Computers, but. To windows. They are Linux I think and they are headless. Minimum computing just for task at hand
And that is a way to guarantee I wonāt be buying into it. With Roon, I use their software for the server but it runs on a system put out by Small a Green Computer. I can use endpoints in any price range imaginable with drastically different feature sets. I can put a super high fidelity and expensive system in my listening room, a medium fidelity medium cost setup in my backyard, a low fidelity low cost setup in my kitchen, and whatever price setup I want in my garage for when I am working on my car. Each might come from a different company. Some might have the endpoint integrated into self-powered speakers with a built-in DAC.
PS Audio canāt possibly meet the needs of the market with their own products yet still will be proprietary. This means you will have limited solutions so your appeal will be limited.
This where Roon has been utterly brilliant.
1 Like
It will be interesting to see if people move from an open system to a closed system.
My system is open, includes Roon and works faultlessly, auto-detecting any digital source. It took about 10 year of development to get the firmware to that stage. I previously used Auralic Lightning, a much more closed system, but still with good networking facilities. However, it was spectacularly good and took about 5 years to get there. It took about 3 years after launch for the networking element to be added. That said, it has never had an Android or OSX app, only iOS.
I think the reality is that people have their favourite apps for operating their systems and something has to be really good to make them change, and really good systems tend to take years to develop.
The biggest unavoidable problem is that PS Audio is developing software for a single expensive product, whereas every other manufacturer I know has a range of several products using the same software platform and often cheaper units selling in high volumes. To design a full singing and dancing open system for a single expensive unit just doesnāt make any sense to me.
1 Like
Whether people like it or not, Roon has by some standards become an industry de facto standard. Iām quite alright with the sound of Roon. Iām sure that Roon has made headway in refining their audible offering. Iām moving Roon to a dedicated i7 NUC this week and weāll see if that makes a noticeable change to the sound.
Iām still open to Octave, Roon or not. Whether itās as a Bridge III or a standalone server. I would love to have the option of Octave and Roon simultaneously. I like options and I like to tinker so that could workout.
I hope that Paul finds a way to enable a Roon endpoint on the newest hardware.
2 Likes
This is were I think Roon has failed. I have a lifetime membership and have used it since almost day 1.
In this time they have done a lot in regards DSP/Integrating Tidal & Qobuz/Roon Partners etc but I donāt think they have spent much time on trying to improve the sound. To me this should be a priority as their target market is people who want the best SQ.
Having a closed system is probably not ideal for the end user as it does have limitations but if the interface is half decent and the sound is superb I know which one I would chose.
2 Likes
Roon passes bit perfect data to the endpoints. The endpoints pass bit perfect data to the DACs. The sound quality is all about how electrically noisy the endpoints are and how the DAC deals with that noise.
1 Like
Well when I play Roon and then Audirvana from the same hardware across the same network to the same DAC I donāt get the same results. So there is something going on somewhere and I am guessing it is software related.
4 Likes
My findings also ,when rarely using audio pc anymore,I prefer to operate Roon and use radio function to find new music ,but it sounds much worse than Audirvana.
With Innuos server, same thing. Roon sounds literally crap vs streaming Tidal through Orange Squeeze.
So I am with Paul here. He wants best possible sound,so do I.
If it means to skip multifunction boxes,then it goes like that. For the same reason I dont use hometheater integrated amps. They do a lot,but sound likeā¦?? Crap 
Well, when I stream PCM to my ultraRendu using Roon and then using Audirvana with no filtering so it is bit perfect, verified using the bit perfect test PS Audio offers, I hear no sound difference.
I would like to know how Roon could sound different than Audirvana when the same exact bits get to the exact same DAC using the exact same hardware. Anyone? Before you respond, please think about it.
Here, let me help you guys. The difference is that the ultraRendu is running āRoon Readyā software when using Roon and āMPD / DLNA Rendererā software when using Audirvana. The DAC is getting the same data from the ultraRendu in both cases so the only viable possibility I can think of is the ultraRendu is putting out different amounts of noise with each piece of software.
Maybe @tedsmith could explain any other possibilities for the sound difference you guys are hearing when running identical hardware but different software. Unless of course you guys are using filters and not sending bit perfect data to the DirectStream.
No filters or upsamplers here,streams 44.1/16 as it is. DonĀ“t know about ultraRendu or how it would be different,never tried one. Just the ānormalā Matrix route.
But that is not all there is. Many use Jplay,Asio Bridge,wasapi and asio drivers ,Audiophile Optimizers,Fidelizer,Lasso,different volume settings ,etc etcā¦so it gets really complicated. Even though they are bit perfect ,so no messing with audio signal,but end results will vary.
Qobuz sounds different ( worse ) than Tidal. How do we explain that? They should sound the same,but dont.
Funny if you cant hear any difference between Roon and Audirvana ,they sound quite different 
But lets see what PS Audio comes out with,soon we will know 
Life was so simple when I took this screenshot in 2014. You tapped on a media source, a device to send it to, selected some music and pressed play.
Now there are disputes about whether the same file sounds the same from Tidal or Qobuz.
Yes, right, but a computer none the less. At the end of the day all servers use computers (ours included). The question is how is that computer configured, how does it get the data out to the renderers, and how does it sound? The computers used in servers matter.
2 Likes
A Mac Mini can run 100 apps at the same time. Analogous to a big truck loaded with 100 packets out for delivery at the same time, making loads of noise and kicking up dust. If you want to deliver just one packet with little or no noise or dust, you wouldnāt use the big truck, youād design vehicle that is lighter, more efficient and that hardly leaves a trace. You might think about modifying the big truck to make it more efficient, but at the end of the day itās still a big truck with a big noisy motor.
I canāt think of a clearer analogy as to why certain manufacturers have spent the last 10 years or more refining hardware and software to deliver just one parcel - a music stream - in the quietest and most efficient manner, with extremely good results.
A fundamental issue is the low noise output, not something that consumer laptop and desktop manufacturers worry about. Auralic developed their Orfeo output module and Innuos a bespoke internal linear power supply. Call these dust collectors, stealth units, whatever ā¦ they are what raise the game on these devices and they are relatively new innovations, the Orfeo in the G series of devices and a new power design in the Innuos Mk3 units.
Yes, but I think when you are using digital signals, issues of sound quality become logically easier. Our thinking has all been conditioned through the years of using analog signals where experience has taught us that everything matters and the best way to determine differences is through listening. I agree with @speed-racer here - with digital signals, if you compare bit perfect data on a DAC that is essentially immune to external jitter (which you can do with the DS DAC -thank you!), the only logical difference that can effect sound quality is noise that is mixed with the digital signal. This noise can be generated by upstream components and software, and/or imparted on connecting cables. It is transmitted to the DAC through those connecting cables and/or the AC distribution system. (Noise can also be created by the circuits that convert whatever āinputā one is using to I2S, so choice of input may be important).
If one can effectively filter this noise (as the Octave Server apparently will), it becomes difficult to logically explain how upstream components or bit perfect software can affect sound quality. Said differently, if there is a difference in sound quality between computers or software, then your noise filtering is insufficient.
This is why dedicated AC circuits (with the computer and audio components sourced separately), power regeneration, and galvanic isolation all improve sound quality and decrease the perceived differences by upstream variables.
Thatās weird. In my Melco N1ZS10->usb->Matrix->HDMI->DSsr setup (no filters/upsampling, with BubbleUPnP as controlpoint) Tidal always sounds worse than Qobuzā¦
To continue your analogy (sort of)ā¦ what makes more sense, what is more efficient, what is least expensive? -to construct a new āsilentā truck, develop and build a quieter engine, develop better fuel, and engineer new roads, or use the same big truck, park a few blocks away, dust and clean the one parcel, and then carefully hand carry it to the destinationā¦ all while you can prove the parcel that arrives is identical.
As it turns out itās much cheaper to build a new āsilentā truck because most of the components already exist and it then becomes a matter of designing good software, which you have to so anyway. The new components are basically the power supplies.
Just a thoughtā¦
using Melcoās ethernet output with Melco handling the IP-adress to the Bridge and delivering flac (or whatever) to BridgeIII would there be any reason this would not work?
Not even close! On my system with Aries G2, qobuz is at least as good as tidal, better on most files.
2 Likes