I think you should choose between Sennheiser HD595 and Alessandro MS1.
The Alessandro MS1 is also made by Grado (based on the SR125 I think), but optimized for a more balanced sound, compared to the usual Grado sound that's on the bright side.
For rock, the Alesandro will perform better, for dance, pop, rap, jazz, classical, the Sennheiser.
If you are considering games/movies, I can only say that the HD595 is the most recommended model here. Comfort side, the HD595 is clearly the winner.
One advantage of the MS1 is its price. Anyway, you can't go wrong with any of these models.
The best thing is to know what kind of sound you are looking for. It's also worth to mention what kind of music do you listen the most.
Grados are the first choice when talking about rock music, but they are a bit overpriced in europe.
I don't think you will have a problem finding the low-impedance version of the 595. The high impedance one was only available for a short time when it lauched.
Here is a good illustration of the three LSD types.
You can see how the pinion, under torque, pushes the plates creating pressure on the clutches. It moves up on throttle and down on coast.
-the 2 way diff has the same locking pressure on coast and throttle
-the 1 way diff puts pressure and moves the plates only on throttle
-the 1.5 way diff has a lower pressure on coast.
The ramp angles on the cam set the locking power. You may remember that from the GPL setups.
I think the preload always has an effect on the clutches, no matter in what state it is.
In this picture you can see the preload springs pushing the pressure rings.
If you add spring hard enough, it will keep the clutch pressed all the time.
I don't think it's possible to have a lower force, in a torque loaded state, then the preloaded one. Of course, I don't think anyone would do such a thing in RL
"To change the ramp angles, new ramps are needed – just exchange one for the other. If preload is needed add shims behind the preload springs, or change to stiffer springs. Locking rate of both sides of the ramp can be lowered by reducing preload, or by rearranging the clutches to reduce the number of active surfaces."