HOLO Audio MAY DAC

[scintilla]

My vinyl is also as important as my digital, but I digitize it, use DSP RIAA correction, apply room correction and drive my amplifier directly from my DAC using software volume control for realtime playback. Same with FM and cassettes. It’s possible to be old fashioned and still keep up with the times... It’s really about motivation and dedication to the cause. 

[Toni-Mang]

Sound Cool...i made some recordings in DSD an in high res, the sound is not bad, but I like to spin records, it is sexy…and I like to buy new carts if the old ones are half the way…but let us talk about the May and how it lights up the rear corners…

 

[lpost]

Preamp (and amp) with input tubes, convolution processing and drive subs high-level from amp. I'm into all of it. I wouldn't be without some tubes in the chain nor without convolution for room fixing. And now add Holo May dac to this list! 550h and she's really coming into her own.

 

Subs simply sound and blend dramatically better when driven with the exact signal as the mains. I use a Jensen transformer as my subs don't have high-level inputs. Works a treat.

[MichaelHiFi]

Finally, my music server is up and running - with HQPlayer no less. I struggled mightily with 2 other computers and never got HQP to work.

 

I'm simply in awe at what is coming out of my speakers except that nothing is coming out of my speakers, music is simply there. Separation, layering and detail in abundance but there's something else. Beauty. The music sounds beautiful. Not sure I've ever used that adjective describing the sound of music in an audio system. And between the speaker and power cables that I've borrowed from my buddy, it seems my "bad" room isn't bad anymore. The bass bloat has been fixed. In fact the texture I'm experiencing in the low notes is simply the best I've ever heard.

 

So where am I at?

I've performed NO adjustments in HQP. Don't even know where to start.

I received my HDPlex 300 watt power supply It was DOA. So I'm using a Dell power brick.

I'm using a USB cable straight out of the server with a few year old SoTM USB card sitting on my table.

My EtherRegen won't deliver network from the B-side moat.

I put my music on a Samsung pci-e SSD drive in a USB input.

My urendu is sitting on the table with my USB card.

I've got a Synergistic orange fuse coming tomorrow for the Kinki

 

So I think there might be some improvement, but I can't imagine what that might look like...

[scintilla]

3 hours ago, Toni-Mang said:

but let us talk about the May and how it lights up the rear corners…

 

I am at the end of week 5, still waiting for a shipment confirmation email. I won't know how it lights up the corners until it gets here...

[Toni-Mang]

1 hour ago, scintilla said:

I won't know how it lights up the corners until it gets here...

I bet, independet of your expectations it will be an aha effect...i had several DAC´s, and this is really hard to beat...

[Quadman]

2 hours ago, scintilla said:

I won't know how it lights up the corners until it gets here...

and yet you helped derail this thread for almost 1.5 pages.  Back on topic tho, you thought your bass was good with what ever dac you have, boy are you in for a real treat.  As Toni-Mang says it "really lights up all back edges of the stage".  In a way you've never heard before,  The energy, the realism the transparency that you will experience will transport you to a musical world you have never been to before, but will happily return to again and again.

[scintilla]

I expect improvements but the showdown between a D90 in DSD direct-dac mode wil be very interesting.  I am not aware of anyone else on the net that has done this head-to-head, and i am going to.  The differences are less than many imagine too.  In both DACs, DSD direct is implemented with a separate low-pass filters that are distinct from the PCM converters.  The difference then is a discrete, hand-made resistor network that is dynamically linearized in the May, vs. on-die resistors in a chip which probably also is using some form of dynamic linearization.  The output stages and power supplies are clearly wide-apart and I suspect are really what separate them in real-world sound quality.  If the AKM4499 was implemented in a May box, they might be nearly indistinguishable in direct DSD mode.  This is on-topic and I suspect will be of keen interest when I am able to put them head-to-head. At least in the form of their current implementations. I truly do not expect them to be that far apart. The D90 is the best DAC that I have had in my system to date in direct-dac dsd-mode, running 128xDSD, 7EC, poly-sinc-ext2.

[Quadman]

14 minutes ago, scintilla said:

I truly do not expect them to be that far apart. The D90 is the best DAC that I have had in my system to date in direct-dac dsd-mode, running 128xDSD, 7EC, poly-sinc-ext2.

Yea well you'll be surprised.  I came from T+A dac8dsd a world class dac when up sampling to dsd 256 EC7 or DSD512, I thought I was in audio nirvana with it.  Then came the Holo and I was schooled again.  It is a whole different league.  You'll see.

[lpost]

25 minutes ago, scintilla said:

I truly do not expect them to be that far apart.

I tend to hear what I expect. Try to feel the differences.

[barrows]

25 minutes ago, scintilla said:

I expect improvements but the showdown between a D90 in DSD direct-dac mode wil be very interesting.  I am not aware of anyone else on the net that has done this head-to-head, and i am going to.  The differences are less than many imagine too.  In both DACs, DSD direct is implemented with a separate low-pass filters that are distinct from the PCM converters.  The difference then is a discrete, hand-made resistor network that is dynamically linearized in the May, vs. on-die resistors in a chip which probably also is using some form of dynamic linearization.  The output stages and power supplies are clearly wide-apart and I suspect are really what separate them in real-world sound quality.  If the AKM4499 was implemented in a May box, they might be nearly indistinguishable in direct DSD mode.  This is on-topic and I suspect will be of keen interest when I am able to put them head-to-head. At least in the form of their current implementations. I truly do not expect them to be that far apart. The D90 is the best DAC that I have had in my system to date in direct-dac dsd-mode, running 128xDSD, 7EC, poly-sinc-ext2.

I agree that the difference between these two will be mostly due to the power supply implementation and the output stage implementation (in terms of DSD direct mode with the D-90, and high rate DSD input), but I also suspect that these differences will be much larger than you might think.  I would expect the May to trounce the D-90 in terms of dynamics, both micro and macro, and in terms of realistic musical presence and tone and timbre.

I found the D-90 to be quite good in DSD direct, but lacking in those areas which I usually feel are due to power supply and analog stage implementation: these aspects of dAC design, to me, often make for much bigger differences these days than the digital differences do.

[GoldenOne]

On 4/8/2021 at 4:27 AM, jeti said:

For Holo May connecting to active speakers, would using HQPlayer to control the volume be a good choice? How does this compare with using a pre-amp, say Holo's Serene?

 

Would appreciate any input and suggestions! After all Serene is also expensive.

I have a holo serene arriving this week so should be able to let you know.

At the moment i'm using a goldpoint SA2X stepped attenuator.
In terms of analog volume control this should be the cleanest way to do it ( https://www.superbestaudiofriends.org/index.php?threads/goldpoint-sa1x-and-sa2x-passive-attenuator-technical-measurements.7303/ ) so it should be an interesting comparison.

I don't know why, but personally I felt that using the goldpoint sounded better than doing DSP volume control via HQPlayer (or Roon).
I know this shouldn't be the case from a theoretical standpoint, but hey.

(I do keep HQP at -3dB to avoid intersample clipping ofc though, but further vol control I do with the goldpoint)

[Extreme_Boky]

Maybe the interconnects/RCA/XLR sockets & plugs and the inherent contact resistance introduced by all of them/those

+

the internal hook-up wiring to that volume pot

and

back from that volume pot...

=> has something to do with the overall impressions you experienced?

 

 

 

[Extreme_Boky]

Oooh, I forgot another important factor:

 

the volume pot should be preceded by a gain (gain of 1) stage of some sort (valve/transistor) to ensure high input impedance for the source

....and....

be followed by a buffer stage (another gain stage), that will ensure a very low output impedance for the amp.

 

Of course, the above is a good practice; however, it can be omitted... but then you'll have a varying output impedance, as seen by the amplifier (both, undesirable scenario and very high output impedance -> at most of the volume settings)

...and...

a non-perfect input impedance, seen by the source. 

 

... those gain stages, that should be used if the volume pot is to be implemented correctly, will require... a power supply, low noise of course,...

 

Summary: measuring a volume pot in isolation does not mean anything.... we have to consider its performance within a contexts of actually controlling the volume in real life situation. And that real life situation includes all of the above elements that will provide sound coloration and will defenetly have a detrimental effect on overall sound reproduction. 

[lpost]

I use HQPe digital attenuation into the May to great effect with a tube preamp. It allows me to get my preamp gain stage at it's optimal SNR and simply sound its best.

 

I did the same thing with my previous dac with a volume control. It's noise floor was not fixed so I would find where it worked well along with attenuation in HQPe combined with same preamp.

 

Getting the gain stages in their sweet spot is key to the 'magic' for me.

[John Hughes]

2 hours ago, scintilla said:

I expect improvements but the showdown between a D90 in DSD direct-dac mode wil be very interesting.  I am not aware of anyone else on the net that has done this head-to-head, and i am going to.  The differences are less than many imagine too.  In both DACs, DSD direct is implemented with a separate low-pass filters that are distinct from the PCM converters.  The difference then is a discrete, hand-made resistor network that is dynamically linearized in the May, vs. on-die resistors in a chip which probably also is using some form of dynamic linearization.  The output stages and power supplies are clearly wide-apart and I suspect are really what separate them in real-world sound quality.  If the AKM4499 was implemented in a May box, they might be nearly indistinguishable in direct DSD mode.  This is on-topic and I suspect will be of keen interest when I am able to put them head-to-head. At least in the form of their current implementations. I truly do not expect them to be that far apart. The D90 is the best DAC that I have had in my system to date in direct-dac dsd-mode, running 128xDSD, 7EC, poly-sinc-ext2.

I had a Topping D90 last year.  The difference between the May and the D90 was dramatic, even in Full DSD. (About a 3 month lag between hearing the topping and the May, so I am fairly confident of my results.  And in NOS mode, leaves it in the dust.   For me moving to an R2R dac made a big difference compared to any of the chip DACs.  Hopefully you will be as pleased as many of us here. 

[Extreme_Boky]

Well, R-2R DAC converts digital electrical current (current flow through a resistor) into potential difference (across that resistor). This potential differences is also capable of creating a current flow further down the line towards analogue amplification, due to that very same potential difference. Electron flow is redirected, if you wish, from one road into another. No damage done (well, apparat from inaccuracies involved with resistor tolerances...)

 

So, there is no silicon involved during this fundamental conversion process, from current level -> into analogue representation; there are no multiple layers of that silicon required to create a transistor structure for the conversion. Here, we have to open the transistor, move it into its conductive state, and then rely on semiconductive change of state (into conductive) and on movement of electrons / holes through that silicon, to achieve the end-result.

 

The above is a fundamental reason why R-2R sounds better than a chip (silicon).

[lpost]

39 minutes ago, Extreme_Boky said:

The above is a fundamental reason why R-2R sounds better than a chip (silicon).

Yes but the description sucks all the fun out of it. Just kidding. It's a good comparative example.

[Extreme_Boky]

Agree, it sucks & hurts; a highly undesirable combination. 

[scintilla]

The professor from my VLSI class is rolling over in his grave right now (he was old then, in the 90's).  I'm ok with it though. While magic fuses and silver wire won't seduce me, a bespoke product that is well implemented by a great engineer gets me all atwitter. I tell my friends this dac is like the ALFA Romeo Spyder I owned in the 90's and the Topping is a Hyundai. 

[Extreme_Boky]

The AKM4499 is a brilliant conversion chip. I monitored very closely Asahi Kasei development, their energy, dedication and the lengths they were willing to go to to design AND build a power supply, adequate rails' decoupling and PCB layout, to be able to give that chip what it really deserves, to be able to actually measure its great performance capabilities. Impressive converter. You see, I do give Kudos where Kudos are deserved and warranted.

 

Maybe if it's implemented in the same way May implemented their R-2R network boards/modules, all else being statues quo within a DAC, it would sound better...? I do not know. 

 

However, I do know that I prefer electron transfer in its original form, used either for the benefit of voltage, current gain (or for R-2R conversion using only a resistor), to any form of signal manipulation / processing done by a peace of silicon.

[Extreme_Boky]

The time-out for edit's.... anyway, just to complete my response.

 

However, I do know that I prefer electron transfer in its original form, used either for the benefit of voltage, current gain (or for R-2R conversion using only a resistor), to any form of signal manipulation / processing done by a peace of silicon, no matter the reduction in length of the traces that a conversion chip brings.

 

The main problem is that people think we have a computer at our hands - where we in fact have an unfortunate evil that was imposed onto humanity in form of a digital reproduction - just so that economy can digress down the path of consumer mentality - indefinitely. Once analogue signal was sampled in time, and converted to digital, the timing errors can only expand from that moment forward, exponentially. It's an error in DNA that can never be corrected. So, in an attempt to retrieve as much as possible of the original analogue information, R-2R conversion will do the least amount of harm. NOS will do the least amount of harm as well. 

 

Manufacturers are moving away from high switching noise - in general. They are also trying to combat the remaining noise that  is unfortunately a requirement for digital reproduction, by utilising extremely low noise supplies. The supercapacitor arrays is the future here, and it's good to see that Chord is doing a great job. Noise (high switching frequencies / high CPU switching voltages)  and timing accuracy -> will never go together well. Silicon IC's and analogue reproduction are not meant to coexist - but they have to in a digital reproduction system.

 

Analog reproduction requires a simple, nonintrusive amplification of an incoming electron stream in its purest form. This is what sounds "natural" to us humans. 

 

But, of course, unfortunately -> it's only digital that we are discussing here, so the rest of the digital silicon IC's (VLSI's...) are a necessity. But, I think that we can try... to use analogue principles wherever we can... within the limitation of a digital reproduction and all that it requires.

 

 

[Miska]

10 hours ago, Extreme_Boky said:

Well, R-2R DAC converts digital electrical current (current flow through a resistor) into potential difference (across that resistor). This potential differences is also capable of creating a current flow further down the line towards analogue amplification, due to that very same potential difference. Electron flow is redirected, if you wish, from one road into another. No damage done (well, apparat from inaccuracies involved with resistor tolerances...)

 

So, there is no silicon involved during this fundamental conversion process, from current level -> into analogue representation; there are no multiple layers of that silicon required to create a transistor structure for the conversion. Here, we have to open the transistor, move it into its conductive state, and then rely on semiconductive change of state (into conductive) and on movement of electrons / holes through that silicon, to achieve the end-result.

 

The above is a fundamental reason why R-2R sounds better than a chip (silicon).

 

It is not really any different from an SDM DAC. Just the resistor structure is different...

 

Quote

NOS will do the least amount of harm as well.

 

It depends... If you run it at 44.1k, it will do massive amount of harm... Because you are not getting anywhere close to the original analog waveform.

 

[StreamFidelity]

2 hours ago, Miska said:

If you run it at 44.1k, it will do massive amount of harm... Because you are not getting anywhere close to the original analog waveform.

 

I am an enthusiastic oversampler and want to understand the process better.

 

If I upsample 44.1kHz to 88.2kHz, will the Nyquist frequency be raised from 22.05kHz to 44.1kHz?

 

The anti-aliasing low-pass filter can then be less steep?

 

And the original analog waveform can be reconstructed better? Or are there other reasons for oversampling?

[MichaelHiFi]

I've got a miniDSP sitting on my garage table. I'm looking to sell but I also have 2 SW260 subs sitting silent as well. Trouble is, my system with the May is producing amazing bass coming from the 11" Eaton drivers which are part of my Usher Towers. Guess I'll hang on to it a bit longer.