Audible Performance Differences in HiFiBerry Digi+ and Digi+ Pro/Digi2

[changer]

Hello all,

The Digi+ Pro is very nice on my Hypex plate amp speakers, no noise at all and best dynamics I have ever heard. When I was routing the sound from PC via a focusrite 2i2 through analogue RCA, it was not nearly as full, clean, articulated and simply ... beautiful! I love it a lot, what a nice device.

Can someone tell from experience what the audible difference between the Digi+ Pro/Digi2 and the Digi+ is? Technically, I know that the latter misses a second clock and has no galvanic isolation. What will I hear of this?

The background of my question is the following:

I am trying to assemble my Raspberry based Logitech Mediaserver and streamer. I acquired a HiFiBerry Digi+ Pro and a Raspberry Pi 4B 2GB. Sadly, and after much of trouble shooting, I learned that the source of my problems was the Revision 1.5 of the Raspberry Pi PCB. The Digi+ Pro HAT was not powering up and consequently could not be used in its intended role as a transport to my Hypex amps. The manufacturer HiFiBerry only most recently got a grip on a new Raspberry board which showed the same behavior and created a blog post on the issue on 15th of February: https://www.hifiberry.com/blog/compatibility-issues-of-the-digi2-pro-and-raspberry-pi-4-rev-1-5/
The different power management circuit in Rev 1.5 is not powering up as cleanily as before. I can verify this problem exists both with Digi2 and Digi+ Pro HATs, as I have tried both. A Raspberry Pi3 that I lent can now power the Digi+ Pro up and the music server is working. Today, I have also checked with a Raspberry Pi4 1GB, which has been re-introduced because of the shortage in chips, but the PCB was already updated to Rev 1.5 and does not work in this combination.

My question is:
I have a neat steel case for the Raspi4, and the equivalent steel case for the Raspi3 is not available anymore in Germany. I can either return all Raspi4 equipment and run the existing Digi+ Pro on a Pi3, which is kind of a backward, but working direction for the Digi+ Pro/Digi2 family at the moment. Or I retain the current Pi4 setup, return the Digi+ Pro and buy a Digi+ until HiFiBerry changes their manufacturing to adopt to the incompatibility issue. They try to find a workaround with existing hardware, but it is hard to say for a laymen if this will work out. What will I miss with the Digi+, which is a lower spec predecessor? Can I hear a difference?

 

[MCH]

Hello Changer,
First of all, sorry i cannot answer your question, and i also dislike when you ask something and people answer telling you to do something else, but just for your consideration. I whish someone had told me some time ago...

I also considered often the hifiberry hats, however now i tend to think they are a bit overpriced for the very little functionality they offer. Have you considered something like a topping d10s for not much more money? You have the same toslink and coax out + dac if you ever needed + a screen if you want to see some info of what is going on (i know, but some people do...) + more versatility + you forget about all these compatibility issues? And if you pair it with a pi zero 2 w you might end up spending a similar amount..
Edit: or even, a external usb powered soundcard

 

[NiagaraPete]

I very much doubt you'd hear a difference in any of those options. Sorry about your luck, I have the Digi Pro and RP4 and steel case and it works and sounds great.

 

[NiagaraPete]

Can you return the Digi2 Pro and buy and DFigi Pro? This supplier has stock https://solen.ca/products/electronics/dac-amp-w-dac/hifiberry-digi-pro-spdif/

 

[changer]

Yes, I can return the Digi+ Pro, and already had returned the Digi2 when I was still thinking it was a faulty board. This is the current solution for Rev 1.5 Raspberry’s, to return to Digi+.

MarcosCh: I only need a digital transport and find the self contained form factor of the Pi + Digi a very attractive and useful solution.

 

[changer]

But I must step down to Digi+. Digi+ Pro and Digi2 share the same sensitivity for slower than before voltage ramp up. Those are nearly the same products, performance wise. The Digi+ is much older.

 

[NiagaraPete]

But I must step down to Digi+. Digi+ Pro and Digi2 share the same sensitivity for slower than before voltage ramp up. Those are nearly the same products, performance wise. The Digi+ is much older.

My digi+ pro works fine. I must have an older RP4.

 

[changer]

Yes, this is also what HiFiBerry did not understand at first, until they got more and more reports. Only since the latest revision of the PCB which included a different power management circuit the Pi4 shows this behavior. Revision 1.1, 1.2 and 1.4 (if anyone ever uses an 8GB Pi for this task) operate properly.

 

[tvrgeek]

"Can I hear the difference"
No one can tell you what you can and can not hear.

Some will invent all kinds of wild terms to tell you what you should be hearing, others will declare with absolute certainty all DACs sound the same. Neither are correct.

A galvanic isolator will only help if you have a ground loop problem, or if it has noise filtering and you have a noisy source. They only work on lower data rates. External isolators for USB powered devices often give you the opportunity to use a higher quality external supply.

Differences in clocks may one of the actual differences in audible effects. Just a guess. Output filtering and IM may be others. It does not seem like we have a good clue on what causes different DACs to sound different on some music, on some amplifiers, on some speakers, but not others.

 

[Tangband]

Hello all,

The Digi+ Pro is very nice on my Hypex plate amp speakers, no noise at all and best dynamics I have ever heard. When I was routing the sound from PC via a focusrite 2i2 through analogue RCA, it was not nearly as full, clean, articulated and simply ... beautiful! I love it a lot, what a nice device.

Can someone tell from experience what the audible difference between the Digi+ Pro/Digi2 and the Digi+ is? Technically, I know that the latter misses a second clock and has no galvanic isolation. What will I hear of this?

The background of my question is the following:

I am trying to assemble my Raspberry based Logitech Mediaserver and streamer. I acquired a HiFiBerry Digi+ Pro and a Raspberry Pi 4B 2GB. Sadly, and after much of trouble shooting, I learned that the source of my problems was the Revision 1.5 of the Raspberry Pi PCB. The Digi+ Pro HAT was not powering up and consequently could not be used in its intended role as a transport to my Hypex amps. The manufacturer HiFiBerry only most recently got a grip on a new Raspberry board which showed the same behavior and created a blog post on the issue on 15th of February: https://www.hifiberry.com/blog/compatibility-issues-of-the-digi2-pro-and-raspberry-pi-4-rev-1-5/
The different power management circuit in Rev 1.5 is not powering up as cleanily as before. I can verify this problem exists both with Digi2 and Digi+ Pro HATs, as I have tried both. A Raspberry Pi3 that I lent can now power the Digi+ Pro up and the music server is working. Today, I have also checked with a Raspberry Pi4 1GB, which has been re-introduced because of the shortage in chips, but the PCB was already updated to Rev 1.5 and does not work in this combination.

My question is:
I have a neat steel case for the Raspi4, and the equivalent steel case for the Raspi3 is not available anymore in Germany. I can either return all Raspi4 equipment and run the existing Digi+ Pro on a Pi3, which is kind of a backward, but working direction for the Digi+ Pro/Digi2 family at the moment. Or I retain the current Pi4 setup, return the Digi+ Pro and buy a Digi+ until HiFiBerry changes their manufacturing to adopt to the incompatibility issue. They try to find a workaround with existing hardware, but it is hard to say for a laymen if this will work out. What will I miss with the Digi+, which is a lower spec predecessor? Can I hear a difference?

As tvrgeek already wrote, different quality of clocks can make a difference soundwise, so is galvanic isolation, but its a big ”maybe” . You have to test it yourself. With just one clock, the soundś gonna be optimized only for 44,1 or 48 kHz, not both.
Im my experience - there is often audible differences with different digital sources, often bigger than differences between dacs. And sometimes there is no audible difference except a high price.…

A serious audiophile is stupid if he dont try different digital sources and listens for himself - sometimes there is no difference to be heard , or there is clearly gains in soundquality , or it will simply sound worse. Buying an expensive dac is a waste of money if the digital source is not a good one.

The tricky part in all this is that a bad digital source ( like, as an example, microsoft windows player ) can prevent a good dac to show its best regarding sound quality.
A bad digital source will make all dacs sound exactly the same - and not really good.
Im not surprised that your digi+pro sounds better than your PC .

 

[mudy]

I read somewhere pi3 is better for audio because simpler circuit.

 

[tvrgeek]

I read somewhere pi3 is better for audio because simpler circuit.

What color is the box? It has about as much relevance.

 

[changer]

All fair contributions, but I find this to be too theoretical for a decision. I was hoping someone could tell me: the missing clock of the Digi muddies the sound, I have heard it, or, the galvanic isolation didn’t matter in my setup. Somewhere, I read the galvanic isolation removed cracks and the Digi+ pro was “darker”, whatever this means, probably less noise? Probably need to test it myself.

The Justboom Digi is quite close to the Digi+ I assume.

MEASUREMENTS: JustBoom Digi S/PDIF HAT for Raspberry Pi. (Plus "Why isn’t digital audio 'just ones and zeroes'?" anxiety...)

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

 

[tvrgeek]

Be careful of reading "somewhere." By your comments, the author did not have a clue what they were talking about.
There is only one absolute correct answer: "It depends"

Again, neither I nor anyone else can tell you what you are going to hear. Your "reading somewhere" may have already set an unconscious bias. It takes a lot of practice to get over them. Bottom line. Try something. If you don't like it, try something else.

WMP on Win 10 is not the evil that it was pre Win 7, but the bias persists. It is also the only freaking player with a UI that makes sense. Set 1 dB down and it won't hit the compressor. I know, I tried about 30 of them. The evil is in the recordings. Many of mine are old from the analog days. Many the mixing engineer was a bone-head. Some the studio work was terrible. Old Clapton. Mix on the Billy Joel records, Airplane 2400 . 60's right and left, not stereo. Horrible. Another big mistake is boosting with a 16 bit processor. Use APO to cut only.

 

[changer]

Picked it up here, but lost my interest fast when I saw the conclusion https://www.bitlab.nl/page_id=130

Anyhow, I do not use Windows for listening anymore, it is a LMS server now. I was asking for experiences btw.

 

[Mnyb]

We are talking digital transport here with spdif out .

I use a couple of them for picore players conected to an lms server works great .

I do not have new enough RasPI to have faced these problems. Mine are of previous generations . So I can’t help with specific issues .

But the general case is that digital transports does not have “sound” , for this to happen you have to search long and hard for especially bad combinations key here is a broken by design DAC as reciever of the coaxial or optical signal .

If you solve your issues these things just works , they migth not be the most perfect transport you can have , but they passed the good enough treashold enjoy

If you want galvanic isolation does not some of these HAT’s offer toslink ? It’s theoretically not as good as coax but no functional DAC would mind .

The purpose of using these is not the same as for example a Topping D10 it’s not for Usb to spdif conversion .
They are endpoint for a distributed music server in this case LMS .

 

[Tangband]

We are talking digital transport here with spdif out .

I use a couple of them for picore players conected to an lms server works great .

I do not have new enough RasPI to have faced these problems. Mine are of previous generations . So I can’t help with specific issues .

But the general case is that digital transports does not have “sound” , for this to happen you have to search long and hard for especially bad combinations key here is a broken by design DAC as reciever of the coaxial or optical signal .

If you solve your issues these things just works , they migth not be the most perfect transport you can have , but they passed the good enough treashold enjoy

If you want galvanic isolation does not some of these HAT’s offer toslink ? It’s theoretically not as good as coax but no functional DAC would mind .

The purpose of using these is not the same as for example a Topping D10 it’s not for Usb to spdif conversion .
They are endpoint for a distributed music server in this case LMS .

Digital signals are RF.
Further, digital transports with spdif is dependent of the clock noise from the source and the ability of the dac to recreate the clock . Asychronuos USB is better .
Spdif outputs can have different qualitys - if the RF noise from the transport is big, its harder for the dac to recreate the clock.
Resampling can also be bad for sound, if done in a bad way.

A bad digital transport has less details in high freq and sound will be less dynamic In the whole music spectra. Listening to music will be boring and not fun.

 

[Mnyb]

Digital signals are RF.
Further, digital transports with spdif is dependent of the clock noise from the source and the ability of the dac to recreate the clock . Asychronuos USB is better .
Spdif outputs can have different qualitys - if the RF noise from the transport is big, its harder for the dac to recreate the clock.
Resampling can also be bad for sound, if done in a bad way.

A bad digital transport has less details in high freq and sound will be less dynamic In the whole music spectra. Listening to music will be boring and not fun.

I must disagree, in both theory and practice , even here at ASR Amir have measured how a good DAC ( almost all modern ones ) produces identical analog output even with a source that’s quite jittery .

In practice using CD material my old Squeezebox 3 my RPi and my MeridianG98DH DVDA player sounds identical as digital transport. Hence the endpoint from LMS to Meridian system is an RasPI with picore player .

Even as you say spdif can vary in quality the DAC sucks it up and it works fine ( they produce identical signals at the analog output).
Less detail and loss of dynamics is also an absurd consequence ( your anthropomorphising it for lack of better word ) .
These changes requires DSP ( apply dynamic compression to lessen the dynamics and roll of the FR at the high end).
A true disturbance caused by bad spdif would probably produce glitches and noise not subtle analog changes .

You have to really lock for a transport bad enough to find a corner case .
Sorry not this audiophile nonsense here at this forum please ?

 

[tvrgeek]

Picked it up here, but lost my interest fast when I saw the conclusion https://www.bitlab.nl/page_id=130

Anyhow, I do not use Windows for listening anymore, it is a LMS server now. I was asking for experiences btw.

You were asking for someone to tell you what you would hear. No one can. If you can, and that is denied by some with a predisposed bias, recognized by others who actually test , only you would know. That's 50 years of experience involved with the pursuit of audio that sounds realistic.

With all due respect to Amir, I know without question of a case where three DAC's sound different in one specific aspect. Once the sensitivity elsewhere was identified ( what the DAC was generating) then they did sound identical to my ears, in my system, with my source material. I know, that seems to really upset a few, but ones belief does not change reality.

 

[changer]

I tend to agree with you that power might be power, signal might be bit by bit identical, but there is sometimes still a different end result. This is why I was asking and thanks for the replies. Got to test it by myself, obviously.