Help with a multichannel Dante setup please

[Rja4000]

But the video player won’t be.

I'm not sure what you mean here.
The player is the PC, isn't it ?

 

[NTK]

Yeah but what’s stopping the player clock from drifting from the Dante leader clock?

How does Dante clocking work? | Dante

Dante devices each contain a very high quality VCXO clock, and are synchronized with one another over the network using the IEEE 1588 Precision Time Protocol (PTP). One device is elected as the Leader Clock to which other devices are synchronized. By default this selection takes place...

www.audinate.com

 

[Ifrit]

player clock from drifting from the Dante leader clock?

Player clock in this case IS Dante leader clock.

 

[MaxwellsEq]

I think he is wavering towards a more conventional solution. We will see.

An all analogue solution will have less latency. This may not matter.

 

[Ifrit]

But the media player isn’t a Dante certified software

But player software is using DVS as the interface. If you use player with PCIe interface, what clock is player using?

BTW I make audio products that have Dante integrated for a living.

Curious to see them. I am mostly (everyday) Ravenna user, but employ DVS and other Dante based audio somewhat regularly. For living.

 

[Ifrit]

If DVS was just as good those devices wouldn’t exist.

Any other [physical] device on the network can be PTP Master. The rest of the network is slaving from PTP Master, or 'leader'. Software player on PC doesn't really care what is the clock master on Dante network, as long as it spits the data when needed. It will be in sync with all devices on the network, if using DVS as an interface.

 

[Ifrit]

Well I’ll be surprised if video on Jriver stays in sync with the audio with DVS.

Can JRiver use ASIO for audio playback? Then there shouldn't be any surprises.
VLC with CoreAudio on Mac works perfect through Dante.

 

[Ifrit]

You tried with streaming services?

Spotify, Tidal, Youtube, Vimeo, medici.tv. Any audio-video source is fine. Didn't test for bit-perfect playback, but it does play.

 

[Keith_W]

Can JRiver use ASIO for audio playback? Then there shouldn't be any surprises.

Yes it can.

Since the last few replies in this thread, I have had some time to think about the Dante solution. I started off discouraging my friend from using it, I said that everyone uses a conventional solution - AVR, or AV processor, or computer + multichannel USB sound card for a reason - it's because it is known to work. I am inherently conservative and risk averse, and I only recommend solutions that I have seen working for myself.

After reading the replies in this thread, especially @Rja4000, I am blown away by the power and flexibility of Dante. If I understand correctly, any combination of Dante devices can be aggregated together, and DVS will present to Windows "I am a device with 80 DAC channels, 80 ADC channels, 200 digital outputs, and 40 microphones". And the variety of Dante devices is massive! There are multichannel power amps with built-in DAC's like the Ottocanali, all kinds of mic preamps, mixers, and DAW's, and powered speakers. For someone who has been in consumer audio all this time, this is as if a whole new world has opened up.

Although I started this thread with the intention of helping my friend, I am now mulling about the possibilities in my own system. I could replace my 8 channel DAC, two pairs of monoblock amps, and one stereo amp with a single Ottocanali. The idea of gaining all that simplicity, reducing the box count, and lowering my power bills whilst making money from the sale of extra equipment is very tempting!

 

[Keith_W]

Only problem with the Powersoft gear is the DAC’s and amps are mediocre at best. The Ottocanali 8K4 has a SINAD of -62dB. This means it can only resolve 10 bits. Crosstalk Is only -65dB, and who knows how much IMD. If that’s fine for you they should work fine.

I wasn't aware of that! -62dB is vinyl territory! Do you have a link to an independent review site with those measurements?

 

[FrantzM]

ASR Rules!!

The level of knowledge and willingness to share is off the charts...
Not a solution I would consider but the thread is worth following.
I would like to see that kind of Audio and Video over IP/Ethernet in the home too. A bit biased: I am an IT person.

Happy holidays.

Peace.

 

[Keith_W]

Their own datasheet claims that.

It appears you are right. Ottocanali datasheet.

Using the formula SINAD = 20 log10 (THD+N/100), we get a SINAD of -66.026dB (assuming THD+N is 0.05), or -61.938dB (THD+N of 0.08%).

 

[Ifrit]

DVS will present to Windows "I am a device with 80 DAC channels, 80 ADC channels, 200 digital outputs, and 40 microphones".

DVS is limited by 64 I/O at base sample rate, half that at x2. As long as you are within those numbers, yes, any and all devices connected/configured through Dante Controller are visible to Windows as one ASIO device, DVS.

 

[Keith_W]

Please forgive my stupid questions What is the advantage of using a sound card like that over, say, using the PC's RJ45 port into a switch to connect to Dante? I understand you are a pro and do Dante installations for a living. Is there any advantage of using a sound card like that for a simple home audio setup?

Also, if my friend decides not to choose the Ottocanali, do you have an alternative product to recommend that is similar? i.e. multichannel DAC built in to a power amplifier.

 

[Keith_W]

Rather than start another new thread, I thought I would ask this question here since it is kind of relevant.

Re: SINAD of -62dB.

It occurred to me that the above SINAD is the sum of the DAC in the Ottocanali and the amplifier stage. I think it is impossible to make a modern DAC with a SINAD that poor, unless we are talking about some kind of weird misguided "high end audio" design. Most likely the DAC performance is adequate, maybe not state of the art, but definitely not -62dB bad. Maybe the poor SINAD comes from the 1000W Class D amps.

So I looked around other high powered Class D amplifiers and I noticed something curious. ALL of the amps I looked at with power output 1000W or greater had SINAD's of -62dB or worse! The only amps with excellent performance and Class D were much lower powered, like the Benchmark AHB2 amp at -113dB but making "only" 185W.

So my questions are:

1. Higher powered Class D amps = worse SINAD? I imagine that they have to bridge several Class D modules to produce this kind of power output, or use a different module altogether with higher power and lower performance. Is this par for course if we want high power from Class D (or any amplifier really)?

2. Do you think my theory that the poor SINAD of the Ottocanali can be blamed on the amplification rather than the DAC?

 

[Rja4000]

Higher powered Class D amps = worse SINAD?

Not really.

NAD C 298 power amplifier Measurements

Sidebar 4: Measurements

www.stereophile.com

Hypex Nilai500DIY Amplifier Review

This is a review and detailed measurements of the Hypex Nilai500DIY class D amplifier module with PS500DIY power supply. Parts were provided by the company and member @Rick Sykora was kind enough to assemble it: As indicated, the single channel amplifier is on the left and the power supply on...

www.audiosciencereview.com

...

 

[Rja4000]

Do you think my theory that the poor SINAD of the Ottocanali can be blamed on the amplification rather than the DAC?

Maybe.
But it might be linked to the (bad) use of the DSP as well.

 

[Keith_W]

Thanks @Rja4000 but both those amps you linked to are far less powerful than the Ottocanali. The NAD is 320W into 4 Ohms, and the Hypex 500 is 500W into 4 Ohms. The Ottocanali claims 1000W per channel into 4 Ohms.

All amps >1000W I looked at have similarly dismal SINAD. I don't know enough about Class D amps to know if this is a fact of life for these types of high powered Class D amps or whether it's just this one!

 

[Rja4000]

The Nilay should probably give you 1kW per channel @4 ohm once bridged. (TBC)

Anyway: https://www.hypex.nl/products/amplifier-families/ncore-family/nc2k-oem

0.002% THD+N at half power is 94dB SINAD at 1250W/4 ohm

And that's >100dB SINAD at 5W

Apollon Hypex NC2K Monoblock Amplifier Review

This is a review and detailed measurements of the Hypex NC2K based amplifier from Apollon. The unit was kindly purchased new by a member and drop shipped to me for testing. The NC2K has different options and starts at 1.990,00€ or US $2,342 as of this writing. The version I have has the...

www.audiosciencereview.com

And this one should have "enough" power, I suppose ?

All amps >1000W I looked at have similarly dismal SINAD.

All amps >1000W you looked at are using internal DSP, aren't they ?

 

[Keith_W]

All amps >1000W you looked at are using internal DSP, aren't they ?

Yes. Something is to blame for the poor SINAD, and I wasn't sure what it was. You think internal DSP might worsen it, even if it's set to bypass?

I am even more concerned about the performance of the ADC in those pro amps. Many of them have an analog input. I think for some of them, e.g. Powersoft, analog inputs go through the ADC, then through DSP, and then DAC again. I suspect the published SINAD is for digital inputs only, and if you use the analog inputs the performance might be even worse.

I badly wish Amir would review one of these units