anybody replaced their Connect with a Port?

Don’t you think that points to your DAC(s) as being the source of the ‘problem’, such as it is?

Again, that sounds like the DAC.

I had ZP80s, from 2007-08, and they’ve all gone now. 

@brotherharry In terms of the two Arcam inputs I meant that if you felt the case swung in favour of the Connect when it was plugged into input A and the Port into B, you could put the Port into A and the Connect into B and run the test again. See if you felt the same way. 

The only true test is to capture the digital outputs of the Port and the Connect. Compare them -- literally a file diff -- and also compare with the WAV equivalent of the original file. Recording an analog waveform is rather meaningless. 

@Finbow The idea that the Port’s digital output could be collapsing the soundstage strains comprehension. Are you suggesting some kind of analog crosstalk, between two entirely separate sets of digital data? As I type I have a set of Sennheiser HD650s clamped round my noggin, fed by a small Schiit stack hung off a Port’s digital out. Lack of separation? Er, no way. 

Or was your complaint about the analog output? 

Hey ratty.

I’m isolating at the moment so no helpers to do blind testing unfortunately. I was trying to get around that by splitting my listening sessions up and being repetitive with sections. Also varying speaker and headphone outputs. 

Good point on the Arcam inputs, I was using the (as labelled) Blu-ray and Personal Video Recorder inputs. Will ask them if there are any hardware/software differences between the two. Physically hot-swapping the patch cables in the same input coax channel would be ‘purer’ approach. I didn’t do it this time round for the ease of source switching via the remote as opposed to crawling around constantly behind the rack.

Agree it doesn’t make sense that a pure bitstream should ‘sound’ different between two devices if all Connect/Port does it pass through to the DAC on the Arcam in either case. I’m convinced I did hear qualitative differences however. 

Maybe the only true test would be as you say capturing the analogue output to speakers and comparing those waveforms? Maybe run a cable from headphone out to computer input and record to something like Audacity to compare waveforms?

Something hinky going on either way. 

Really interesting thread @Finbow and @ratty.

I ordered a couple of Ports to replace my two Connect units and I’ve just spent a couple of days A/B testing Connect vs Port. I was really interested to hear them side by side. 

I use the Connect units to stream my ripped-from-CD FLAC collection off a PC to a couple of hifi system setups: one in my office (Marantz NR1504 feeding B&W 685 S2s on stands) and my main listening system in lounge. (Arcam AVR450 feeding bi-wired Mordaunt Short Mezzo 6 floor standers).

I saw the WhatHifi review after I’d placed the order but wanted to try and keep an open mind so went pretty hardcore on my test process.

• I wired both Connect and Port using the coax digital out into two separate but equivalent inputs on the Arcam.
• Both inputs were configured in amp to precisely the same neutral settings, no processing or dynamic stuff  applied.
• Fixed level output on both Sonos units. No EQ applied anywhere.
• Source files were all CDs ripped to FLAC.
• I grouped the Port vs Connect ‘rooms’ into one and ran the Sonos playlist.
• Listened both live through speakers, but also repeated up using a semi-decent pair of over-the-ear headphones plugged straight into amp.
• I’d identify sections of tracks with interesting stuff going on in them and then switch sources on the amp to change between Port vs Connect, listening in parallel with fast switching and then repeating longer sections in serial.
• I took breaks between listening sessions to avoid ear fatigue.
• Where I had impression of one source being 'better', I also tried taking the volume down 1db or 2db on that source to compensate for possible simple volume mis-perception. 

Test notes

My audition tracks are long standing, really familiar pieces that I’ve used over the years to test systems out. They’re well engineered (in some cases Grammy winning), good tests of articulation, control, presence, bass handling and musicality.

1. Muse - Reapers
   Section 2:38 to 3:38. The solo is noisy with a lot of powerful stuff going on. Port sounded woolier, less articulate and in control.
2. Bomb the Bass - Switching Channels
   This is a busy, bass heavy, rhythmical track with a lot of complex production.
   Section 0:21 to about 0:30, the ratchet sound in left ear. Connect sounds brighter, more presence than Port. 
   Section 3:50 to end, listen to articulation of the guitar lead, much cleaner and resolved on Connect.
3. Tears for Fears - Woman in Chains
   Section 1:47 - 2:20, listen to the pizzicato synth sounds in centre right background. Complex, musical, light needs articulation
4. Tears for Fears - Badman's Song
   Section 5:03 - 6:23. Steel guitar twang at 5:05. Bass guitar melody from 5:20 onwards has body and articulation. Bass licks at 6:23 and 6:35. Separation, articulation, control.
5. Brian Culbertson - Get it On (Live from the Inside)
   Organ break Section 3.07 - 3:45 listen for articulation of brass centre left and the organ not sounding woolly it’s tough as it’s so rounded. Trombone break 3:45 - 4:14. Listen for the articulation of the second guitar in left ear (yes there is one, not the rhythm on right)
6. Open Up - Leftfield
   Section 0:36 - 1:06, bass line articulation, the rhythm train track chugga is tight (there's a little click on the last 16th beat before the 1 and 3 count that’s clearer on Connect). The reversed gong sample just before the vocal should be clear too.
7. Are We Really Through - Ray LaMontagne
   Intro, should feel like the guitar is in the room. Alive. Presence. Steel guitar 2:52 to 3:32 should be resonate richly but cleanly.
8. Shape of my Heart - Sting
   2:19 harmonica solo, listening for the main melody on acoustic guitar underneath, should have enough separation to pick out. Port found this harder and easier to lose it. 

My Conclusion

If you’d never heard a Connect, you’d probably be happy with Port. But, for me, the Connect’s ability was demonstrably superior.

There were consistent, subtle and to me important quality differences. The thing that kept coming up for me was a sense of articulation and control. The Connect consistently managed to lift out instrumentation and present the music with more control and separation, particularly with complex pieces. I could pick out more background instrumentation, identify more in the tracks at varying points of the mix. In lighter, sparser tracks it was articulation, space and presence (musicality?) where the Connect always seemed a step above the Port.

A case of older and wiser, newer is not always better it seems.

Now there is the argument that I should accept a small reduction in quality for as yet unspecified S2 future capabilities of Port. I’m not sure I can though, knowing I’m missing out on the final 10% that counts for everything in HiFi and loving music.

It was a disappointing result really and I’m intending to send the Ports back. 

As everything is software these days, maybe there’ll be a second gen Port with improvements. Think I’ll have to hold back my upgrade until then as if it aint broke... 

Footnote

Great music on great HiFi should be cherished. Giving myself the excuse to just sit and really listen was a fabulous exercise. I reconnected with and fell in love all over again with some wonderful tracks. I’d thoroughly recommend a dedicated listening session to anyone looking to fill a few hours these days. Nile Rodgers is right, it might be a trap, but getting lost in music is a wonderful thing and the whole point of investing in HiFi.

Well, bare with me! Think I’m going to have to start a new thread called ‘ Sonos Port an apology!’.

Had to send back the Connect I bought from eBay as it was faulty. Hadn’t sent back the Port yet so dug it out for one last try. It updated when I plugged it in & now it sounds wonderful! Don’t know if, it was confirmation bias all along or whether simply rebooting the device triggered something but I am now 100% happy with the Port! 

What Hifi are talking out of their backsides!!

Cheers for all of your help,

Totally appreciate what you are saying & it could be I guess but the difference is really marked to my ears. Tried the analogue outs & the Port definitely wins that battle. I’ve even tried variable out & setting the volume on the connect slightly lower. It’s space & separation that’s the difference. It’s like I’m listening to a more compressed version of the track. There are times that I forget I’m doing the test & just got lost in the music. Every time that happens without fail I’m listening to the connect, it’s just more ‘open’. Going to use the port for a few more weeks & not listen to the connect. I may just get used to the sound of it? Might just be that I’m so used to the connect sound that the Port sounds ‘alien’. Who knows?! 

Absolutely no doubt about it. The connect sounds significantly better than the Port via Digital out. Instrument separation is nowhere near as defined the soundstage is much narrower & overall sound seems less dynamic as a result. Had both set to fixed volume & difference is obvious. The sound is definitely more compressed somehow coming from the Port. Is it possible it’s being ‘throttled’ somehow? Will they ‘open it up’ with S2 in June? Or is it possibly just cheaper parts? The external build quality seems lacking compared to Connect. 

Hi Ratty, hope all is good?

Received my Port today plugged it in, sat down & felt underwhelmed. The sound just didn’t seem as engaging as the connect. As a comparison here is what I did. 
 

I had my connect plugged in via coax & optical both going to my DAC. Often did a quick A-B test between these & there is no audible difference between the two connections.

Plugged the coax into the Port & left the Connect wired with the Optical.

Been switching between the Port & the Connect & there is quite a difference between the two even running through the DAC. Had these set up as a group & the volume set to the exact same level in variable mode. 

The Port has a much narrower soundstage compared to the connect, sounds a lot more ‘boxy’. 

Surely the Port is just throwing out 0&1’s?

How can this be possible?

Definitely there. Had my son switching between the two & not telling me. Can pick out the Connect & The Port every time! 
 

Also, plugged the Coax back into the connect & no difference between the Optical & Coax there. 
When I’m paying £280 for a inferior product to the one I’ve already got it kind of sticks in the craw.

Especially as this is pretty much been forced upon me. 

Any thoughts? You seem like a guy who knows about such things. 

I used to use the Pre-outs with ‘Y’ connectors into the sub. I changed it to the connect outs as I was worried that somehow splitting the signal would degrade the sound. The Kef Sub is also connected to my Denon Receiver. I have a switched input so can use the same sub with my AV set up & the Nad.
 

So, if I keep the volume below 85% in variable mode the output into the DAC will be the same as using ‘fixed’? Or is ‘fixed’ mode the absolute best way of feeding the DAC? 

As the Nad doesn’t have a remote the convenience of using the App for volume is one that I’m loathe to lose. 

I use variable volume on the connect at the moment. I use the coaxial out to my DAC/NAD then use the analogue outs to my Sub. Set the volume of the Nad to about ¾ & then use the volume control on the app. Works really well. I thought there might be a slight delay when streaming through the DAC but it syncs with the sub beautifully. I know technically I get a ‘bit perfect’ output in fixed mode but that would discount me using the Sub in this way. 

You may be recalling a fault on some models of Connect a while back where a soft knee limiter was found to be in use for Fixed volume (which is clearly wrong). As I understand it this has now been corrected. 

Soft knee limiting (compression) with variable volume is not uncommon, particularly with upstream EQ. It’s to avoid the much more unpleasant distortion that would arise should hard clipping occur.  

The 3020 was well known for being able to deliver a heck of a lot more than 20W into difficult low impedance loads during transients, assisted by the soft clipping feature. I had one back in the 80s and, even when I’d replaced it with more expensive pre and monoblocs, I’d still dust it off for party music. Pretty much indestructible.

Good luck with the Port into your Arcam Rdac. A small tip: either run the volume in Fixed mode or if you use Variable keep the volume below about 85-90%. Above that the Port will begin to apply soft-knee limiting on the very loudest digital samples, which can under extreme circumstances introduce just a smidgen of harmonic distortion.