off (ours or anyone else’s) the power supplies must charge up, or discharge, as the case may be.  The raw power supply voltages run through regulators to stabilize them at some known number (+ and – 18 Volts, in our case).  Total capacitance and current draw varies amongst the circuits and cannot be guaranteed to stay at the exact same voltage during the charge and discharge cycles.  An imbalance between the + and – can cause noise.  Again, all products are subject to this phenomenon.  The volume control is essentially the “valve” which controls a preamp’s output level.  If it is all the way up then any instantaneous noise which could come out before the MUTE relay closes can be heard as a POP. "

 

 

Cheers John. I can eliminate the hardware as a variable because the same hardware running Jussi’s image is dead silent. I get the same differences with other hardware too.

 

And in my post I emphasised this is a quiet couple of pops between tracks, as opposed to the classic PCM to DSD transition loud pops, which was fortunately solved a long while back now. 

 

You can try every other Linux distro running NAA including Audiolinux and they will be there between tracks. Easier to hear on headphones than speakers, as I mentioned.

 

Except Jussi’s bootable image is flawless in this regard (and free..) so it’s an easy decision for me.

 

Thanks for sharing the SQ observations, that Audiolinux ramroot is no different - I noted the same.

 

[asdf1000]

4 minutes ago, Johnseye said:

 

How have you solved the PCM to DSD transition pops?

 

Hi John

 

With Roon and HQPlayer and the hardware I’ve had, I just haven’t had that issue in over 2 years.

 

It’s not something I personally solved myself of course - it seems to have been solved by all software and gear I’ve had and the problem went away for me (fortunately).

 

So you’ve had the transitions pops all this time and you now also hear the quieter couple of pops between tracks too, even if staying at DSD constantly (no PCM transition)?

 

 

[asdf1000]

21 minutes ago, Johnseye said:

 

All are relatively quiet, nothing loud, but yes IIRC.  I need to futz with some things this weekend to sort stuff out.  Will have definitive answers then.

 

Assuming everything stays in DSD, it's likely the quieter couple of pops between tracks has always been there with Audiolinux but maybe you never noticed it... as mentioned a few times, it's nowhere near as loud as the classic PCM to DSD transition pops. Easier for me to hear with headphones but they are there with speakers too with Audiolinux.

 

When you do your testing you may find these quieter pops with Audiolinux (and everything else) disappear with Jussi's bootable image - that's what I found.

 

I'm also at DSD512 for everything - it's possible it's less of an issue at lower sample rates with Audiolinux (and others) but I haven't bothered testing because with Jussi's bootable image it's 100% pop free at DSD512 between tracks (for me).

 

 

[asdf1000]

1 minute ago, elan120 said:

This is interesting, as I have not experienced any pops with either NAA images, and this is running DSD512 (non-2S), dead quiet between tracks with headphones.  Wonder what the difference is.

 

When you say either NAA images, do you mean Jussi's bootable?  And also with Audiolinux?

 

I've heard it with NUC running Audiolinux NAA, Sonore Rendu running NAA, Jussi's recommended UpBoard Gateway running DietPi and Auduilinux in RAM etc etc. Lots of different hardware running NAA have the same small pops between tracks for me - with different DACs too.

 

I have to re-emphasise these are quiet pops between tracks - nowhere near as loud and as the old classic and sometimes scary PCM-to-DSD transition. I live with these small pops between tracks fine with my Rendu because it's not a big deal and the Rendu wins out on overall SQ for me.

 

But with Jussi's bootable image running on NUC, UpBoard Gateway etc - flawless.

 

That's why I made a request for @Miska to one create a bootable image for the Rendu's - it's a lot of work though I guess.

 

[asdf1000]

4 minutes ago, elan120 said:

Yes, this is correct.  I ran Jussi's ramroot image first, was happy with the result, and changed to AL HQP NAA later.  Either case, no pops in my setup upsample all music files to DSD512 with non-2s filters with headphones.  The only pops I got is switching between PCM and DSD, but this is not the image issue anyway. 

 

Noted and thanks. Do you hear any SQ difference between Jussi's free ramroot image and AL ramroot HQP NAA?

 

Or no difference of significance?

 

[asdf1000]

3 hours ago, kyrill said:

because of Audiolinux I was worried about the pops remark

 

I've mentioned I think 5 times already but will do one last time because I can see it getting blown way out of proportion.

 

My pops comment is not relating to pops during music playback at all and nothing to do with the classic loud pops of PCM/DSD transitions. I fortunately haven't had these pops for over 2 years now (touch wood it stays that way...).

 

My pops are specific to NAA and starting a new track and it's two small pops - very quiet. I imagine most people wouldn't hear them, especially on speakers, they are that quiet. Even with headphones, if my HQP volume is at -3dB and I'm using a separate headphone amp it is hard to hear. It becomes more obvious when using HQP's digital volume control for me. It's not just specific to Audiolinux for me - as I already mentioned above a few times. It happens with my Rendu and DietPi running on NAA on NUC and UpBoard Gateway and other hardware. And other distro's on different hardware too. DietPi NAA on an Allo USBridge is another example.

The pops are there with my Rendu but it's not enough to annoy me much at all - just to give an idea of how annoying this thing is to me, in the big picture.

 

There is an Audiolinux love fest here so I want to make clear my own very minor issues observed are not at all specific to Audiolinux and not specific to the NUC even.

 

For me, Jussi's bootable NAA image is free from all pops completely and SQ is no different to Audiolinux in RAM and is free and is without these annoying vert minor pops.

 

I hope that squashes pop gate.