My experience is pretty much novice compared to most of you guys. I was reading up in the PI2AIS Hat and was wondering if it would be worth integrating into my existing Pi4 setup.
Currently running Roon Core via a dedicated NAS
Which pushes storage FLAC files or Tidal via RAAT to Pi4 connected to a Khadas Tone Board as Roon endpoint.
Tone Board outputs analog to Cambridge CNR120 in Amp mode only feeding Anthology DYI) speakers
Would there be any benefit in purchasing a PI2AES Hat for the Pi4 then feeding the Tone Board via S/PDIF, if I can get the tone board to accept and operate with that connection, from a sound perspective?
Additionally, would this hardware configuration still work as a Roon endpoint?
Read up on several forums on the PI2AES but still am cloudy on Roon integration and benefit.
Thanks for your input.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
2
I’d find it hard to imagine. You’re pushing the same bits with USB and AES. AES is an older protocol, though, and it’s one-way versus two-way with USB, so the receiver (the Tone Board) can’t adjust the clock frequency of the sender with AES, as it can with USB. I realize there are some Pi2AES enthusiasts on the forum, but I don’t understand what they see in it, unless it’s the “dancing bear” thing – the dancing isn’t great, but it’s amazing it dances at all!
Far enough.
I was thinking, “how could it make sound better?” or “$150 better not counting implementation” and it wasn’t really clicking for me logically.
I’ve done blind tests of USB vs AES3 with several DACs over the years. With some source/DAC, the subject could hear a consistent difference; with other combinations they could not. All of these are electronic circuits that may be able encode bits equally well, but have different timing and other electrical characteristics at the interface between the source and the actual D2A circuitry — you can’t avoid physics. Our most recent such test was between Pi2AES AES and Allo USBridge+Sbooster LPS USB into a Schiit Yggdrasil. Small but consistent differences made the subject prefer Pi2AES AES. After that test, I confirmed the impressions for myself, but of course that’s not a blind test any more.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
5
Well, again, that’s between two devices I can’t imagine the need for. Adding stuff to a system that performs no obvious function tends to degrade it, not improve it. The Allo USBridge is another thing I don’t see the point of – just use the USB port that’s already built into the Pi, I would say.
To be sure, some extra measure of electrical isolation may be necessary for crufty old DACs which don’t perform this function well, and by the same token there may be DACs which handle S/PDIF well but not USB, as USB Audio is a more complicated protocol. But there’s no good reason for a well-designed DAC to suffer from either of these deficiencies.
I tend to listen to music, not various bits of hardware, so maybe I’m not the target audience for either of these devices. But in any case, I wouldn’t waste my time listening to a Shiit Yggdrasil:
there are fundamental problems in Schiit multibit DACs when it comes to doing what they are told to do: convert digital samples faithfully to analog. Whether we use balanced or unbalanced output, problems remain. And not just in one measurement but practically everywhere we look.
So given the disappointing engineering ability evident in this device, perhaps it benefits from one or both of these devices. It wouldn’t surprise me.
This would be more useful if you reported measurements or listening tests you have personally conducted rather than repeating online sources whose accuracy you have no way of evaluating. The particular source you linked to has been amply debunked by professional audio measurement engineers, if you care to search the Web.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
7
Searching the Web is a sometime thing, Fernando. I’m certainly willing to expand my horizons, if you’d care to provide some links.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
8
I’m not sure why one stranger’s reports would be more trustworthy than another’s, but in any case this isn’t what I do. I tend to rely on reports from others who enjoy that pastime.
Fernando - So you indicate there is a positive difference using the PI2AES with the Schiit yggdrasil which is a $2.5K. That DAC is way out of my price range. I mean even if the funds were available, I doubt I would lay it down for a DAC. Which makes me a cheap music listener.
My inquiry on the PI2AES was to just understand the application from a hardware perspective and if it would make a difference in my current setup or standalone.
Reading into your statements, it seems you indicate to have made improvements on some DACs and not others.
In my case, I would not even tried the Pi2AES with my Yggdrasil if I did not have the Pi2AES already for a different purpose. Since I had it, the minor improvement we heard was worth getting a dedicated Pi2AES given the overall cost of that system. But even then the difference was small, I can’t see the value of getting the Pi2AES unless you have strong reasons to believe that your DAC’s USB input is poor, which is unlikely for recent well-designed DACs (old DACs is a different story). So, in this I agree with @Bill_Janssen.
Thanks for your feedback Fernando, have a much better understanding now.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
14
This may be a bit over the top. The source I linked to was a response to the measurements by atomicbob (which, by the way, thanks for the pointer to). Not sure how you can “debunk” a source in advance.
Well you never know, could be another dancing bear, lol.
I have a Cambridge CXNv2 that sounds great to me. However, I am building some new speakers for the living space so will be moving existing speakers in the mancave. So was looking for an affordable Roon end-point and a DAC. There was a lot of good forum talk and reviews of the Tone board. So, RPi4 and tone board seemed like an affordable option.
I have no regrets. It isn’t as good as the CXNv2 IIMHO but not far away either. This says much considering the $200 price tag for both the RPi4 and Tone board.
Was just looking for a way to squeeze a little more performance out of it. However, after thinking about it, realize it’s just a different DAC and you really can’t make a DAC do everything you want. It does an excellent job for what it is.
I need to focus on planning some subwoofer builds anyway. Got side tracked with this PI2AES thing.
There were several sets of measurements from different people for different versions of the Yggdrasil (A1 vs A2). A legitimate concern for most resistor array DACs is that they take quite a while to stabilize thermally. Good measurements require waiting for that, which was part of the argument. Anyway, it’s old story, not relevant to the OP’s concerns.
Does anyone have any suggestions on what to use instead of a Pi2AES since it now seems discontinued and replaced by the Mercury? I use RopieeeXL and I care about rock solid AES and USB output, but otherwise don’t need extra features. Thanks.
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
20
The S/PDIF protocol is almost the same as the AES serial protocol, but they are both antiquated. I just use the USB ports on the Pi, myself. It’s a short run from Pi to DAC.
But because of the demand, which certainly took us by surprise, the PI2AES will remain in production. The new price for the PI2AES will be $249. We will begin pre-orders in late November or early December once the current batch of PI2AES units have been shipped.
Whether USB, AES, or I2S depends on what your DAC does best with. In theory it should not matter, but much experience, which included blind tests, has shown me and many others that it does. While the bits are transferred correctly no matter what input is used, the electrical characteristics of each receiver circuit interact differently with the analog side of the DAC, even in fancy DACs.