Some facts about Sonos that are relevant to this theory.  With the RCA inputs you can use on various Sonos products, there is a slight delay between when the data is received by the unit and when the audio actually plays on the device or any speaker that’s in the group.  This is need to convert the analog to digital signal, but also to allow for buffering.  I think of the buffering sort of like a group of joggers where some members go slower than they are capable so that the slower members of the group can keep pace. When you have a ‘digital source’ such as optical input or HDMI-ARC, the speaker that’s connected plays the audio as soon as possible to play in sync with video on a TV, but any grouped rooms are delayed for required buffering. 

So, following your theory of having the S2 system just pickup the digital stream from S1, if that were even possible, it would need to buffer what it receives so that the speakers in the S2 system could all play it in sync, and thus would be noticiably behind your S1 speakers. Maybe the single S2 speaker that picks up the S1 stream could play it faster and the lag wouldn’t be noticable, but not the entire system.

You maybe wondering why something like Airplay can be used to play audio on both S1 and S2 systems at the same time?  Because airplay has a rather large buffer built in at the source.  It will send out the data and tell it to play at a timing a bit further out so that every speaker has a chance to get the data and play at the same time with less coordination required.  And if there is video involved, it can force a delay for that too.

Also, important to know that Sonos made a significant change to how devices in a S2 system communicate to each other, in order to better facilitate hires streaming audio.   Sure an S2 device could sort of do some translating between the S1 language and the S2 language, but it’s still another bit of processing that would have to be done.

Sonos employees are clearly marked as such.

For me, it’s more of a black box knowledge.  I can accept that it works this way or that way without knowing all the details of what bits are going where at every instance.  And something, like playing TV audio in groups out of sync, is clearly something Sonos would want to avoid if it was technically possible and reliable.

Which part of the above would you say is wrong? Buzz has desscribed a scenario below which seems much the same as what I describe in the first paragraph.

The second paragraph seems like common sense, why would Sonos have reinvented the wheel when moving to S2 - they may have added extra stuff to support newer technology but do you really think they would have thrown away years of work and started from scratch?

Standing by it does not make it helpful!

On a S1 system with several zones playing the same media it seems obvious that the master unit must be “streaming” whatever is playing to the other zones. I have called this an S1 Stream. Seems perfectly reasonable if perhaps simplistic.Are you suggesting that the master unit doesn’t stream media to the others?

I agree with what you are saying, but given that Sonos probably started S2 with the guts of the S1 code in there it would have been relatively simple for them to have provided backwards compatability in some way, whether this was by emulating an S1 system (albeit with limited features) or otherwise. This would be simplified by the fact that the S1 protocol is - as you have said - not going to evolve.

What I am suggesting is that not doing so was a “business” decision rather than a technical one. Given that providing a means for S1 users to add new S2 only equipment would allow them to buy new equipment from Sonos, they must have decided that the revenue from this would be less than the revenue from people binning all their S1 kit and buying full S2 systems.

Nothing has to be binned. This is surely the most common piece of misinformation on this most common of troll topics, the S1/S2 split. It was 2 years ago. It's not going to change. This thread is a complete waste of all our time and I'm not wasting any more of mine..

Or it actally was a technical reason, and Sonos had to give up on S1 people buying more Sonos devices or cease growing as a company.  Because that's exactly what it was.  As @buzz said, each unit has to communicate with all others, meaning each device must be recognized by every other device.  And there's no way for S1 devices to know how to recognize a device which didn't exist when S1 was frozen.

Otherwise, if Sonos really wanted to force people to upgrade, why allow S2 capable devices which existed at the time of the split to remain on S1?  Why not force all S2 devices to ugrade?  

If you want to ponder the wisdom of perpetual backward compatibility, hackers can easily exploit some ancient code that has been brought along. A prime example is the SMBv1 protocol required by S1 to access NAS (Network Attached Storage) drives. Hackers love this protocol. Moving away from SMBv1 requires (at least) more RAM be installed into the S1 player because its operating system kernel must be replaced. This cannot be done in the field. The trip to a service center could easily approach or surpass the cost of a replacement unit. Since very old units might begin to fail due to unrelated issues, this upgrade would be a questionable investment.

Sure, a few DIY elites might be able to handle some of this at home, but only if they have surface mount soldering skills and tools. And, this assumes that new IC’s have the same pin-out as the IC’s they replace.

“Backwards Compatibility” is a great marketing claim, but our global computer security could be greatly enhanced if we would trash everything and start over, rather than attempting to maintain compatibility with ancient hardware and software that was never secure.