Bkool trainers adjust the speed as the incline changes to simulate inertia, such as when going from a steep decline to an incline the speed falls away gradually rather than immediately adjusting to the new gradient. This simulates real world inertia, which will carry you some short way up an incline faster than you would otherwise pedal.
It seems that other trainer users are finally going to have a level playing field in this regard, possibly because of the fuss made here!
For those of you who haven't updated the simulator yet (or didn't read the intro) here is what it says about the new version.
I like the free spectator mode (as I'm not currently riding but I can hopefully watch you guys having fun).
Edit: Just tried this out, and of course it is limited to the 2D and map modes so not quite as nice as it seems. I would think allowing spectators to watch in 3D would make sense even if you are not a premium member, as it would show you how nice it is and encourage you to upgrade to premium to be able to do your own rides in 3D? I can't see that they would lose any premium subscribers and may gain more as a result.
I also love the comment "we have decided that users without a Bkool Trainer can also enjoy", especially the "we have decided" bit. Pretty much confirms the comments here that there was no real reason to exclude the non Bkool trainers from inertia and gravity assist.
Still not 100% clear from the comment "The inertia when starting a climb or descent, and the speed that it generates" that this means non Bkool trainers will simulate free-wheeling down a slope (it may just mean the initial inertia), so I guess I'll have to try it out to confirm.
Geoff
There was a discussion a while ago here about how a more level playing field between those on different trainers could be achieved - maybe by adjusting the times by some formula. I think it ended up sounding too complex to reconcile both the disadvantage non Bkool trainers got on the descents AND the advantage some with lower max resistance get on the climbs. If this update does solve the descending issue then maybe it will make it easier/possible to come up with a formula to account for just any climbing differences?