Hi,
Another aspect of this to consider is our pedalling efficiency. We all have some rpm ranges where we are more effective. When you go below or above that range your watt output relative to heart rate will go down. It’s quite easy to see if you do a drill to find you max rpm, in light gears, spin as fast as you can. What you will see is typically at your top RPM, you will start bouncing in the saddle. What is happening is that the nerves and muscles to release the pressure (or event lift) on the rear pedal is not able to respond fast enough. This causes there to be some downward force on the rear pedal at the same time as we are pushing down at the front pedal. Thus our efficiency is drastically falling as one leg is fighting the other.
The good news is that it’s easy to train this max rpm. Doing a drill where you spin up your rpm until you start bouncing in the saddle, then sightly reduce the rpm so you are not bouncing and keep it there for a while. Then slow down, rest and repeat. We adapt quickly to this because we are training primarily the nervous system and that is the fastest to respond. So doing this drill a couple of times per week for a couple of weeks probably will result in good progress of increasing your max RPM. Now I am not certain this is the casue of what Oddvar has observed here, and especially since the higher rpm downhill is just 75 rpm, that is usually not at the RPM limit, but it may contribute if his preferred and usuall rpm is a bit lower than 75.
Yes, Ivar, this is certainly a plausible explanation.
This is one of the variables I was hoping to remove in a later test by riding on a less steep hill.
As it seems I’m the only one experiencing this, it seems extra plausible 
I had the same issue Saturday with my Tacx Vortex. The uphill climbs were way to easy, but struggled with the downhills. This was not the issue the previous week, or even the previous Tuesday. I’ve been on Zwift since beta with the same trainer and have never felt this before.