[Krutsch]

I feel that there is nothing to ask Roon. Protocol is UDP based. Why there is possibility to recover broken packet, there is no way to recover lost packets. The only way to improve their delivery is sophisticated network setup with managed network switch. Otherwise you can't be sure that your audio stream is 100% bitperfect. Its a fun to see audiophile zines advertising Roon. They tell nothing about network setup and packet loss.

UDP doesn't normally "drop" packets on a reasonable network (read: your home); this is more of a problem on congested data center networks. If you are grabbing a random segment and seeing PCM samples being lost, that sounds like a serious protocol deficiency or there is something wrong with your home network (e.g. bad switch or something on the network that blasting out a lot of noise).

[vklyushnikov]

UDP doesn't normally "drop" packets on a reasonable network (read: your home); this is more of a problem on congested data center networks. If you are grabbing a random segment and seeing PCM samples being lost, that sounds like a serious protocol deficiency or there is something wrong with your home network (e.g. bad switch or something on the network that blasting out a lot of noise).

Actually there is no warranty of delivery. No at all. Every network may have its own problems. I only guess why Roon went this way. They could use TCP protocol in price of small delay on start of playback. For example I have no problems with DNLA or squeezelite traffic.

[Krutsch]

Actually there is no warranty of delivery. No at all. Every network may have its own problems. I only guess why Roon went this way. They could use TCP protocol in price of small delay on start of playback. For example I have no problems with DNLA or squeezelite traffic.

I am very familiar with UDP, but I can tell you from a lot practical software development experience in networking that it's reliable, most of the time, unless you have the issues I mentioned above.

I can guess why Roon when this way: they need to stream audio to multiple endpoints that are synchronized. With TCP, you can have re-transmits and/or back pressure flow control issues that make timing across multiple receiving sockets difficult. With UDP, you just broadcast to all end-points and on a stable network it will end-up just working.

[vklyushnikov]

I am very familiar with UDP, but I can tell you from a lot practice software development experience in networking that it's reliable, most of the time, unless you have the issues I mentioned above.

I can guess why Roon when this way: they need to stream audio to multiple endpoints that are synchronized. With TCP, you can have re-transmits and/or back pressure flow control issues that make timing across multiple receiving sockets difficult. With UDP, you just broadcast to all end-points and on a stable network it will end-up just working.

Have you heard about Merging Nadac and Ravenna protocol?  It is UDP based too. They strictly require managed switch and have a list of certified switches. They strictly require 100% UDP delivery. If Roon goes this way they should provide their own list of switches with sample setup and calibration procedure. Otherwise they risk to repeat USB audio way - it works great but only in theory. Reality is far worse. Anyway let's wait Roon response.

[Krutsch]

Have you heard about Merging Nadac and Ravenna protocol?  It is UDP based too. They strictly require managed switch and have a list of certified switches. They strictly require 100% UDP delivery. If Roon goes this way they should provide their own list of switches with sample setup and calibration procedure. Otherwise they risk to repeat USB audio way - it works great but only in theory. Reality is far worse. Anyway let's wait Roon response.

Or... you just plug a USB stick into your BDP and enjoy the music, which is where I've returned to 

[vklyushnikov]

Please see discussion with Roon: https://community.roonlabs.com/t/huge-number-of-lost-pcm-samples-in-network-endpoint-output
In short they claim that their clocking mechanism does not work well with virtual devices like ALSA tee device. So this is the cause of measurement errors. Regarding UDP packets - their protocol is heavily modified and is reliable like TCP. So that is. I don't have audio analyzer hardware to check samples directly from BDP AES output, so let's trust Roon. I appreciate the responsiveness of their support.

[Grit]

So back to no explanation for why the difference in sound...?

[vklyushnikov]

So back to no explanation for why the difference in sound...?

Brian from Roon said: "I would be interested to see some measurements (or a blind ABX trial with a reasonable population) that quantified the differences more concretely". I absolutely agree. I think that Bryston could perform such measurements and publish results. They should have all required measurement equipment. I only afraid that James and Chris gave up reading this thread

[James Tanner]

Brian from Roon said: "I would be interested to see some measurements (or a blind ABX trial with a reasonable population) that quantified the differences more concretely". I absolutely agree. I think that Bryston could perform such measurements and publish results. They should have all required measurement equipment. I only afraid that James and Chris gave up reading this thread

No I have been reading it with interest and it is a very interesting subject.  Not sure there are any definitive answers though 

james

[Grit]

James, do you hear a difference between Roon and MPD in your system?

[CanadianMaestro]

Brian from Roon said: "I would be interested to see some measurements (or a blind ABX trial with a reasonable population) that quantified the differences more concretely". I absolutely agree. I think that Bryston could perform such measurements and publish results. They should have all required measurement equipment. I only afraid that James and Chris gave up reading this thread

Heck, I don't have any vested financial interests in either Roon or Bryston. Why do I need to do ABX/"blind" tests to back up my subjective impressions? Rather spend my time enjoying my library.

It's a typical manufacturer/seller's response to criticisms that don't jive with their marketing agendas.

 

[vklyushnikov]

Heck, I don't have any vested financial interests in either Roon or Bryston. Why do I need to do ABX/"blind" tests to back up my subjective impressions? Rather spend my time enjoying my library.

It's a typical manufacturer/seller's response to criticisms that don't jive with their marketing agendas.

 

Because there are pros and cons in both Roon and MPD. I was skeptical about Roon initially, but I really need better UI for managing my library and playback than semi broken open source controller software for MPD.

[CanadianMaestro]

The measurements posted earlier here, demonstrated clearly that fidelity is seriously compromised by Roon playback. That's pretty damning, if one wants high fidelity over operational smoothness and library management.  I know I do.

Not aware of a single software suite that does both to near perfection.

[vklyushnikov]

The measurements posted earlier here, demonstrated clearly that fidelity is seriously compromised by Roon playback. That's pretty damning, if one wants high fidelity over operational smoothness and library management.  I know I do.

Not aware of a single software suite that does both to near perfection.

They claim that this measurement method is incorrect for Roon. Strange that it works perfectly for MPD

[tishbite]

Heck, I don't have any vested financial interests in either Roon or Bryston. Why do I need to do ABX/"blind" tests to back up my subjective impressions? Rather spend my time enjoying my library.

It's a typical manufacturer/seller's response to criticisms that don't jive with their marketing agendas.

 8)

Absolutely bang on, CanadianMaestro.

I was going to say exactly that but you put it forth first: "Why do I need to do ABX/"blind" tests to back up my subjective impressions?".

Roon marketing lot can say what they want (though I regret that attitude: essentially they're saying "we don't believe any claims from users that our system is less than perfect; we'll only take anything seriously if you show it a+b") for me in my system Roon is heavily flawed. Do I stand by my senses? A resounding yes.

[tishbite]

They claim that this measurement method is incorrect for Roon. Strange that it works perfectly for MPD

Reminds me of VW and their own "measurements" done in-house and/or for them: look where it landed them.

Anyway, I picked up something noteworthy in Roon's response:

"(...) One notable difference: with Roon+RAAT: the media file is being decoded in the Roon Core. With MPD decoding work is taking place on the BDP.(...)"

Indeed this was my initial thought and (reasoned) observation re. the sound difference. I'd suggest the problem lies with the Roon Core and/or the underlying hardware (i.e. PC, in my case).

[zoom25]

Heck, I don't have any vested financial interests in either Roon or Bryston. Why do I need to do ABX/"blind" tests to back up my subjective impressions? Rather spend my time enjoying my library.

It's a typical manufacturer/seller's response to criticisms that don't jive with their marketing agendas.

 

I do understand the need for ABX and blind tests, like when I was comparing uncompressed and compressed lossless. On other sites, followed protocols and posted my results and still got crap that I'm just hearing things or something is broken in the system, or there must be DSP going on or it's not bit perfect etc etc.....I get it, it's a close difference between those two.

However, for Roon, the moment I pressed play, it wasn't about hearing a difference but more so of an entirely different feeling of having a seat way back. It seemed aggressive compared to what I had with MPD. The difference is still there. Once it's not that drastic, maybe I'll entertain ABX or blind testing, but as of now with how things stand there is absolutely no need in my system.

[TJ-Sully]

Question for James or Chris:

any plans in the works for future improvements to the BDP's user interface...? just curious.

thanks, Sully

[James Tanner]

Question for James or Chris:

any plans in the works for future improvements to the BDP's user interface...? just curious.

thanks, Sully

Hi

I know Chris is constantly working on it based on feedback.

james

[Marius]

I fear I must agree.
Cancelled my Roon trial today.... because of this.
Also, because in the end, Roon is no more than an interface to my own library. Nothing I can't do without Roon, either for free or a small amount.
Its an extensive interface, some might call it beautiful. But also very confusing and distracting one, and that's without counting in Tidal pushing its preferences down our BDP's.


Hope it will get better over time, and will check back when it is for sure.


Cheers,
Marius

Heck, I don't have any vested financial interests in either Roon or Bryston. Why do I need to do ABX/"blind" tests to back up my subjective impressions? Rather spend my time enjoying my library.

It's a typical manufacturer/seller's response to criticisms that don't jive with their marketing agendas.