she (the support engineer) did not give specifics - and I did ask her if the engineering documentation indicated what side the problem was on - the PC side or the HomePod side. ***
That said, I have pcap files (Wireshark) from the PC indicating the PC is sending the audio stream to the HomePod...all of the UDP packets in the capture are coded RED - but the flagging of the packets SEEMS TO BE (my emphasis) related to the TTL (Time-to-live, usually router hops) of the UDP packets being set to 2. That SHOULDN'T BE (again, my emphasis) a problem, as the PC and the HomePod are on the same subnet, so there should be no decriment of the TTL before it hits the HomePod. I can in fact see the HomePod in the ARP table of the PC, so IP-to-MAC resolution appears to be working as intended.
Waxing philosophical:
First, I find it incredulous that IF this is the result of the HomePod firmware update (to ver 27) it was not caught in regression testing before the product (new HomePod firmware) was released. If this is indeed the case, the entire product team should be dressed down, and the release manager ought to be put on some sort of PIP (performance improvement plan).
Generally, I love Apple products - they, again, generally, just work...until you introduce a device (like a PC) into your ecosystem that isn't Apple. I've been in the IT industry - working in very large enterprise environments, for over 30 years. And every time I've been involved with introducing an Apple product into an enterprise environment, it has been a major headache. My experience consistently has been that Apple tends to play fast and loose with their implementation of industry standard protocols, usually after being somewhat late in implementing them at all. This isn't meant to be a discussion about whether the Apple way is better than the industry standard or not, but rather that when eschewing industry standards for your own methods of doing things tends to result in issues like this.
[Edited by Moderator]