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

[lmitche]

1 minute ago, AmusedToD said:

 

Well, I think you should listen to some true high end DACs that will smash any DSD upsampling voodoo with simple redbook playback. In fact - there is no comparison at all. As a matter of fact, the Ifi DAC will look and sound like a toy next to those machines, but we are talking very big bucks. 

 

My computer (i7 iMac) can also upsample to DSD512, and while not bad, it’s not necessarily better than regular non upsampled PCM, it’s just different in presentation (and often “softer” than PCM, lacking in energy and slam). Not my cup of tea.

So what is your cup of tea?

[AmusedToD]

42 minutes ago, lmitche said:

So what is your cup of tea?

 

Non upsampled PCM all the way. DSD files sound great too, but I don’t upsample PCM to DSD. As I said, I tried it, and while it sounds refined, it somehow lacks energy, so I ultimately stopped doing that.

[afrancois]

2 hours ago, AmusedToD said:

 

Well, I think you should listen to some true high end DACs that will smash any DSD upsampling voodoo with simple redbook playback. In fact - there is no comparison at all. As a matter of fact, the Ifi DAC will look and sound like a toy next to those machines, but we are talking very big bucks. 

 

My computer (i7 iMac) can also upsample to DSD512, and while not bad, it’s not necessarily better than regular non upsampled PCM, it’s just different in presentation (and often “softer” than PCM, lacking in energy and slam). Not my cup of tea.

I agree. If a DAC sounds better playing DSD, it is because PCM playback is badly implemented on that DAC. Sadly most DAC designers have been concentrating on DSD, because it is hyped as the next big thing. Of course people may prefer DSD, the same way others prefer vinyl because the way it sounds.

[ElviaCaprice]

I still think regardless of which you prefer (I definitely will take good PCM over DSD any day), HQP upsampled and filtered from the mobo or bit perfect mobo, the sCLK-EX modified mobo will benefit both.  And probably the upsampled mobo more so. 

Why?

Because mobo's are not audiophile constructed components, they are inherently noisy.  When upsampling/filtering from the mobo, you are creating even more noise with the needed demands of more power into an inherently noisy component, the mobo.   But by replacing the clocks on the mobo with clean powered clocks such as the sCLK-EX you are now making the mobo a much better audiophile component.  Thus I think the upsampling/filtering servers will benefit more so than the low power bit perfect servers.

 

As far as the DAC's that do the upsampling/filtering within themselves.  These are audiophile components that are built to handle this function and so will do a much better job than a mobo at not damaging the audio stream.

 

As far as pricing in either server method, I don't see how you can come up with the need for 5 times the price as a necessity?  Also as far as Chord DAC's go, you have to be mindful of the benefits of it's built in low impedant preamp/amp that are a SQ benefit to the direct driving of high efficient speakers. 

Also, my 2Qute only cost $900.  Hardly a huge runup on cost over a ifi Micro.  And there are other higher cost DAC's being used for server upsampling/filtering systems.  Also it is costing less for a low powered mobo.  DAC choice is a personal choice.  I hardly see the cost difference between upsampling/filtering mobo system vs. low power bit perfect mobo system being an argument point of preference for one over the other, and definitely not the more expensive mobo/cpu over the lower priced embedded mobo.

 

But, as Larry said, It all comes down to how it sounds to you. 

[lmitche]

I certainly don't want to get into another tedious dsd vs. pcm debate. My post was about where to do upsampling, not about pcm vs. dsd upsampling. Plenty of people upsample pcm with Hqplayer and Audivana.

 

BTW, it would be unacceptable if energy was lacking in the presentation here.  It is not.

 

While I appreciate the concept that a dedicated purpose built audio computer sounds better, my experience with various purpose built streaming Linux boxes was not convincing.

 

And yes the Hugo is not 5x the price of the microIDSD, but I was talking about extremes, not the massive grey area between the two points. Sorry if I didn't make that clear.

[afrancois]

38 minutes ago, lmitche said:

I certainly don't want to get into another tedious dsd vs. pcm debate. My post was about where to do upsampling, not about pcm vs. dsd upsampling. Plenty of people upsample pcm with Hqplayer and Audivana.

 

BTW, it would be unacceptable if energy was lacking in the presentation here.  It is not.

 

While I appreciate the concept that a dedicated purpose built audio computer sounds better, my experience with various purpose built streaming Linux boxes was not convincing.

 

And yes the Hugo is not 5x the price of the microIDSD, but I was talking about extremes, not the massive grey area between the two points. Sorry if I didn't make that clear.

Don't know what Linux streaming boxes you are referring to, but it can't definitely be the sMS-200 ultra.

[Johnseye]

7 hours ago, lmitche said:

My view of the motherboard debate, low power, (j1900 or n3700) or high power (i7-6700k - i7 8700k), is a basic choice between up-sampling in a dedicated and proprietary processor, in or close to the DAC, vs. up-sampling in a commodity machine using a third party tools like HQplayer.

 

Curiously, this thread seems to prove that the benefit of various software and hardware tuning tweaks contribute to improved sound quality in an equal fashion with either solution.  This includes clock replacements, software tweaks like AO, USB cables and gadgets and high quality external power supplies.

 

So the real difference is where the up-sampling occurs. With my "high powered" i7-6700k machine, Hqplayer and low cost IFI microIDSD DAC, up-sampling in a PC to DSD512 results in jaw dropping sound quality that I have never heard in any other system, analog or digital. Many other long term audiophiles that have heard this solution say the same.

 

And of course there are those of you that up-sample with proprietary devices like the Chord Blu MKII and Dave DAC.  While I have never heard this combo, I am sure it is spectacular given what we hear from Roy and the other long term audiophiles posting in this thread.

 

Obviously there are many solutions between these two relative extremes. And lastly, the cost of one solution vs. the other, could be close to 5 or 10 times, with the proprietary solution most expensive.

 

Objectively, it is not clear to me in a one to one comparison which will sound better. I know of no other tool then a pair of ears that can tell us the difference.

 

Anyway, my two cents.

 

The clock replacements, software tweaks, USB cables, gadgets and high quality external PSUs are in order to reduce noise and in some cases improve timing.  A completely separate effort from upsampling.

 

My DAC sounds amazing with PCM 192.  It's only capable of DSD64.  It doesn't really do DSD imo and what I've heard of DSD is like others have commented, a bit too soft for my taste.  This is personal preference and if I had a DAC that excelled at DSD512 maybe I'd like that.  I don't know right now and can't have it until I buy a new DAC.  One thing to also consider is how you design the sound of your components.  If all you do is listen to DSD you're selecting components that fit that sound signature because you like it.  Or, if you find it may be too soft you select components to brighten it up.

 

 

4 hours ago, ElviaCaprice said:

I still think regardless of which you prefer (I definitely will take good PCM over DSD any day), HQP upsampled and filtered from the mobo or bit perfect mobo, the sCLK-EX modified mobo will benefit both.  And probably the upsampled mobo more so. 

Why?

Because mobo's are not audiophile constructed components, they are inherently noisy.  When upsampling/filtering from the mobo, you are creating even more noise with the needed demands of more power into an inherently noisy component, the mobo.   But by replacing the clocks on the mobo with clean powered clocks such as the sCLK-EX you are now making the mobo a much better audiophile component.  Thus I think the upsampling/filtering servers will benefit more so than the low power bit perfect servers.

 

As far as the DAC's that do the upsampling/filtering within themselves.  These are audiophile components that are built to handle this function and so will do a much better job than a mobo at not damaging the audio stream.

 

As far as pricing in either server method, I don't see how you can come up with the need for 5 times the price as a necessity?  Also as far as Chord DAC's go, you have to be mindful of the benefits of it's built in low impedant preamp/amp that are a SQ benefit to the direct driving of high efficient speakers. 

Also, my 2Qute only cost $900.  Hardly a huge runup on cost over a ifi Micro.  And there are other higher cost DAC's being used for server upsampling/filtering systems.  Also it is costing less for a low powered mobo.  DAC choice is a personal choice.  I hardly see the cost difference between upsampling/filtering mobo system vs. low power bit perfect mobo system being an argument point of preference for one over the other, and definitely not the more expensive mobo/cpu over the lower priced embedded mobo.

 

But, as Larry said, It all comes down to how it sounds to you. 

 

In my limited understanding upsampling is going to get you to that higher frequency.  It's the filtering that HQPlayer includes which makes the biggest difference and why so many flock to it.  There can be filters built into the DAC as well.  Perhaps the upsample to a higher frequency aside from the filtering can be done on one device better than the other, and given the high resource requirements of DSD512 on a PC I suspect the more horsepower you give it the better.

 

The downside of that is you introduce noise.  Here I think is the crux of Larry's point.  He prefers the benefit he receives by upsampling to DSD512 over all the things one can do to reduce noise as much as possible.  To each their own.  I suspect if I liked the sound of DSD better I would make sacrifices myself.  To your point Elvia, he would still benefit from noise reduction where possible without sacrificing his upsampling performance.  No matter what, noise elimination is a benefit.  I just slapped in some of that 3M EMI-RF absorber in my PC and DAC and I can hear a clear significant improvement.

[afrancois]

26 minutes ago, Johnseye said:

 

The clock replacements, software tweaks, USB cables, gadgets and high quality external PSUs are in order to reduce noise and in some cases improve timing.  A completely separate effort from upsampling.

 

My DAC sounds amazing with PCM 192.  It's only capable of DSD64.  It doesn't really do DSD imo and what I've heard of DSD is like others have commented, a bit too soft for my taste.  This is personal preference and if I had a DAC that excelled at DSD512 maybe I'd like that.  I don't know right now and can't have it until I buy a new DAC.  One thing to also consider is how you design the sound of your components.  If all you do is listen to DSD you're selecting components that fit that sound signature because you like it.  Or, if you find it may be too soft you select components to brighten it up.

 

 

 

In my limited understanding upsampling is going to get you to that higher frequency.  It's the filtering that HQPlayer includes which makes the biggest difference and why so many flock to it.  There can be filters built into the DAC as well.  Perhaps the upsample to a higher frequency aside from the filtering can be done on one device better than the other, and given the high resource requirements of DSD512 on a PC I suspect the more horsepower you give it the better.

 

The downside of that is you introduce noise.  Here I think is the crux of Larry's point.  He prefers the benefit he receives by upsampling to DSD512 over all the things one can do to reduce noise as much as possible.  To each their own.  I suspect if I liked the sound of DSD better I would make sacrifices myself.  To your point Elvia, he would still benefit from noise reduction where possible without sacrificing his upsampling performance.  No matter what, noise elimination is a benefit.  I just slapped in some of that 3M EMI-RF absorber in my PC and DAC and I can hear a clear significant improvement.

What exact type of 3M paper did you use? I myself was thinking of AB5100SHF. So many different types to choose from!

[Johnseye]

6 minutes ago, afrancois said:

What exact type of 3M paper did you use? I myself was thinking of AB5100SHF. So many different types to choose from!

 

Yes, AB5100SHF.  I bought 2 sheets which was just enough for my PC and DAC.  I'll have to buy more if I choose to do this with other equipment.

[afrancois]

50 minutes ago, Johnseye said:

 

Yes, AB5100SHF.  I bought 2 sheets which was just enough for my PC and DAC.  I'll have to buy more if I choose to do this with other equipment.

Thank you for confirming.

I just did a small test with a portable radio set to LW 144 Khz and I can confirm that my audio equipment is subject to much noise. The closer I get to the electric distribution box, 24 port router, security camera boxes and the cable modem the more noise I get. Almost all my audio equipment is in the garage. There are clearly quieter places in my house. So I hope that the 3M paper will help.

[Johnseye]

17 minutes ago, afrancois said:

Thank you for confirming.

I just did a small test with a portable radio set to LW 144 Khz and I can confirm that my audio equipment is subject to much noise. The closer I get to the electric distribution box, 24 port router, security camera boxes and the cable modem the more noise I get. Almost all my audio equipment is in the garage. There are clearly quieter places in my house. So I hope that the 3M paper will help.

 

It will help.  I was surprised how dark the background is by comparison.  The instruments stand out more as a result.  My equipment is side by side and stacked and surrounded by 3 speakers and a sub.  A lot of cabling and equipment in the area.

 

I'm also tempted to try ferrite on some cables.

 

My only question and concern is can there be too much shielding?  That was something Roy mentioned Lee stating; not to overdo it with the EMI paper.  How could it be overdone and to what effect?

[afrancois]

7 minutes ago, Johnseye said:

 

It will help.  I was surprised how dark the background is by comparison.  The instruments stand out more as a result.  My equipment is side by side and stacked and surrounded by 3 speakers and a sub.  A lot of cabling and equipment in the area.

 

I'm also tempted to try ferrite on some cables.

 

My only question and concern is can there be too much shielding?  That was something Roy mentioned Lee stating; not to overdo it with the EMI paper.  How could it be overdone and to what effect?

Where did you put the paper? On the inside of the cover of the DAC? I've just ordered 3 sheets :-)

[BigGuy]

1 hour ago, Johnseye said:

 

Yes, AB5100SHF.  I bought 2 sheets which was just enough for my PC and DAC.  I'll have to buy more if I choose to do this with other equipment.

Not sure how much "shedding" there might be from the ABxxxx sheets but I know there is a cautionary statement on using ERS paper that it contains conductive fibers which potentially cause problems with circuitry.  I actually put mine into clear sheet protectors before placing on top of components.  The ABxxxx do contain metallic particles  which may have the same issue.  There are some of these sheets which are a foil which I would expect to have less of a problem in this regards.

[afrancois]

5 minutes ago, BigGuy said:

Not sure how much "shedding" there might be from the ABxxxx sheets but I know there is a cautionary statement on using ERS paper that it contains conductive fibers which potentially cause problems with circuitry.  I actually put mine into clear sheet protectors before placing on top of components.  The ABxxxx do contain metallic particles  which may have the same issue.  There are some of these sheets which are a foil which I would expect to have less of a problem in this regards.

The datasheet clearly says non-conductive. But you are right, it contains metal flakes.

[BigGuy]

6 minutes ago, afrancois said:

The datasheet clearly says non-conductive. But you are right, it contains metal flakes.

As a wise old Roman once said..."Caveat Emptor"  or, in our frame of reference, "Caveat Audiens"!

[Johnseye]

19 minutes ago, afrancois said:

Where did you put the paper? On the inside of the cover of the DAC? I've just ordered 3 sheets :-)

 

The inside top and sides.  Basically where there were large spaces for application.

 

8 minutes ago, BigGuy said:

Not sure how much "shedding" there might be from the ABxxxx sheets but I know there is a cautionary statement on using ERS paper that it contains conductive fibers which potentially cause problems with circuitry.  I actually put mine into clear sheet protectors before placing on top of components.  The ABxxxx do contain metallic particles  which may have the same issue.  There are some of these sheets which are a foil which I would expect to have less of a problem in this regards.

 

I don't know if there will be shedding either.  Maybe blow out the chassis with an air canister every 1/2 year or so.  This product is advertised to be stuck directly on the back of chips so the only concern would be if enough metallic shedding fell to create a bridge somewhere there shouldn't be one.  I suspect the likelihood of that to be slim.

[Lebouwsky]

This might help

 

http://jplay.eu/forum/index.php?/topic/1128-can-the-guys-whove-used-emirfi-paper-sotm-3m-etc-post-photos-here-of-how-you-did-it/

[mozes]

@Johnseye

when you have some time, can you pls share your impressions of the PH SR7.

[beautiful music]

49 minutes ago, mozes said:

@Johnseye

when you have some time, can you pls share your impressions of the PH SR7.

+1

[Johnseye]

8 hours ago, Lebouwsky said:

This might help

 

http://jplay.eu/forum/index.php?/topic/1128-can-the-guys-whove-used-emirfi-paper-sotm-3m-etc-post-photos-here-of-how-you-did-it/

 

Thanks for sharing Lebouwsky.  After I read through that thread I went back and applied individual pieces to each chip in my DAC, on top of my server's proc heatsink and SSD chips.  I have another motherboard and memory on the way which will go to SOtM along with a switch so will look to covering more of my server, including under the mobo as well as the RAM and other select chips on the mobo itself.  I honestly can't tell if covering each individual chip helped.  The change was so dramatic by covering the top and sides of the cases that by comparison it's minuscule.  That tells me there was significant EMI or RF noise coming from outside the devices.  If there were a lot coming from inside then covering the chips would have had a bigger impact.

 

I'm glad I opened up the DAC again.  We have been discussing what would happen if shedding occurred.  Well the powder coating of my DACs case turns out not to be very good for adhesives.  The top and one side of the 3M absorber had fallen off.  I suspect in an enclosed case without any vents it heats up inside and that also affected the adhesion.  I reattached them more firmly this time and will check tomorrow.  I may have to superglue the pieces on.  The pieces in my server stuck just fine, but it doesn't get very hot inside that case. 

 

In addition to the darker/blacker/quieter background which I would equate to a lower noise floor the bass has opened up significantly.  I also hear a softening or maybe the opposite of there being less of an audible digital edge describes it better.  I would consider this as a step closer to what I hear when listening to vinyl.

 

7 hours ago, mozes said:

@Johnseye

when you have some time, can you pls share your impressions of the PH SR7.

 

6 hours ago, beautiful music said:

+1

 

I need to do this.  I've opened up the case and taken pictures of everything.  I've made those recent changes with the EMI shielding which impacted the sound so I didn't want to confuse any A - B comparisons on the SR7 until that was sorted.  I have to put back the HDPlex power supply into my server to do the comparison.  I should be able to do that tomorrow, but I'm also working on a paper and project for my MBA so time's been limited.

 

What to compare with the SR7:

 

SR7 12v DR vs HDPlex

SR7 19v vs HDPlex

SR7 12v DR vs SR7 19v

SR7 9v vs LPS1 @ 7v (I still have to figure out how to lower the SR7's voltage with the potentiometer.  It requires the adjustment of a dial inside the case along with a voltage measurement)

 

The SR7 vs LPS1 is also challenging because I'm using them both currently to power the sMS-200 and ISO Regen.  If I test one against the other I'll have to stop using the Regen to compare and my ears have become accustom to its sound.  Not a big deal, just another variable.

 

[Johnseye]

9 hours ago, Lebouwsky said:

This might help

 

http://jplay.eu/forum/index.php?/topic/1128-can-the-guys-whove-used-emirfi-paper-sotm-3m-etc-post-photos-here-of-how-you-did-it/

 

I specifically liked this comment.  Looks like one can't overdo it based on this person's experience.  Although there is the risk from increased heat.  Something I'll have to monitor in my DAC.

 

It's also worth noting that when you cut the sheet the cut edges are conductive but the face of the sheet is not.  This guy kyrill recommends taping the edges in certain situations.

 

"Crucially, though, computer transports aren't producing analog output. We don't need to worry about tailoring a frequency response or spoiling a pre-set balance. We're looking only to tame and contain the white noise melée of RF radiated by the board and its power supply that will otherwise bounce around the case and feedback into signal paths. The more absorption you use, the quieter the electrical environment of the case and the closer we reach the holy grail of the computer delivering 'only noughts and ones'. It's hard to see (and hear) how you can overdo absorption in a digital device."

[afrancois]

4 hours ago, Johnseye said:

 

I specifically liked this comment.  Looks like one can't overdo it based on this person's experience.  Although there is the risk from increased heat.  Something I'll have to monitor in my DAC.

 

It's also worth noting that when you cut the sheet the cut edges are conductive but the face of the sheet is not.  This guy kyrill recommends taping the edges in certain situations.

 

"Crucially, though, computer transports aren't producing analog output. We don't need to worry about tailoring a frequency response or spoiling a pre-set balance. We're looking only to tame and contain the white noise melée of RF radiated by the board and its power supply that will otherwise bounce around the case and feedback into signal paths. The more absorption you use, the quieter the electrical environment of the case and the closer we reach the holy grail of the computer delivering 'only noughts and ones'. It's hard to see (and hear) how you can overdo absorption in a digital device."

Does this mean you will be inserting some 3M in your sMS-200?

[beautiful music]

Does anyone attend the “Beyond Off-The-Shelf DAC Chips” seminar by Rob Watts in RMAF occasion?

[Johnseye]

7 hours ago, afrancois said:

Does this mean you will be inserting some 3M in your sMS-200?

 

My intention is to eliminate the sMS-200.  With the new motherboard I'll add a tx-USBexp and either convert that to an Ultra with a sCLK or txUSBultra.  Haven't decided which yet.  It's either add the txUSBultra and don't update my switch clock or add the txUSBultra because of the limited number of taps available.  Theoretically I shouldn't need the txUSBultra with a txUSBexp so that's likely the route I'll take.  Just waiting for confirmation from May.

 

I will be putting the 3M absorber in all devices possible except speakers of course.

[Johnseye]

6 hours ago, beautiful music said:

Does anyone attend the “Beyond Off-The-Shelf DAC Chips” seminar by Rob Watts in RMAF occasion?

 

Better yet, has anyone made any video of this or other events?  Nothing up on YouTube yet.