I’ve had a TVS for a year and got a pair of V3s as a holiday gift. I use Sufferfest/Zwift to control the TVS and record the V3 using a Garmin 920XT - so I get power from both devices. Initially, the TVS was 10% higher than the V3 - but this is my old trainer bike with Shimano Biopace (nonround) chainrings. I though that each power meter might respond differently to the nonroundness. So I mounted my road training bike (round chainrings) and did not see much change - TVS still aout 10% lower than V3. When I put the trainer bike back on, the V3 power was all over the place and high. Hmmm…well, problem # 1 was that I forgot to change the crankarm length from 170 mm (road bike) to 165 mm (trainer bike). Having fixed that, the V3 power was still weird, with crazy R/L proportions, for example. Well, despite the ‘it doesn’t matter how tight the V3s are, just tighten them, calibrate them (which I did before each ride) and ride them’, I discovered that loosening the V3s and really tightening them on the crankarms made all the difference - so this was problem #2. Now, the V3s and TVS are within a couple of percent. I just did a ride where the average power was within 1% and the normalized power was identical. So - lesson - if you have a short pedal wrench (as I do), tighten the V3s down with some force. I’m going to get a long handled pedal wrench…