Review and Measurements of miniDSP SHD DAC, DSP And Streamer

[Krunok]

I'm impressed with this device as it offers some functionlities which are not being offered by other DACs, like Dirac Live digital room correction and Volumio. In my eyes this is worth much more to mention than some jitter spikes at -125dB and below.

I do however agree with @amirm that analog inputs and ADC functionality should be optional to make the price more affordable.

 

[andreasmaaan]

Funny way of counting outputs.

Ha, yes. I think the "4 out" denotes that the processor is capable of outputting a max of 4 different (analogue or digital) audio streams, making it suitable for e.g. a 4-way speaker, a 3-way speaker + sub, 2 x 2-way speakers, etc.

 

[andreasmaaan]

I do however agree with @amirm that analog inputs and ADC functionality should be optional to make the price more affordable.

The SHD does have 2 x analogue inputs, no?

It's the SHD Studio that is a digital processor only.

 

[Krunok]

The SHD does have 2 x analogue inputs, no?

It's the SHD Studio that is a digital processor only.

Correct.

 

[andreasmaaan]

Correct.

Ah I think I misunderstood. You'd like to see a product with digital input only, but with analogue outputs?

 

[Krunok]

Ah I think I misunderstood. You'd like to see a product with digital input only, but with analogue outputs?

I'd like to see analog inputs and ADC functionality as optional. As I don't need them why pay for them.

 

[Deleted member 2944]

I'd like to see analog inputs and ADC functionality as optional. As I don't need them why pay for them.

You'd want them to generate a third version of this unit??
Your choice seems very simple. Select the SHD Studio model and add your own DAC(s) to it. As the ASR website is well outlining, there are multiple inexpensive well-performing DAC's that would make that an extremely attractive setup.

Dave.

 

[Olli]

Or even better: Select the SHD, use it as a network player and active XO AND add your favorite DAC.

 

[amirm]

And I wonder if it i still there when using its network functionality with Volumio?

It was there on S/PDIF without USB connection. But I will test Volumio interface also and report back.

 

[amirm]

To what purpose is to measure something that cannot be heard?

What we want to measure is non-linearity above 11 kHz but we can't do that if we limit the measurement to 22 kHz. Those non-linearities manifest themselves also as intermodulation which does not need higher bandwidth to be seen.

 

[Krunok]

What we want to measure is non-linearity above 11 kHz but we can't do that if we limit the measurement to 22 kHz. Those non-linearities manifest themselves also as intermodulation which does not need higher bandwidth to be seen.

If they don't need higher bandwidth to be seen why extending the frequency range beyond 22kHz?

I would like to offer my general opinion regarding measurements, hopefully you won't mind, it's just my personal view:
If we compare, say 2 devices, one of THD+N of 0.0008% vs other 0.0003%, we have 2 options - we can conclude that the device with 0.0003% is performing better, which technically speaking it is, or we can conclude that both of these devices are performing far below hearing threshold thus the cheaper one is obviously doing a better job. So, what do we conclude? What do you, as audio professionalist advise the forum members to buy? Is it to spend their money on Topping D10 or on some semi-professional DAC that will beat D10 for a fraction of the spec that cannot possibly be heard?

We have seen on many occasions that accent has been put on subjective impressions of the audio equipment, and most of us probably agree that doesn't make much sense. But, on the other hand, should we incline to the other extreme, which is chasing the measurement figures, which IMHO are in the same degree detached from the reality as the subjective impressions? The ulitmate question being here is something like this: is the $100 DAC having THD+N of 0.0008% doing a better or a worse job than a $2000 DAC having THD+N of 0.0003%?

Of course, I'm not implying there's right or wrong answer here, so feel free to choose for yourself..

 

[Krunok]

You'd want them to generate a third version of this unit??
Your choice seems very simple. Select the SHD Studio model and add your own DAC(s) to it. As the ASR website is well outlining, there are multiple inexpensive well-performing DAC's that would make that an extremely attractive setup.

Dave.

No, I want them to offer true modular design so each of us can choose modules based on our needs. Why would I have to pay for, say headphone amp or ADC converter, if I'm not going to use it. With todays technology this shouldn't be a problem for manufacturers to offer so I think we should send such message.

 

[amirm]

If they don't need higher bandwidth to be seen why extending the frequency range beyond 22kHz?

Because that is the requirement of the measurement system to detect non-linearities below that frequency. As I explained, alternative measurements like IMD don't require that.

 

[amirm]

If we compare, say 2 devices, one of THD+N of 0.0008% vs other 0.0003%, we have 2 options - we can conclude that the device with 0.0003% is performing better, which technically speaking it is, or we can conclude that both of these devices are performing far below hearing threshold thus the cheaper one is obviously doing a better job. So, what do we conclude? What do you, as audio professionalist advise the forum members to buy? Is it to spend their money on Topping D10 or on some semi-professional DAC that will beat D10 for a fraction of the spec that cannot possibly be heard?

Reviews have far more data than that one number. So people should look at the full picture and not just that.

THD numbers are not psychoacoustically aware so you can't make clear judgements from them with respect to audibility. What you can do is use them to see which design has more care in it. If that extra care doesn't come at higher price, then it is an easy choice. If there is a premium, then you have to decide if you want to spend the money or not. To the extent the budget is not an issue, my recommendation is to buy the higher performing one as to not worry about audibility concerns. Today's DACs are not liable to get obsolete anytime soon so it makes sense to buy the best and live with it.

 

[Krunok]

To the extent the budget is not an issue, my recommendation is to buy the higher performing one as to not worry about audibility concerns.

And why would you want to do that? Why spending money on something you won't be able to enjoy in any possible way? Why not spend money on some expensive Kobe steak or some fine wine instead? Or why not spend money on some quality speakers where you could actually HEAR the difference?

I'm strugling to find sound logic in what you just said but somehow it sounds "snobistic" to me. No pun intended, but let us not become "measurement snobs" here..

 

[amirm]

And why would you want to do that? Why spending money on something you won't be able to enjoy in any possible way? Why not spend money on some expensive Kobe steak or some fine wine instead? Or why not spend money on some quality speakers where you could actually HEAR the difference?

You can't determine the "any possible way" from THD numbers as I explained.

As to good food and wine, some can afford to get that and good audio gear. Just the same way you do versus someone who just uses the free IEMs that came with their phone. I know I have had plenty of Kobe beef while also owning good audio gear.

Anyway, let's not have debates in this thread. We have a dedicated one here: https://audiosciencereview.com/foru...-debate-thread-about-audio-measurements.2580/

 

[Krunok]

You can't determine the "any possible way" from THD numbers as I explained.

Is that so? Are we really having ANY doubts anyone can hear the difference between say Topping D10 and some so called "highend" DACs costing north of $1500 you have tested?

On the other end, are we having ANY doubts that anyone can hear the difference between loudspeakers that cost $100 and $1500, or the ones that cost $3000 or more?

 

[amirm]

Is that so? Are we really having ANY doubts anyone can hear the difference between say Topping D10 and some so called "highend" DACs costing north of $1500 you have tested?

There is a difference between what we believe, and what we can prove. To prove inaudibility, you have to show that you have a spurious and noise free response of 120 dB at mid-frequencies (see: https://audiosciencereview.com/forum/index.php?threads/dynamic-range-how-quiet-is-quiet.14/). This is not easy to achieve.

Do I think people can hear such artifacts as a rule? No. But I can't prove that no one with any content in any situation (referenced to 120 dB) can't hear it.

Again, let's have these debates in the other thread.

 

[Krunok]

There is a difference between what we believe, and what we can prove. To prove inaudibility, you have to show that you have a spurious and noise free response of 120 dB at mid-frequencies (see: https://audiosciencereview.com/forum/index.php?threads/dynamic-range-how-quiet-is-quiet.14/). This is not easy to achieve.

Do I think people can hear such artifacts as a rule? No. But I can't prove that no one with any content in any situation (referenced to 120 dB) can't hear it.

Again, let's have these debates in the other thread.

IMHO, the first rule of measurements - of ANY measurements - is to have well defined thresholds for each parameter what levels are acceptible and what are not. Don't get me wrong, but If we don't have that defined for all the parameters you're measuring than these measurements fall into cathegory of "l'art pour l'art" (art for art's sake) instead of real science.

I'll stop here as I don't see anything more to state to make this discussion constructive.

 

[Fitzcaraldo215]

Amir - in the past, many miniDSP units, as well as prepros, AVRs supported hires sampling rates like 96k or192k, but when you turned on the Room EQ function, everything was resampled to 48k. This was usually not specified or published. The miniDSP specs imply it is 96k for this unit, but not clearly.

If you still have the unit, could you test for this with Dirac Live? I realize many people do not care much, but for others it might be good to know.