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

[lmitche]

16 minutes ago, RickyV said:

 

Very curious to hear about the result of these changes from various people.

Are you @lmitche, @austinpop @bobfa going to make these changes?

I was about to buy Euphony but I’ll wait a bit 

The only change, amongst many, that I hadn't done before was step 6 - adding the asound.conf file. It does indeed increase clarity but was harsh until the real-time priority value was lowered from extreme to standard from the AL menu as recommended by Piero.

[lmitche]

20 minutes ago, seeteeyou said:

Thread gone quiet lately?

CTU,

 

We are busy listening to music on our absolutely awesome NUC/Audiolinux systems. The recent software upgrades are fantastic! SQ exceeds my best expectations.

 

Larry

[lmitche]

22 minutes ago, mourip said:

 

 What would be the command to return the current link speed and duplex values? I looked at Network in the Status menu but those values are not displayed. Perhaps they could be? 

 

Thanks!

ethtool -s eno1 speed 1000 duplex full autoneg on

[lmitche]

51 minutes ago, mourip said:

 

Does that command set new values or return/display the current values? Thanks.

Sorry, I read your question as return to old values.

 

Ethtool eno1

 

shows current values

[lmitche]

2 hours ago, rickca said:

In the USA, I now see it at Newegg for $150.

In a vertical orientation this case puts the cables a long way from the surface of the table or shelf underneath. In the horizontal orientation it's better, but roughly double the height of the X7 or X8  pizza box design.

 

Outside of aethestics, a matter of taste, what are the benefits of this new case? Better cooling maybe? 

[lmitche]

34 minutes ago, seeteeyou said:

 

98.66 quids and 879 RMB respectively, $130 or so

 

https://www.lambda-tek.com/Akasa-A-NUC45-M1B~sh/B43300798

https://item.taobao.com/item.htm?id=592557668564

 

Basically there's a previous generation called Baby Canyon while the processors were quite from each other

 

https://ark.intel.com/content/www/us/en/ark/products/codename/95586/baby-canyon.html

 

7.5W to 15W TDP

https://ark.intel.com/content/www/us/en/ark/products/95442/intel-core-i3-7100u-processor-3m-cache-2-40-ghz.html

 

9.5W to 15W TDP

https://ark.intel.com/content/www/us/en/ark/products/97539/intel-core-i5-7260u-processor-4m-cache-up-to-3-40-ghz.html

 

23W to 28W TDP

https://ark.intel.com/content/www/us/en/ark/products/97541/intel-core-i7-7567u-processor-4m-cache-up-to-4-00-ghz.html

 

Then Akasa started selling their Plato X7 despite such significant gap in TDP, in the end they just added this kinda "friendly reminder" specifically for i7 model(s) even though that's pretty much like BS

 

http://www.akasa.com.tw/download/compatibility/Akasa 7&8th Gen NUC Chassis Compaibility V13.pdf

 

Could Domino's / Papa John's actually take the heat? Not THAT much room inside the box and therefore we could tell what the deal is. IMHO you're a sucker if you're expecting their Plato X8 to be any good for Bean Canyon since all of them are sharing the same 20W to 28W TDP (i.e. throttling should be no surprise?)

 

https://ark.intel.com/content/www/us/en/ark/products/codename/126382/bean-canyon.html

 

BTW, they also "experimented" with Skull Canyon while we're talking about 35W to 45W TDP for an i7-6770HQ

 

https://ark.intel.com/content/www/us/en/ark/products/89187/intel-nuc-kit-nuc6i7kyk.html

https://ark.intel.com/content/www/us/en/ark/products/93341/intel-core-i7-6770hq-processor-6m-cache-up-to-3-50-ghz.html

 

Honeymoon didn't seem to last very long and that NUC ended up with premature death

 

https://www.reddit.com/r/intelnuc/comments/6zq4yo/my_akasa_galactico_case_review_for_intel_skull/

 

What about Turing then? My uneducated guess would be turning the entire case into a gigantic heat sink and see how that might / might not be able to handle something with 20W to 28W TDP.

 

IMHO that could be potentially a logical explanation since their Galactico was more like a joke, though we still couldn't tell if Turing were good enough for Bean Canyon yet

 

https://www.notebookcheck.net/Intel-NUC-Kit-NUC8i7BEH-i7-8559U-Mini-PC-Review.360356.0.html

 

Obviously that's something to be taken into consideration if we're running something like Extreme 2 with AL, otherwise the role of Roon Endpoint etc. shouldn't consume that much power to begin with.

Thankfully the 15 watt Comet Lake U-series is on the way, so we can skip the 28 watt Bean Canyon release and can continue to avoid high current machines. In the meantime the 15 watt Kaby Lake R NUC machines sound terrific and run cool in the Akasa X series cases.

[lmitche]

On 5/6/2019 at 2:40 PM, shahed99 said:

Can anyone recommend which Optane drive to use with my NUC 7i7DNHE? I was looking into this one on amazon and the compatibility tool there says it's not sure if Intel NUC is compatible with that particular Optane. 

Yeah, the 32 GB Optane works in the NUC7i7DNXX series.

[lmitche]

7 hours ago, Ricardo007 said:

What is the interest of bridging if you use audio switch like sotm?

At the risk of another "death by a thousand cuts" episode, one can think of the software bridging of two nics rather like a type of isolating switch but in the server.  The level of isolation depends on  the physical isolation and power source used for the nic cards used to create the bridge(switch).

 

Bridged nics in a server complements the isolation provided by isolated switches like the SOTM device.

[lmitche]

2 hours ago, seeteeyou said:

Holy Smoke, it's one hella setup and look at those DC cables

 

http://dxpwr.com/product-range

https://www.facebook.com/110747222358641/posts/1804350272998319

Wow, CTU is this your system?

[lmitche]

9 hours ago, rettib2001 said:

Hi

 

I'm currently running AL headless in a server (nuc7i7bnh) + endpoint (fanless nuc7pjyh powered by SR4) configuration, upsampling Tidal to dsd512 via roon.

 

The server is connected to an Orbi rbk50 router via wifi at one end and the endpoint is linked to the accompanying Orbi satellite via ethernet (set to 100mbps) at the other. Both Orbi units are powered by their own Lpsu. 

 

I recently started using hqplayerd with Roon and unfortunately for my wallet I prefer what I'm hearing.

 

With my 2 core i7 nuc I can get upto dsd256 using the - 2 filters, I'm surprised it can do that. It sounds more involving and organic (subjectivity alert) than Roon at dsd512.

 

I'd now like to build a more powerful server with a view to using hqplayer at dsd512, in an ideal world without resorting to the - 2s filters. 

My dilemma is that if I aim for a powerful server I'll be straying away from the realm of fanless cases and clean power in favour of brute computational force. 

 

What are peoples thoughts on either:

 

a) Waiting for the next powerful multicore i7 nuc, probable the comet lake gen, sticking that in a fanless case and powering it by my Paul Hynes Sr4 and hoping I can get far enough up the hqplayer filter hierarchy. 

 

b) Building a powerful server using a cpu that has a proven track record of being able to upsample to dsd512 and hoping that the additional noise or general electrical fallout generated by the more 'untamed' computing environment doesn't do too much damage to general audio quality. 

 

I'm torn. 

Do you have a preamp or do you use Roon or hqplayer volume control?

[lmitche]

19 hours ago, rettib2001 said:

 

I understand why you would ask that question. 

 

Funnily enough volume control is something I've spent quite a bit of time 'fiddling' around with lately

 

The dac I use is an LKS 004 which allows for on-board volume control, I initially compared that functionality to my fairly decent passive preamp which has a stepped attenuator. At lower volumes the passive had the edge but as I crept closer to 0db (usually upto about - 20db) I felt the Dac's volume control had the edge. Roon volume control was set to fixed and Hqplayer was set to - 3.

 

My current setup however is: Roon fixed volume, Hqplayer - 3db  and Dac at fixed/0db. The dac goes into a tube buffer with volume control the latter is powered by a Paul Hynes Sr4. It sounds so much more alive than anthing I've used in the past. 

 

Satisfied with this setup, I've put aside any tweaking with volume control and I've turned my attention to gaining access to hqplayer's more demanding parameters hence the question about server options. 

 

 

Before building an uber machine you may want to consider Roon bit perfect playback with underclocked processors using your existing equipment. 

[lmitche]

2 hours ago, rettib2001 said:

 

Thanks for that advice, I played around with cpu frequencies in audiolinux for both the server and the endpoint today. No upsampling, just bit perfect playback via Roon. 

 

I didn't expect the underclocking to make much of a difference but I was pleasently surprised to find out that it did.

 

That being said the Dac chips in my LKS (dual mono es9038pro) just seem to show their full potential when dsd upsampling is used, I wish it wasn't the case as it would save me a fair bit of money. 

 

I don't want to come across as a 'it's dsd512 or nothing' kind of guy. I have a chord Hugo which obviously sounds best when left alone so that the in-house Chord sound can shine through. 

 

If others have similar ESS based Dacs and have arrived at a different conclusion please feel free to tell me that upsampling is useless so that I can give my credit card a rest...

 

Interesting, I have a protoype of a new ESS 9038pro DAC here on tour. I'll crank things up to DSD12 and have a listen.

[lmitche]

40 minutes ago, cool_chris said:

 

It took me 7 years to read it all 😄

I am slow and You are posting so many things a day It was almost impossible to cach up with everything. 

 

Again. I did many things back in 2010 -2015 and tested all the options.

I had to quit because of to much time I burned to keep reinstalling everything after the system was not booting anymore. I had addition to the short latency.

it looks like I was right to go this direction , but hardwere could not keep the owners needs.

After I quit I purchesd totaldac server + twelve DAC and stayed with that till now.

Unfortunate that perfect sound I was getting old times is still on the back of my head . 

I have totaldac top DAC so no need for upsample.

 

After reading all of the topicks here I think it is a good time to get back and 

1. Purchese top server soch as Taiko Extreme 

( I wish this is linux based server not windows that I hate, because w12 server plays well once installed and optimised but it starts to DEGRADE the sound with time.

Reinstalling everything brings the SQ back but it takes a day or two to install and optimise everything.

please correct me if I am wrong.

 

OR

 

2. Build the cost no object server (euphony preferable) that WILL NOT turn down or degrade the SQ with time.

 

Aditionally 

I should add the Sotm swiches ( 2 pieces ) or wait for the Uptone ether regen to come.

Get the Taiko Audio SGM Extreme. It is the best of the best. Did you know the Totaldac D1-twelve mk2 is one of the DACs used when testing the SQ of the Extreme?

 

While you are at it, checkout the Taiko Audio Daiza anti-vibration platforms. I have three here and the Panzerholtz construction has an immediate impact enriching SQ. My DAC, Amp and NUC/ISO Regen ride on these platforms. 

 

TheTaiko Audio guys are the top players in computer audio.

[lmitche]

1 hour ago, cool_chris said:

I am just unsure if stripped Windows will work as good as the new installed device years later.

It would be very surprising if Windows automatic updates are not disabled on the Extreme. Also, the Taiko people will bend over backwards to keep your system up to their latest release. Check for yourself, but I wouldn't have any worries about Windows or Taiko service ethic.

[lmitche]

5 hours ago, austinpop said:

 

Thanks for reminding me about SSC (spread spectrum clock). I'll have to go back into the BIOS of my H370-ITX/ac and see if it's exposed, and enabled/disabled. As I recall now, @lmitche had found it doesn't always help to disable it everywhere - for example: PCIe.

 

If you wouldn't mind - please post a pointer to where in the Z390 BIOS the SSC controls are.

My Asus server motherboard has three SSC controls and I've disabled all of them.

[lmitche]

18 minutes ago, Chopin75 said:

I have z370 board for AMD with Ryzen CPU, not intel., so it means as mentioned above, that Optane won't run as cache, and some of its features would be disabled, correct ? But it can still as  euphony OS ? In other would benefits still apply if one uses non-intel based motherboard/CPU ?

 

Yes,  it is highly likely the benefits of running the Optane as SSD will work in an AMD board. I've not tested it, so can't guarantee anything.

[lmitche]

20 minutes ago, Chopin75 said:

You have a web link to these speakers ? 

https://www.focal.com/en/home-audio/high-fidelity-speakers

[lmitche]

On 6/17/2019 at 3:42 PM, Neal.Audio said:

FYI, I've posted a NUC that Romaz built for sale in the classifieds if anyone is interested! It takes the best of his lessons in this thread and puts them in a nice Akasa case.

 

NUC7i7DNBE NUC Akasa X7D

 

17 hours ago, austinpop said:

 

It's not a case of moving on. The NUC7i7DNBE remains an outstanding endpoint option, so your moniker of "near end-game" is very apt.

 

However, we only evolve when we absorb the lessons learned.

 

I now see how it's possible to improve much further than the i7DNBE NUC, but it's not easy, and it's not for everyone. To understand why, let's talk about the lessons learned:

1. PSU quality is paramount. This appears to be the one constant in this journey! But while there are excellent PSU options for single rail, low current (<3A), the options dwindle for multi-rail high current, especially ATX.
2. Assuming a high-quality, high-current PSU, the more powerful the CPU (in terms of #cores, peak frequency, cache size), the more space, dynamics and body the music has.
3. PSU quality is the limiting factor, and low PSU quality yields harshness as CPU power rises.

Right now, the NUC7i7DNBE is a widely achievable high-water mark in terms of SQ, as its 15W TDP means it can easily be powered with something like an SR-4 and still yield outstandingly clean SQ, with very respectable dynamics. Certainly head and shoulders better than we have experienced with the underpowered Atom/Celeron/Pentium based commercial streamers of past and present.

 

For people looking past NUCs, achieving #2 is the big challenge. I can clearly hear how much bigger and more dynamic my i7-8700T server sounds when directly connected to my DAC (through the tX-USBultra, of course). And this is not even with the i7-8700K or i9-9900 CPUs @romaz and others have tried. The challenge, of course, is that my HDPlex 400W ATX Linear PSU, while very good, is not in the same league as an SR7 and cannot completely tame the harshness. So I am currently in this state of flux, where I can choose more dynamics, but pay for it with more harshness. Yes, as I've explained in earlier posts, I can tune this somewhat using the max frequency control in Euphony.

 

Now, if you have the luxury of multiple SR7 DR rails like Roy does, where he is using a 19V/6A DR rail to power an HDPlex DC-ATX converter for ATX, and another SR7 DR 12V/6A rail for EPS, you can truly reap the benefits of greater dynamics, with no harshness. This is the context of Roy's comment of his NUC being "retired." If I had those rails, I would too! In my case, this is achievable, as I do have a 3-rail SR7DR on order, and based on my queue position I should hopefully receive it this summer.

 

Can this be improved even further? It remains to be tried, but I am very hopeful it can be. Here are some thoughts:

• If a PSU vendor like Paul Hynes (unlikely), Sean Jacobs, or Mattijs at Farad were to deliver a truly audiophile DC-ATX converter, then that would be a game changer
• Extrapolating from the improved dynamics observed going from i7-8650U (i7DNBE NUC) to i7-8700T to i7-8700K to i9-9900, would the Ryzen 3900x (12 cores, 4.6GHz peak, 64MB L3 cache) be the next logical step?
• Of course, each step, raises the TDP, and the demands on the PSU.
• ideally we need dual PCie card slots, without risers, one for Ethernet (JCAT or the upcoming SOtM), and one for USB (JCAT or SOtM). Each of these need PSUs too!
• It's possible that with an SR7 powering the ATX input, then "bus" power may in fact be cleaner for these PCIe cards than external PSUs. This needs to be validated. I know Roy is playing with this.

One final thought. If you look at commercial systems based on these CPUs like the i9-9900K, you'll see they're provisioned with PSUs often in excess of 1000W. Our custom builds work with much lower power PSUs (even a 19V/10 SR7 rail for ATX, and a 12V/6A rail for EPS adds up to less than 300W) because we are running these systems mostly idle. Roon, Stylus etc. without upsampling, room correction, or other DSP, barely run at 1-2% utilization. This allows us to use these lower powered PSUs with impunity.

 

But once you crank this server up to high utilizations with DSP, all bets are off. This is also the reason why commercial music servers built to this principle are going to a while coming. A manufacturer cannot just assume a customer will never run high CPU workloads. Thus, they have to build for the worst case. This explains the extraordinary lengths the Innuos, the Taiko and the Pink Faun servers have to go to for power supplies, and why they are so expensive.

 

That is why my advice to people who ask me if it's time to move past the NUCs is to focus first on improving the PSU on the NUC and the rest of the chain. Only if you have multi-rail, high current, and high-quality PSUs should you venture further, or otherwise be prepared to ride the "dynamics vs. harshness" wave!

 

 

OK, so the hoarder of SR7 rails sends messengers to tell us our NUC machines, that he first recommended, are now inferior as they can't reproduce the "dynamics" that a single 8 core machine can when powered by the unobtainable SR7s.

 

LOL!!!

 

Really? What is the purpose of posting this? None of us can acquire these power supplies. I am usually grateful for Roy's posts but posting by proxy about unobtainable devices feels over the top to me.

 

For the rest of us, when tuned properly the SQ of our NUC/AL machines is fantastic without  any harshness or lack of dynamics.  Indeed AL just keeps getting better with time. Power supplies and added regulation, isolation devices and cabling choices continue to add value.

 

I know dozens of people that are thrilled with the SQ of their Celeron, Pentium and I7 NUC machines.  For them, I says thanks to Roy for introducing us to the NUCs last August.

 

Sadly it appears our favorite audio NUCs are being discontinued, and they are likely to be unobtainable shortly as well.

 

I look forward to hearing the new Ice Lake NUCs. In the meantime with these supply issues this conversation seems rather academic.

[lmitche]

On 6/20/2019 at 1:51 PM, austinpop said:

 

It's not a case of moving on. The NUC7i7DNBE remains an outstanding endpoint option, so your moniker of "near end-game" is very apt.

 

However, we only evolve when we absorb the lessons learned.

 

I now see how it's possible to improve much further than the i7DNBE NUC, but it's not easy, and it's not for everyone. To understand why, let's talk about the lessons learned:

1. PSU quality is paramount. This appears to be the one constant in this journey! But while there are excellent PSU options for single rail, low current (<3A), the options dwindle for multi-rail high current, especially ATX.
2. Assuming a high-quality, high-current PSU, the more powerful the CPU (in terms of #cores, peak frequency, cache size), the more space, dynamics and body the music has.
3. PSU quality is the limiting factor, and low PSU quality yields harshness as CPU power rises.

Right now, the NUC7i7DNBE is a widely achievable high-water mark in terms of SQ, as its 15W TDP means it can easily be powered with something like an SR-4 and still yield outstandingly clean SQ, with very respectable dynamics. Certainly head and shoulders better than we have experienced with the underpowered Atom/Celeron/Pentium based commercial streamers of past and present.

 

For people looking past NUCs, achieving #2 is the big challenge. I can clearly hear how much bigger and more dynamic my i7-8700T server sounds when directly connected to my DAC (through the tX-USBultra, of course). And this is not even with the i7-8700K or i9-9900 CPUs @romaz and others have tried. The challenge, of course, is that my HDPlex 400W ATX Linear PSU, while very good, is not in the same league as an SR7 and cannot completely tame the harshness. So I am currently in this state of flux, where I can choose more dynamics, but pay for it with more harshness. Yes, as I've explained in earlier posts, I can tune this somewhat using the max frequency control in Euphony.

 

Now, if you have the luxury of multiple SR7 DR rails like Roy does, where he is using a 19V/6A DR rail to power an HDPlex DC-ATX converter for ATX, and another SR7 DR 12V/6A rail for EPS, you can truly reap the benefits of greater dynamics, with no harshness. This is the context of Roy's comment of his NUC being "retired." If I had those rails, I would too! In my case, this is achievable, as I do have a 3-rail SR7DR on order, and based on my queue position I should hopefully receive it this summer.

 

Can this be improved even further? It remains to be tried, but I am very hopeful it can be. Here are some thoughts:

• If a PSU vendor like Paul Hynes (unlikely), Sean Jacobs, or Mattijs at Farad were to deliver a truly audiophile DC-ATX converter, then that would be a game changer
• Extrapolating from the improved dynamics observed going from i7-8650U (i7DNBE NUC) to i7-8700T to i7-8700K to i9-9900, would the Ryzen 3900x (12 cores, 4.6GHz peak, 64MB L3 cache) be the next logical step?
• Of course, each step, raises the TDP, and the demands on the PSU.
• ideally we need dual PCie card slots, without risers, one for Ethernet (JCAT or the upcoming SOtM), and one for USB (JCAT or SOtM). Each of these need PSUs too!
• It's possible that with an SR7 powering the ATX input, then "bus" power may in fact be cleaner for these PCIe cards than external PSUs. This needs to be validated. I know Roy is playing with this.

One final thought. If you look at commercial systems based on these CPUs like the i9-9900K, you'll see they're provisioned with PSUs often in excess of 1000W. Our custom builds work with much lower power PSUs (even a 19V/10 SR7 rail for ATX, and a 12V/6A rail for EPS adds up to less than 300W) because we are running these systems mostly idle. Roon, Stylus etc. without upsampling, room correction, or other DSP, barely run at 1-2% utilization. This allows us to use these lower powered PSUs with impunity.

 

But once you crank this server up to high utilizations with DSP, all bets are off. This is also the reason why commercial music servers built to this principle are going to a while coming. A manufacturer cannot just assume a customer will never run high CPU workloads. Thus, they have to build for the worst case. This explains the extraordinary lengths the Innuos, the Taiko and the Pink Faun servers have to go to for power supplies, and why they are so expensive.

 

That is why my advice to people who ask me if it's time to move past the NUCs is to focus first on improving the PSU on the NUC and the rest of the chain. Only if you have multi-rail, high current, and high-quality PSUs should you venture further, or otherwise be prepared to ride the "dynamics vs. harshness" wave!

 

 

Rajiv,

 

I was just lamenting that our world seems to be splitting into SR7 haves and have-nots. Perhaps we should start a new thread for the have-nots. Too bad you did not see the humour in my expression of this lament.

 

Larry

 

The original promoter of "one-box" solutions.

[lmitche]

35 minutes ago, luisma said:

Honestly Larry I get that from the  comments the SR7 "seems" to be the pinnacle of the SQ (which not necessarily means it is) but that doesn't mean there is a split between haves and have nots, Rajiv and Romaz are establishing reference systems which most of us we cannot try (now I kind of see your point) but there are things in between which are completely within reach, and I personally I'm perfectly happy with it, like for example AL, it has made a big difference on my system. 

My point is Rajiv have been part of the community and very valuable like yourself and your comment although kind of understandable don't really apply, it started like a joke but then you went into assumptions

My apologies for trying even to provide opinions between you two but I appreciate you both contribution. 

Luisma,

 

The whole point of the NUC direction, and the reason I have been promoting it, is recognition that as a low current device a NUC can use an obtainable, high quality, low current power supplies to get SQ that is superior to prior one-box solutions.  Supplies like the LPS1.2 and Hdplex 200 are examples. On top of that with galvanic isolation on the network and USB connections ground loops are gone. This is easily accomplished with a 5ghz WIFI card and an ISO Regen with any NUC.

 

Maybe someday we can find an obtainable and affordable high current power supply as good as the SR7. In the meantime, I will be recommending NUCs, if they are available.

 

As far as I know, the only DIY one-box solutions rumored to beat the NUC approach are powered by SR7s.

 

Larry

[lmitche]

1 hour ago, seeteeyou said:

 

 

https://audiophilestyle.com/forums/topic/55873-shootout-at-the-linux-corral-audiolinux-vs-euphony/page/8/?tab=comments#comment-960794

 

Hi CTU,

 

Funny, when I am asked to quote/build a custom machine,  an Intel 8 core processor and a Hdplex ATX 400 watt power supply is the recommended solution. Running AL of course.

 

I did once build a Euphony machine but it didn't sound as good as AL. Maybe I did something wrong.

 

Thanks,

 

Larry

 

 

[lmitche]

On 6/22/2019 at 5:13 PM, austinpop said:

You may consider insulting me by calling me Roy's "messenger" and "proxy" to be humor, but I don't. Not everybody earns my respect, but you had. Now you've lost it. 

Hi Rajiv,

 

When you convey a message about another persons SQ observations you are being a messenger. You did that in your original post and there is nothing wrong with that. Indeed you appear to share both my support of NUCs as a high quality music source, and my lament about the lack of SR7 supply.

 

Given this, I can't understand how you found my message insulting especially when I sent a clear signal that it was meant in humor.

 

Best of luck on your audiophile, online moderation and hifi reviewer endeavors.

 

See ya,

 

Larry

 

 

[lmitche]

4 hours ago, Blackmorec said:

Hi Imitche.

I can understand your frustration on reading that according to one opinion leader, the very best available sound is now coming from power hungry 8 core machines, whose ideal power supplies are no longer available. But getting riled at Austinpop for neatly summarising the current situation with 8 core and NUCs is just ‘shooting the messenger’ . 

I’ve read and enjoyed all these ‘novel way to massively improve the Sound Quality....’ posts since day one; or at least since Austinpop took over their administration. But it was always clear that NUCs weren’t designed as audio boxes, rather as consumer electronics that fortuitously did audio rather well. Fortunately the person who discovered this happenstance was gracious enough to write up their discovery and share it via this Forum. Further, no one really understood what exactly made these NUCs sound so good, which was rather ominous for the future, given the rate at which computer hardware is superseded. Essentially with NUCs, audiophiles are at the mercy of pure chance that the audio capabilities of these units are preserved and indeed bettered through future generations. 

In the case of the Innovation Adoption Lifecycle, its very clear that NUCs are used exclusively by Innovators, followed later by Early Adoptors.  As there is absolutely no audio marketing development by their manufacturers, that’s as far as adoption will go. There will never be an early majority, a late majority or a group of laggards, so NUCs will never gain a true foothold in the audio market and will only appeal to a few tech-savvy technicians. 

So when the next development is discovered and adopted by a few technology pioneers, don’t expect them to give much of a fig whether their discoveries are easily adopted or not. The only thing they care about is moving forward. Discovery and progress are the only names of their game. 

Blackmorec,

 

Many thanks for your thoughtful post. There is a lot here, and it deserves a detailed response. I will send you a PM with a link when I have found an appropriate place to respond.

 

Larry

[lmitche]

1 hour ago, Superdad said:

The technique is hiding in plain site in the data sheet.   John has wired LT3045s both ways and measured about half the output impedance when it is done right.  For some reason Alexy at ldovr does not believe it. Oh well...

Hi Alex,

 

FYI, the board pictured above does not look like one of Alexey's Ldovr.com brand boards. It is likely a Chinese knockoff of his design.

 

Larry

 

 

[lmitche]

2 hours ago, Dutch said:

It’s got nothing to do with the power applied externally which you mentioned so that’s why I explained.

There are power controls in the NUC bios with sliders to set amperage limits. They work too.