HOLO Audio MAY DAC

[lpost]

LNS15 in both cases. I asked the US distributor, Tim at Kitsune, and said he's experienced the same with some USB ports feeding the May dac. He suggested disabling PLL circuit in the May dac, this made no difference for me. The sound is thinner and a bit less warm using 1.5M vs 768k.

 

I experienced this thinner sound with both the motherboard ports and two Asmedia boards with different chipsets. All sync at 1.5M but don't sound as good. Everything else the same.

 

I also tried 1.5M/20 with all the various USB ports and none would play smoothly. Sound would dropout every 1/2 second or so. No pops or cracks just a dropout with no sound then it would be back. Setting to 16 bit allowed music to play without these 'dropouts'.

[Diavolo]

1 minute ago, lpost said:

LNS15 in both cases. I asked the US distributor, Tim at Kitsune, and said he's experienced the same with some USB ports. The sound is thinner and a bit less warm using 1.5M vs 768k.

 

I experienced this thinner sound with both the motherboard ports and two Asmedia boards with different chipsets. All sync at 1.5M but don't sound as good. Everything else the same.

I'm just using my motherboard usb port on a recent Asus/Intel based motherboard and 20bit at 1.5mhz sounds great, but honestly 16/44.1hhz sounds great too.  I'm not noticing much difference between the two, truth be told, but I'm not noticing anything deleterious either.

[Quadman]

1 hour ago, lpost said:

Did you grab the one on Usaudiomart for just a couple hours yesterday?

Yes, shipping today but UPS is snail, they project Monday arrival to E TN😥

1 hour ago, fds said:

maybe even direct to power amp with software volume control

I bi-amp and will not use an RCA splitter to do this, I run to a buffered preamp then RCA to each amp and I use HQP volume control currently.  Of course DSD512 will be tested.  Currently my T+A is so real, so solid and so palpable, if it gets any better I'll never sleep.

1 hour ago, lpost said:

I find DSD to be quite rolled off with the May compared to the same track PCM.

Interesting I will run pink noise upsampled to DSD512 and use audiomatica's Clio to measure RTA frequency response and compare to PCM 768k.  I would be shocked if there is a difference, we'll see.  I believe the firmware will be 30.14 which cannot do 1.5K or 1024DSD, I may roll back to 30.12 in the future if I feel the need to do so.  

 

I will also be building a new i7-10700K server to replace my older ryzen 7-1800x so more filters to play with...Whoo Hoo.

[lpost]

I think I’ll just run NOS full path for a while and remove the HQPe alterations. The dac is still changing so it’s chasing a ever changing beast at this point.

 

like diavolo said, it sounds fine without upsample and shapers.

[Iceaero]

And in the opposite direction, I will be attempting tomorrow to run USB via an Ethernet extender (30ft) and try to compare 768k/1.5M USB with 192k AES (which I’ve been using for the past few months).

 

I expect it to be too subtle to distinguish, but who knows!

[Extreme_Boky]

I have spent a lot of time experimenting as well. The most relaxed, detailed and natural sound I get is with LSN15 shaper, upsampled to 1.536Mhz, using poly-sync-lp filter and bit-depth of 20bits. It is an exceptionally enjoyable experience. Moving from 768KHz to 1.536MHz is like going from standard crystal oscillators to low phase noise ones (NDK SD to SDA). The bass frequencies are also fuller, better defined and fully integrated. Nothing sticks out; everything sounds just right. 

 

There's a caveat to the above....  I get much better sound (more natural and relaxed) with my laptop - battery powered, compared to NUC11 powered with a low noise linear power supply. The difference is striking. I noticed that many manufacturers are moving towards "zero noise" supper capacitors based power supplies, which they also use to do upsampling inside the DAC unit itself. 

 

I was just about to give up on digital and go back to analog (turntable and reel-to-reel)... and then along came the May DAC and the HQ Player.

[fds]

14 hours ago, Quadman said:

Currently my T+A is so real, so solid and so palpable, if it gets any better I'll never sleep.

I know exactly what you mean ;-) ... same experience here with the T+A ... leading already to a significant reduction of sleeping time.

[Quadman]

Hey @Iceaero and @Extreme_Boky did you roll back your firmware from 30.14 which I believe comes installed on the May dac's and is unable to do DSD1024 or PCM 1.536Mhz to version 30.12 which does handle those sampling frequencies?  If so how did you do that? I downloaded the driver and firmware package found on the spring 2 page and the directions at a glance seem a bit awkward.

[lpost]

3 hours ago, Quadman said:

Hey @Iceaero and @Extreme_Boky did you roll back your firmware from 30.14 which I believe comes installed on the May dac's and is unable to do DSD1024 or PCM 1.536Mhz to version 30.12 which does handle those sampling frequencies?  If so how did you do that? I downloaded the driver and firmware package found on the spring 2 page and the directions at a glance seem a bit awkward.

Connect a Windows 10 machine to the May USB. Run the HoloAudio_v5.0.0_2020-12-15_setup.exe to install ASIO driver and programs. Go to program folder for 30.12 or 30.14 respectively. Run holousbaudioDfu.exe. It's pretty self-explanatory from there and only takes a few seconds to update/switch fw.

 

I upgraded to 30.12 for 1.5M support a few days ago. I found that 30.12 worked fine for 768k/20bit however, this morning I went back to .14 as my motherboard USB ports sound absolutely terrible at 1.5M/20 and still not quite right at 1.5M/16. I suspect this is all dependent on your particular USB output et al.

 

 

[Quadman]

25 minutes ago, lpost said:

I upgraded to 30.12 for 1.5M support a few days ago. I found that 30.12 worked fine for 768k/20bit however, this morning I went back to .14 as my motherboard USB ports sound absolutely terrible at 1.5M/20 and still not quite right at 1.5M/16. I suspect this is all dependent on your particular USB output et al.

I thought in the notes in the firmware/software file and on the holo spring 2 page they mention .14 has better USB compatibility among various boards.  Of course you lose 1.5M/1024 ability, (only .12 allows that).  Guess it is MB dependent.

[lpost]

Yep, I understand this to be the case too and probably why they ship with the more compatible 30.14 version. I simply wanted to try 1.5M to see if it would play at all.

 

I have an ASRock x370 Pro Gaming board with AMD 1700x I believe. It's a few years old now. The 3.1 Gen 2 port works, the other 4 3.0 ports do not sync at all.

 

I'll go back to I2S in a few days to have a listen but it's limited to 192PCM but I'm not finding HQPe to be a HUGE advantage over NOS all the way. Sure, it's a little better but not if my USB port is making a mess of it.

[lpost]

On 12/11/2020 at 5:20 PM, GoldenOne said:

I don't believe there is no. Sorry, probably could have worded that post better. 

 

What I meant is that the pll puts basically any digital source on equal footing with the may. Be it an expensive streamer or basic optical from my pc. 

 

USB sounds the same regardless of if the pll is on or not. 

Tim tells me emphatically that the PLL is used with USB and the May USB runs in 'master' mode.

[Diavolo]

5 minutes ago, lpost said:

Tim tells me emphatically that the PLL is used with USB and the May USB runs in 'master' mode.

This is a pretty typical way to remove jitter from USB these days.  Benchmark was using PLL to get jitter <-50ps 20 years ago.  It's always been centered around usb to my knowledge, but I'm not an expert. Benchmark provides great white papers on their jitter reduction PLL which at the time was quite remarkable as it bested jitter measurements of many much more expensive (over $10k) DACS and Atkinson was blown away by the original Benchmark DAC 1 - USB.  

[lpost]

1 minute ago, Diavolo said:

This is a typical way to remove jitter from USB these days.  Benchmark was using PLL to get jitter <-50ps 20 years ago.  It's always been centered around usb to my knowledge, but I'm not an expert. 

It's just contrary to what Jussi stated: There shouldn't be any PLL involved when using asynchronous USB....

 

I have no idea. I don't hear any difference with it on or off and it doesn't change the locking times for better or worse for me using USB.

[Diavolo]

4 minutes ago, lpost said:

It's just contrary to what Jussi stated: There shouldn't be any PLL involved when using asynchronous USB....

 

I have no idea. I don't hear any difference with it on or off and it doesn't change the locking times for better or worse for me using USB.

Asynchronous communication is defined as transmission of data without the use of an external clock signal. This allows data to be transmitted intermittently rather than in a steady stream

From Audio Misconceptions blog. This doesn't reduce jitter without some buffering.  PLL just reclocks packet data to the DAC, mitigating jitter effectively. 

 

[John Hughes]

1 hour ago, lpost said:

It's just contrary to what Jussi stated: There shouldn't be any PLL involved when using asynchronous USB....

 

I have no idea. I don't hear any difference with it on or off and it doesn't change the locking times for better or worse for me using USB.

I've also discussed this with Tim and he says that PLL is used with USB.  I've also disabled it for some sources using the USB and AES inputs and my ears detected only a very slight improvement with PLL.  So little it wasn't worth worrying about whether it was on or off. 

[Extreme_Boky]

I can hear the difference with PLL ON, USB input. The sound is more spacious. Frequency extremes are more pronounced, they go higher and deeper. I like it better:)

 

[lpost]

Just now, Extreme_Boky said:

I can hear the difference with PLL ON, USB input. The sound is more spacious. Frequency extremes are more pronounced, they go higher and deeper. I like it better:)

 

My May is a long way from 500+ hours so I'm not judging it too harshly yet though I did have a day or two that was pretty rough in the high frequencies.

 

What USB output are you using? When I try 1.5M/16 I get a very strained etched sound. Completely different than using the same USB port at 768K/20.

[lpost]

3 minutes ago, Extreme_Boky said:

I can hear the difference with PLL ON, USB input. The sound is more spacious. Frequency extremes are more pronounced, they go higher and deeper. I like it better:)

 

My May is a long way from 500+ hours so I'm not judging it too harshly yet though I did have a day or two that was pretty rough in the high frequencies.

 

What USB output are you using? When I try 1.5M/16 I get a very strained etched sound. 1.5M/20 is unlistenable and drops out. Completely different than using the same USB port at 768K/20.

[Diavolo]

7 minutes ago, lpost said:

My May is a long way from 500+ hours so I'm not judging it too harshly yet though I did have a day or two that was pretty rough in the high frequencies.

 

What USB output are you using? When I try 1.5M/16 I get a very strained etched sound. Completely different than using the same USB port at 768K/20.

Shouldn't be sounding like that even out of the box. Mine was very natural sounding from the get go. Maybe there's something wrong with your unit or the source. This applies to any pcm sampling rate through roon or hqplayer or DSD up to 512. 

[lpost]

It's my source. Just swapped from AMD to a Intel desktop machine 1.5M/20 no sweat and sounds good. Time to troubleshoot.

[sledwards]

22 minutes ago, lpost said:

What USB output are you using? When I try 1.5M/16 I get a very strained etched sound. 1.5M/20 is unlistenable and drops out. Completely different than using the same USB port at 768K/20.

I am seeing the same behavior from my JCAT USB-XE card. Switched to mobo USB and 1.5M sounds nice.

[lpost]

My AMD with good linear power supplies sounds better with motherboard ports at 768/20 than mini desktop Dell running either NAA or Window Server 2019 Roon Bridge at either 768/20 or 1.5M/20, no big surprise. At least the Intel plays at 1.5M/20 with NAA and doesn't sound horrible. Roon is limited  to 768 but I didn't even try upsampling with it as it's a non-starter for me compared to HQPe/NAA.

 

I've got 3 inexpensive cards on the way based on 3142, 2142 and 1142 Asmedia chipsets to test. 

[Miska]

7 hours ago, Diavolo said:

This is a pretty typical way to remove jitter from USB these days.

 

How do you remove jitter from something that doesn't have it's own clock? Asynchronous USB runs off from the DAC's master clock.

 

You need it for S/PDIF and AES/EBU though.

 

7 hours ago, Diavolo said:

Benchmark was using PLL to get jitter <-50ps 20 years ago.

 

That wasn't asynchronous USB, that was the old slave-clocked USB (Audio Class 1.0) where clocking was similar to S/PDIF.

 

7 hours ago, Diavolo said:

Atkinson was blown away by the original Benchmark DAC 1 - USB

 

Which was totally different technology than USB interfaces used today... Clocked in totally different way.

 

[Miska]

6 hours ago, Diavolo said:

Asynchronous communication is defined as transmission of data without the use of an external clock signal. This allows data to be transmitted intermittently rather than in a steady stream

From Audio Misconceptions blog. This doesn't reduce jitter without some buffering.  PLL just reclocks packet data to the DAC, mitigating jitter effectively. 

 

 

That external clock signal is DAC's conversion clock. There is nothing to reclock or nothing to remove jitter from. Because the USB stream doesn't include a clock. Since there's only single clock, there's nothing to reclock. There's no point in reclocking clock with itself.