What do you think the actual improvement of torque sleep will be for the P85D when it is released? A bit of context on this question - My P85D is my second Tesla. The first was an early model 85. For the 2 years I had my 85, my lifetime driving average was 348Wh/mi after 20k miles. My driving was a mix of city and highway driving, with probably 65% of the miles in the city, and 35% on the freeway. When taking long trips (e.g. the south peninsula bay area to Napa, or from Mountain View to Monterrey), my Wh/mi would average closer to 325. Overall, my 85 gave me little range anxiety and I had little trouble planning and executing long trips with the car. Since getting my P85D last month, I have averaged 398Wh/mile with the same driving patterns. I have noticed that highway driving is not very good, and my Wh/mi there can range from 350 to 410 depending on if I'm going 60 or 75+mph. I don't find my driving to be any more aggressive than it was in my old 85 and I don't regularly do launches in Insane mode. Since getting my P85D I have been trying to think through why the car is so much less efficient. My old 85 had 19" wheels, and wasn't a performance version. However the wheels should only account for 3% according to Tesla's website. My car has 700 miles on it so I'm mostly through the 1000 mile 5% tax for new tires. And of course we don't have torque sleep yet. So, how much will torque sleep actually provide, and how will that vary under highway vs city driving conditions? I've wondered how much the front and rear motors end up competing with each other in actual driving - particularly at cruise. In a hypothetical wholly inefficient calibration model, one could envision a scenario where one motor is fully providing power, while the other is fully regenerating. In that case, you'll lose a lot of power simply to heat as you transfer power to mechanical energy and back to electrical power through the motor. In a fully efficient model, the two motors are optimally calibrated and share workload perfectly. Torque sleep seems like it would enable this by shutting down one motor entirely, and assuming there is minimal cost in terms of the drag the motor would generate, would return the car to efficiency levels similar to a single motor car. However, what about city driving where there is a lot of acceleration/deceleration? Will torque sleep make a difference here? Is it possible that Tesla could release a more optimal algorithm for balancing workload between the two motors? In fact in this scenario, why would there be so much inefficiency between the AWD and RWD models? Wouldn't power be shared efficiently upon acceleration? Wouldn't regen always favor the AWD because you have two motors doing the work and therefore have even less need to use the brakes? Most importantly - what will our Wh/mile improvement actually be? Tesla says 10% - for my driving that would get me to 358Wh/m. But will we actually see that much? Might we see more? or should we expect less? What do you think?