Best try based on a listening test, people here claim to hear differences even using the digital out that is measured bit perfect. Which does not imply that they are right or wrong, people hear what they want to hear.

Can you confirm that if the output is set for fixed volume there is no digital processing by either unit?

Fixed output is bit-perfect to the source. 

Thanks

Use of Variable Volume on the digital output always involves some processing, whether on Connect or Port. Simple volume adjustment of course changes the samples from their original input values, never mind using EQ.

Along with most such devices, Connect had some dynamic range limiting on its digital output to avoid clipping. Port does too, but for some reason it’s evidently more aggressive.

Fixed Volume bypasses all such processing.

Reviewers have remarked that Port’s analog output is also a suitable way of avoiding the effects of Variable Volume on the digital out. Port’s DAC is based on the AK4490, which is a well regarded “premium” chip.

Sorry to reopen this can of worms….

To clarify the use of Fixed Volume to obtain “bit perfect” output. Does this also require the volume limit to be set to 100%?

A quick test shows the volume limit reduces the analog and digital outputs together so i’d assume it’s doing some processing via DSP. If so, would expect that the limiter should be reset to 100% to avoid this.

If the volume variation is applied at the digital signal level (pre-DAC) I can’t see how the use of the analog outputs would bypass the problem (if there is actually a problem to bypass, i’m not sure there is anything audible) as some reviewers have suggested. Surely the only difference is that the signal is going via an internal rather external DAC.

I like the use of a fixed volume with a 66% limit to give a less sensitive dial on the external amp. However, analog attenuation might be “superior”, technically speaking.

Although I’ve never analysed its effect, one would logically expect a volume limit below 100% to simply scale the digital samples down linearly. They would obviously no longer be bit-perfect to the source, but I can see no reason why the soft-knee protective limiter in the DSP would be engaged. It’s there to stop samples being scaled up over Full Scale and clipping on the output. That can’t happen if there’s no EQ. 

I’m was in the same boat so I bought a Sonos amp to replace a zp90. Seems it is bit perfect to the source too. It sounded good even with my B&Ws from 1999. 
However, just got a new iPhone and casting via iPhone / AirPlay 2 sounds stellar; way better than the Amp as a source.
How is this possible? Is there a default setting messing up the sound from the Amp alone? 

I’m not sure I understand. The Sonos Amp cannot be ‘bit perfect’ as such. It contains a DAC, and EQ, and drives analog passive speakers. 

Are you using AirPlay2 to stream from the iPhone to the Amp? If so, it’s clearly part of the chain, and is doing decode and DA conversion much as before. Perhaps the source you’re comparing with AirPlay2 was in some way compromised? 

In bit perfect, I mean the bit before the digital stream hits the DAC in the amp.

Im using Airplay 2 to stream from the phone. So why in the name of all that is holy does it sound better? Not possible surely?

Just to clarify I’m comparing Amp playing files from a NAS to iPhone streaming files via Airplay 2 to the Amp.

It doesn’t sound different, it sounds way better.

Again, just to clarify, Airplay2 is lossless and sends the digital stream from iPhone to Amp?

So… the NAS files are in what format? And is there any EQ being applied by the iPhone before transmission over AirPlay2 (which as far as I can make out is a lossless transport)?

By the way “way better” can be explained by all kinds of factors: such as a different mix/mastering of a piece of music, whether it’s simply louder, or whether the listener believes it’s “way better” (the technical term is confirmation bias).

Files are flac. 
The difference goes beyond EQ - you can hear each individual instrument, soundstage is far more pronounced, lyrics are easier to interpret etc etc.

I was using an iPhone 6s Plus the same way previously and it just sounded the same as the Amp.

Bizarre. And I was thinking all you need is Shannon/Nyquist and 16/44 and you had the original analogue sound.

For the clarification I’m comparing the same files played  in both ways. Never expected it, just carries on as before and noticed immediately how much better it sounded. Does that excuse me from bias?!

AirPlay2 from iPhone 6s sounded the same as the NAS files, but with a different phone it sounds “better”? To me that sounds like Apple is doing some audio ‘enhancement’ processing.

If you prefer things that way, just stick with it. 

I thought that. Looked at Amp EQ and it’s flat. Looked at player EQ on the iPhone and it’s off.

I thought I knew that all bit perfect streamers would sound the same, now I’m not so sure. 

PS whatever enhancement it is, it works a treat.

You said the Amp sounded the same, when comparing playing the NAS files directly or via AirPlay2 from the previous iPhone. With a different iPhone it sounds ‘better’. Being a simple fellow, to my mind the obvious conclusion is that the later phone is doing something different, assuming it’s the same content in all cases. 

I agree. But what? It makes no sense if it’s better components / chip etc as it sends a bit perfect digital stream - it sends the data on the NAS to the Amp’s DAC.

And there’s no EQ.

Beats me. I’m no iPerson, but if phone A sounded different from phone B -- and both were delivering a  stream losslessly -- I’d assume some processing was going on somewhere at the phone end. 

Perhaps it’s the Amp end? Sliders on EQ set to 0, but iPhone EQ is switched off. Surely that wouldnt make that much difference?

Assuming Airplay bypasses the Amp EQ of course. Probably doesn’t!

Just observing this thread, I’m curious as to what App player you’re using to AirPlay your tracks, so I can perhaps try this too… thanks.

Plexamp. Unfortunately it costs money.