Hi Paul, I have an older generation macbook that was working fine with Ant+ dongle and my previous Neo 2T but through various forum suggestions it has been a process of elimination. Zwift support suggested this companion setup instead of ANT+ because it is apparently more stable. On the Wifi issue the same, 2.4Ghz was suggested instead of 5Ghz. I have not tried the suggested deactivation of Race mode which is my next point of call.
Rich I did an hour test on the Garmin trainer software and had no dropouts at all which pointed me to Zwift and Click & Cog which I have now eliminated as Click & Cog hardware being a problem.
I would test without pairing via Companion - pair directly via Bluetooth to the Mac and see if you get a different result. Whether or not the Mac has Bluetooth 5 hardware will depend on its age. If it’s newer than 2018 then it probably supports Bluetooth 5 and might work better.
Testing with everything paired via Companion is a fine idea, it’s just not the optimal setup if your WiFi service is not great. If going to 2.4GHz WiFi is necessary, that implies that signal quality in your training area may be poor, since you might be depending on the superior range of 2.4GHz. Personally I would take that as a sign that I should work on improving my home WiFi coverage.
Testing in the Tacx software doesn’t really tell you much about whether your setup is adequate for using virtual shifting which changes how the trainer interacts with the software and eliminates a Bluetooth connection. It also doesn’t pair via the same pathway through Companion, so it eliminates the use of WiFi as the pairing bridge.
Thanks Paul, the Macbook is a 2014 hence trying to eliminate the age as a cause but as I say it was working well before I bought an entire new setup (Tacx Neo 2T and C&C), I will use my day to day M1 Macbook to eliminate. Home wifi is great, it is a Mesh Wifi 6 with strong signal in the paincave with a 400mb data connection measured at the Macbook. Will can the 2.4GHZ network and create a 5GHZ dedicated guest wifi as the following test.
Sounds like a good plan. The 2014 MacBook will have Bluetooth 4 hardware, and that is known to be inferior at managing multiple paired devices, it has half the bandwidth of Bluetooth 5, and its frequency-hopping support is inferior as well so it is more vulnerable to interference. The M1 is much better for Zwift in a great many ways beyond just Bluetooth.
Something doesn’t add up.
My garage setup current is an A1398 Macbook Pro (Early 2013 model) with a Tacx 2T and I’ve had virtual shifting with it before, not anymore since I had an issue with a play controller going berserk; but currently using the 2T and an Alpine with steering, along with HRM, all through bluetooth, along with Airpods also on bluetooth, and I’m not experiencing any connection issues.
All through the macbook, not using companion as a bridge.
Racemode on the 2T is enabled.
Never used the clicks with the macbook though, only when I was using an iPad… not a fan of them; but I do have them for testing potentially.
Update 3 - created dedicated 5Ghz Wifi, ran Zwift off my M1 chip notbook, switched back to direct to notebook bluetooth connection and after starting the group ride for 2 minutes the power dropped out for 20 seconds. I quit the group ride and immediately started a training program on Zwift and rode for 30 minutes with no problem at all with exact same setup. Program obviously uses ERG mode so technically it eliminates the Neo2T as a potential cause. I forgot to deactivate race mode on Zwift. That will be my last test tomorrow.
Update 4 - deactivated race mode on the Taxc Neo 2T. Rode for 15 minutes in a group ride with various wattages and no power drops. Could not test longer but results look very promising, does anyone have the release notes for 2.70, I would be very interested to check them out. Will perform further testing when I have more time
Update 5 - After deactivating the Race mode on my Tacx I rolled back all the changes I have made, shared wifi, old macbook, bluetooth direct connection. All working fine.
Conclusion: The random power dropouts were caused by the racemode setting on my Tacx 2T.
Out of curiosity, do you ever disconnect your trainer from its power source so it’s fully de-powered at any point?
I use a switched outlet for my equipment, both kickrbike v1 and tacxneo2t to be specific, so they’re always fully de-powered until I end up using them, and de-powered when I’m done.
I have a going theory that keeping devices plugged in causes strange behavior.
This theory extends to all trainers.
I live in a place where keeping electronics plugged in when not using them is asking for trouble. Dirty power is real. I don’t just depower, I physically unplug from wall including ethernet. This has led to power supplies lasting for the lifetime of the trainer and a much more consistent and reliable experience. Previously, power supplies would stop working and lead to erratic trainer behavior. I am sure others have ‘cleaner’ power and this is a complete nonissue. Just keep in mind if you’re burning through power supplies and those electrons are acting strange. And, no, I am not talking about quarks. We could talk about quarks if you want to… 
and it just occurred to me that by unplugging from the ethernet between rides one can avoid those pesky and broken firmware updates. 
This is something I do as well. If the device is not in use it is unplugged. I do not leave TVs or others like that even in sleep mode.
Hi Andrew, no i don’t but will most definitely moving forward. Today’s ride was with everything reverted back and when I was in the starting pen for a group ride I had a power drop and once we started it was fine until 58 minutes into the ride. I believe your theory may have some merit because I’m at my wits end.