Thank you!
I’m not sure if the oval rings are a red herring. I’ve a few dual recordings with my IRL race bike on the trainer (which has round rings) and there didn’t seem to be much of a difference. All my outdoor power PB’s are on round rings, and they line up well with my indoor power PB’s.
I’ll do a few more recordings and see where things are heading 
I had the same experience with the Neo 2T, reading about 12 watts lower than my Assiomas at around 300w. Returned and got a Kickr V5, which is spot on with two different sets of Assiomas.
Then the Kickr or Assiomas, or both, are inaccurate.
They measure at opposite ends of the drivechain and you should see some difference.
I could have been a bit less opaque about what I meant by “spot on”… at low wattage, both the Neo 2T and Kickr read about 5w lower than the Assiomas, which I attribute to drivetrain losses on the clean drivetrain on my trainer bike. Up around 300w, the Kickr is still reading about 5w lower, but the Neo 2T was reading 11ish watts lower, with progressively more negative skew as the wattage increased beyond that.
So just to be clear, we’re talking about a 6w difference at 300w between the two trainers? I have no intention of defending the Neo 2T, I’ve owned over a dozen trainers at this point, but to me you’re just chasing numbers with no actual point.
If you use power match, you have no issue. If you do a ramp test on the Neo and use that number as your FTP, no issue. There’s literally no issue with having either trainer be +/- 11w other than your ego telling someone what your FTP is (which is real, I get it).
For the most part you are absolutely spot on. It doesn’t matter what the number is, as long as 300W is 300W each time you ride. However, there are some applications where is actually does matter. Racing on Zwift in the Invitational races where you must use your trainer as the primary power source, a 3% difference between trainers is significant. A five minute effort at 400W is very different from a five minute effort at 388W. That’s the difference between being able to stay in the group to getting dropped.
Yeah, difference of 6w at 300w, but around 13w at 500w, and so on, with increasing wattage. So let me ask ya, if you’re gonna drop 1200-1400 on a trainer, and you’ve got two choices, one that consistently reads 5w lower than your pedals across all power outputs, and one that has a (presumably) linear increase in the offset between pedals and trainer, which are you gonna choose? Call me crazy, but as a consumer, when I spend that kind of money, I want something accurate. Some zwift races require you to use trainer based power, and I don’t want to have to take my Assiomas off my other bikes to do powermatch. So yeah, not just “chasing numbers with no actual point”. As for ego, though, guilty as charged.
But isn’t the loss in fact proportional to the power output and not a fixed number of watts?
TBH, there is no perfect trainer. I would drop $2000+ on a solid trainer without any issues. But, I’m on my third Neo 2T (one had buzzing/electrical sounds with super loud Zwift ground effects, one was overheating and fried the cadence sensor, third one so far so good), I had two Neo 2 before that (cadence sensor issue), and before that I had an Elite that was so incredibly slow at changing power it was unusable for TR, and before THAT I went through seven Kickr 18s because of their disastrous QC problems.
FWIW, I’m not convinced that Wahoo has really fixed their QC issues and is any more reliable or accurate than the Neo. They both have issues. You solve one thing by switching companies and end up with 2 new things. 
That’s the truth. I went through 3 Kickr 18s before I got a solid one that is working great now. The new Kickr I got for the office is working fine thus far, but I dread the onset of whatever the newest glitch will be…
My understanding was that drivetrain efficiency was a fixed number, given that frictional losses are constant, and proportional losses due to drivetrain (in)efficiency should actually decline with increasing power output (i.e., 5 watts at 300w is a higher percentage loss than 5w at 500w). I think that’s why they talk about savings for pulleys or chains in watts, not percentages of power input.
Some interesting stuff on these 2 links (one of which was already posted). I think if you see specific wattage, it’s likely given at an average output (probably something like 200w), otherwise, frictional losses are indeed proportional. As power goes up, friction and strain on bearing and the chain also goes up, so power loss goes up as well.
This is about choosing gear sizes related to friction loss, same concept applies to increasing power I think:
The smaller gears magnify the chain tension which must directly magnify the link bearing friction and thus the losses.
It should be noted that many models of cycling speeds use a default value of 3% for total drivetrain losses. The standard model used in the simulations reported in this section used the data published by Friction Facts that suggests that at the level of power output of an average-strength bikepacker, a more accurate value is about 5% (including chain, derailleur pulleys, bottom bracket, and pedals). Of course, these losses are only realized when pedaling, which is why the pie charts in the <Types of Resistance page show that only 4% of the total resistance experienced across the entire route is due to drivetrain losses.
My Neo 2T packed up after 9 months so now on my second 
What happened?
I’m getting the sudden urge to put on a new chain… 
Just stopped working. All the lights when out and no resistance. Phoned Garmin, they said it was not repairable and I’ve just received a new one
Oh no, do you have an older trainer you can use in the meantime?
I was without it for 2 weeks but am doing offseason weights so only needed bike in the gym for warm up and cool down. Gym bike is also a wattbike so not too bad!
Does anyone know when the Neo 2T is coming back into stock? I have been checking daily for two months now 