Minidsp SHD Review Updated (DSP, DAC & Streamer)

[Photony83]

ok so my current set up:

SHD - Digital 1&2 output Coax output to Denafrips Pontus II ; analog 3 & 4 output to 2 Rythmik subs.
-bought a Bluesound Node to see if that is a good interface.. not really so that's going back
-I signed up for Apple Music and playing HiRes lossless audio via Airplay straight into the SHD. this seems to work well but not entirely convinced the playback is getting top signal like 192kHz. I checked Volumio app while it was supposed to play a 192kHz track but it only showed 44.1khz.
-TV attached via optical input as well

volume set at -10db on the LAN input for airplay and playback volume controlled via Apple Music volume control.

So end result: it sounds different: overall better, deeper soundstage, clearer instruments, female voice seems more real, less fatiguing but the biggest thing for me is that my wife noticed a difference and she never notices a difference... ugh sound like a YouTuber.. but I would be lying by saying it sounded the same as the set up prior. I have to rerun Dirac since I am running the previous settings.

not sure how to explain this to my rational side but the sound is better, in the end maybe about 5%-10%. my thought is that my system and room are not perfect. I have glass and mirrors built into this rooms I can't get rid of. my equipment is not top of the line. all these imperfect things have to work together by the time the sound gets to me, who might be the most imperfect thing in the room. it just gels better to my ears after its has passed through the denafrips than without it.

 

[mdsimon2]

Nice anecdote and I am sure that you hear a difference but from what you done so far you have no idea why there is a difference. Likely explanations include:

1) Level difference between Denafrips and SHD DAC
2) Phase differences affecting main / sub crossover as a result of different delays between Denafrips and SHD DAC, especially as the Denafrips has a FIFO buffer
3) Expectation bias

In addition if you are using airplay hi-res sources will be resampled to 44.1 kHz / 16 bit.

Michael

 

[Photony83]

Nice anecdote and I am sure that you hear a difference but from what you done so far you have no idea why there is a difference. Likely explanations include:

1) Level difference between Denafrips and SHD DAC
2) Phase differences affecting main / sub crossover as a result of different delays between Denafrips and SHD DAC, especially as the Denafrips has a FIFO buffer
3) Expectation bias

In addition if you are using airplay hi-res sources will be resampled to 44.1 kHz / 16 bit.

Michael

I would pick number 2 out of those as the most likely case. I had the volume set to 75db with a test tract. I tried to best internally to stay objective, but was open to see if there was difference. If I were to revert back to just the SHD I am not sure I would be able to replicate this result. I will try running the speakers only without subs and test that way next. my Tekton do go down pretty deep on its own.

I have a lot more listening to do. but the Denafrips is a keeper for me.

was hoping apple changed some protocol to allow the hi res audio stream to airplay capable devices without downsampling, but even with that conversion it sounds great.

 

[nothingman]

I hooked it all up, didn't touch Dirac config (which, tbh does need updating cause I moved the speakers a bit) and it sounds good. The stock tubes, however, sometimes seem to emit a noticeable hi-pitched whine, which means I haven't been using the tube mode a lot.
I've since ordered and received different tubes, but haven't yet had the time to do the rolling.

At first, I didn't have another pair of balanced cables to connect SHD and Freya, so I used single ended... It was noisy as hell. Now with balanced interconnect it's perfect. I listen 10+ hours through this, Freya in passive mode. I'll post an update there when I switch out the tubes.

One has to wonder why you’re bothering with the Freya. I don’t mean that in a snarky way, but really, what useful role does it have in that signal chain? It has noisy tubes? The RCA signal is compromised? Everything I’ve connected to my SHD via RCA has been dead silent.

ok so my current set up:

SHD - Digital 1&2 output Coax output to Denafrips Pontus II ; analog 3 & 4 output to 2 Rythmik subs.
-bought a Bluesound Node to see if that is a good interface.. not really so that's going back
-I signed up for Apple Music and playing HiRes lossless audio via Airplay straight into the SHD. this seems to work well but not entirely convinced the playback is getting top signal like 192kHz. I checked Volumio app while it was supposed to play a 192kHz track but it only showed 44.1khz.
-TV attached via optical input as well

volume set at -10db on the LAN input for airplay and playback volume controlled via Apple Music volume control.

So end result: it sounds different: overall better, deeper soundstage, clearer instruments, female voice seems more real, less fatiguing but the biggest thing for me is that my wife noticed a difference and she never notices a difference... ugh sound like a YouTuber.. but I would be lying by saying it sounded the same as the set up prior. I have to rerun Dirac since I am running the previous settings.

not sure how to explain this to my rational side but the sound is better, in the end maybe about 5%-10%. my thought is that my system and room are not perfect. I have glass and mirrors built into this rooms I can't get rid of. my equipment is not top of the line. all these imperfect things have to work together by the time the sound gets to me, who might be the most imperfect thing in the room. it just gels better to my ears after its has passed through the denafrips than without it.

As @mdsimon2 notes, whatever the source, AirPlay on your device is converting that to 44/16 ALAC in real-time and then sending that file to the SHD‘s Airplay receiver. So, you haven’t listened to an ounce of even 24/48 in that setup, much less 24/96 or 24/192.

I don’t see the benefit of adding an extra, expensive DAC with known variable latency issues between the mains when the SHD could integrate it all at once, but that’s just me. REW and a microphone and you could get it all going nicely with proper integration, not “I have no idea why this works but it seems to and I have no chance of replicating it later” type integration.

 

[eisenb11]

I hooked it all up, didn't touch Dirac config (which, tbh does need updating cause I moved the speakers a bit) and it sounds good. The stock tubes, however, sometimes seem to emit a noticeable hi-pitched whine, which means I haven't been using the tube mode a lot.
I've since ordered and received different tubes, but haven't yet had the time to do the rolling.

At first, I didn't have another pair of balanced cables to connect SHD and Freya, so I used single ended... It was noisy as hell. Now with balanced interconnect it's perfect. I listen 10+ hours through this, Freya in passive mode. I'll post an update there when I switch out the tubes.

Looking forward to your update with the rolled tubes! I read that the GE tubes sound good as well as combinations of others. Do you notice any difference in passive mode (runs it through JFET I think) vs not having the Freya?

 

[Photony83]

As @mdsimon2 notes, whatever the source, AirPlay on your device is converting that to 44/16 ALAC in real-time and then sending that file to the SHD‘s Airplay receiver. So, you haven’t listened to an ounce of even 24/48 in that setup, much less 24/96 or 24/192.

I don’t see the benefit of adding an extra, expensive DAC with known variable latency issues between the mains when the SHD could integrate it all at once, but that’s just me. REW and a microphone and you could get it all going nicely with proper integration, not “I have no idea why this works but it seems to and I have no chance of replicating it later” type integration.

ok I will get some measurements today with REW with Dirac off.. Apple Music via airplay 16/44. will disable to subs and keep volume 75db measured with a separate meter.. theoretically shouldn't have any differences at least measurement wise. I can also try running off my NAS with Volumio for higher bitrates. will post results

 

[mdsimon2]

Acoustic measurements are not going to be very useful in quantifying why things are different, too much variation. If you want to determine if things are different at the DAC output I would start with electrical measurements using an ADC.

Michael

 

[kokoon]

One has to wonder why you’re bothering with the Freya. I don’t mean that in a snarky way, but really, what useful role does it have in that signal chain? It has noisy tubes? The RCA signal is compromised? Everything I’ve connected to my SHD via RCA has been dead silent.

I wanted to have the option to add tubes to the soundpath, mainly to hear with my own ears what the fuss is about. I've never listened to a good system with tubes before.

Now that I have it set up, it's just a press on my harmony remote and I'm in "tube mode". One press again, and I'm back to essentially direct connection between the SHD and my nCore power amp. I made it so that it also switches the SHD preset, so the tube gain is compensated.

Now that the interconnects are balanced, it's indeed dead silent, just like yours - even when running it thru tubes.

 

[kokoon]

Looking forward to your update with the rolled tubes! I read that the GE tubes sound good as well as combinations of others. Do you notice any difference in passive mode (runs it through JFET I think) vs not having the Freya?

Me too, just have to clear my schedule for a day
Btw I got a matched pair of GE coin base and a matched pair of Sylvania Baldwin black plates.

Freya+ has three modes. Passive is just passive, no buffers or anything afaik. Sounds completely transparent to me, when no attenuation (volume knob to max). Active is JFET, I haven't used that a lot but I didn't notice any difference in sound when I did.

 

[eisenb11]

I wanted to have the option to add tubes to the soundpath, mainly to hear with my own ears what the fuss is about. I've never listened to a good system with tubes before.

I just had a deal fall in my lap that I had to hop on. Picked up a McIntosh C2300 tube preamp for significantly under market value. Paired that with my McIntosh MC152 solid state amp. Preamp being fed by the MiniDSP SHD. I configured the SHD with a high pass to the tube preamp and a low pass to the active sub (class D). Ran Dirac to room correct the whole thing.

Wow oh wow. I'm absolutely floored. So happy to learn that tubes do play well with Dirac!

 

[eyesofra]

my minidsp shd has just arrived ! yet to run Dirac but it's been working like a charm off LAN
In Amir's measurements, the best SINAD of112.x db was reached at 4v @ Vol -6.5 balanced XLR out. It started clipping there after.
Question pls - with unbalanced RCA out, will the behaviour be similar ? Meaning will it start clipping after vol -6.5 as well ?

 

[mdsimon2]

Those results never made much sense as every miniDSP product I’ve ever had clips at 0 dB. If I had to guess I bet that the routing matrix was setup so left and right inputs were summed and that is why it clipped at volume levels above -6.5 dB.

Anyone have a SHD and a scope to verify?

Michael

 

[Iwf]

Could I ask about the analogue inputs on the SHD?
Is it possible to bypass everything so the output from the SHD is exactly the same as the input? I’m thinking of using the SHD as a pre amp and trying different DACs against the built in DAC within the SHD

 

[f1shb0n3]

Could I ask about the analogue inputs on the SHD?
Is it possible to bypass everything so the output from the SHD is exactly the same as the input? I’m thinking of using the SHD as a pre amp and trying different DACs against the built in DAC within the SHD

My understanding is that SHD will always digitize the analog input so it can digitally process the signal for PEQ, crossover, Dirac, etc.

Why you need to feed analog to SHD? Digital to SHD and you are all good to try out different DACs. I did that myself comparing SHD’s DAC to D90, D90SE and RME ADI-2. Did not hear a difference to reliably distinguish any of them with music. Only difference was the noise level when I put my ear to the tweeter - predictably the better measuring ones have less hiss at the tweeter but the difference disappears at 2-3 inches from it as noise is so low even for SHD’s DAC.

 

[radix]

Could I ask about the analogue inputs on the SHD?
Is it possible to bypass everything so the output from the SHD is exactly the same as the input? I’m thinking of using the SHD as a pre amp and trying different DACs against the built in DAC within the SHD

I am not aware of any setting like that. AFAIK, the analog input always goes through the ADC (which I think is 32-bit). Here's the figure from the manual (p. 44).

 

[Iwf]

Not having a pre, means I can’t easily a/b test the DACs, but it’s a bit academic as I plan on selling my Mytek to part fund the SHD.

I though there might be a bypass option

 

[Sir Sanders Zingmore]

Could I ask about the analogue inputs on the SHD?
Is it possible to bypass everything so the output from the SHD is exactly the same as the input? I’m thinking of using the SHD as a pre amp and trying different DACs against the built in DAC within the SHD

It’s not possible.
I have asked miniDSP about whether the SHD volume control is analog or digital. They told me it is digital which means analog inputs must go through the ADC

 

[Iwf]

That’s what I thought, but I thought I’d ask.

I ordered mine this morning direct from HR. Not sure how long itll take to get here though.

 

[f1shb0n3]

Not having a pre, means I can’t easily a/b test the DACs, but it’s a bit academic as I plan on selling my Mytek to part fund the SHD.

I though there might be a bypass option

If you use USB or send digital signal to SHD (or SHD Studio) it becomes a great tool for comparing DACs because you can adjust gain on each output separately in 0.1dB increments. DACs always have some minor differences in what exact voltage they produce for 0dB signal and often don't have a volume control or it's at 0.5dB increment if they do. Level matching is essential when A/B comparing DACs and you need to get them to 0.1dB of each other otherwise the volume difference is all you are going to hear as difference when switching.

 

[ebslo]

Those results never made much sense as every miniDSP product I’ve ever had clips at 0 dB. If I had to guess I bet that the routing matrix was setup so left and right inputs were summed and that is why it clipped at volume levels above -6.5 dB.

Anyone have a SHD and a scope to verify?

Michael

I measured mine with a DMM. It's a very old and cheap DMM so I'm not sure how much I trust it, but it read 4.1V RMS with a 0dBFS sin wave playing from the REW signal generator.