Perhaps this will be of interest, I found it on the Skylinesdownunder board. It's Mario GTR's views on dyno's, transmission losses etc. I found it interesting to read.
Seriously, one thing your all forgetting is that the GTR is not a 4WD car. It is never completely 4WD, so you are always going to get a variation when running the car on a dyno and on the track.
You have less driveline loss when the car is 100% RWD. As traction and G's dictate how much power is sent to the fronts, you incurr a greater loss through the transfer case, fron diff and front tyres, in proportion to the amount of power sent to the front wheels.
So, all these calculations of acceleration over time, which take the vehicle weight and power at the wheels, require you to factor in the amount of torque split that the GTR will be at any single moment in time during the run.
Easy to do on a RWD car, and if you locked your GTR into RWD and did the complete run, you'd find that you would be pretty close to the mark with your HP to ET/Trap Speed estimates.
On the subject of power loss through the drivetrain. A lot of power is "lost" through the drivetrain. There is a common misconception that all of the "loss" is just converted into heat energy, it is not. A majority is, but there is also the conversion to sound (noise) and slipping of clutch surfaces (the clutch and clutch packs).
Take into consideration that from the flywheel, there is the clutch (which has some slip, no matter how strong it is), then there is the gearbox with all of its rotating shafts and fluids, then there is the transfer case (which is a clutch pack and has a parasitic drive to the front prop shaft) then onto the rear prop shafts (joined with a CV joint) then onto the rear differential, which has to split the drive 90 degrees through whatever clutch/split mechanism your diff is constructed from, then through the rear drive shafts, via two sets of flex joins to the hubs, then the hubs and brake rotors, then the wheel rims, which contain tyres, which then mate to the road surface.
Now back to the front prop shaft, it goes into the front differential, which splits the drive 90 degrees, much akin to the rear differential, through to the front drive shafts which have the same layout as the rears, through the the tyres.
When on a dyno (or the track) all of these interface points will combine to create a parasitic loss, not just from friction events, but also from the acceleration of the weight of the assemblies (have you any idea how much power it takes to accelerate just a 10 Kg brake disc rotor? Keep in mind that on disc brakes, there is always some touching of the pads to the disc surface as well...)
The majority of the "loss" (I should say energy conversion, but you get the idea) is actually from the wheel and tyre interface, this is why all of those "hub dynos" are a waste of time and provide figures that are not a real indication - as I don't know of anyone who drives their car on the street without wheels and tyres, so any power figure that they produce is missing these vital components - especially the tyres...
They tyres are responsible for the greatest portion of lost energy simply because they are the main interface that provides traction - feel your tyres (and check the differences in tyre pressures and internal temperature) between dyno runs for an indication of how much power they are able to sap and dissipate into the air.
Basically, a hell of a lot of energy is lost through the drivetrain, and whilst not in direct linear proportion to the power output of the motor, the "losses" are greater with higher output. I find that a majority of the "losses" are attributable to the speed of rotating assemblies (with shafts, tyres, etc all have friction points that occur more frequently with greater speed, be it engine shaft speed, gearbox output shaft speeds or wheel speed) the faster these items move, the greater the number of friction events in the same time frame, hence the greater the power lost to these events.
Torque also plays a major part in power lost - as an increase in torque (at the same speed) will induce a greater load onto items like shaft bearings, wheel bearings, mated gear tooth surfaces, torsional diff mounts, engine and drive train mounts, axial loads on drive and prop shafts, are all increased, and as a result, they have a profound effect upon the previously mentioned friction events, and provide a magnification of the same said events, hence that the occurance of those same events now incurr a much greater loss per event.
In short, there is no hard and fast rule to calculate drive train losses, infact, I'd go so far to say that it is virtually impossible to calculate to any reliable degree how much loss a particular vehicle will incurr. It can't be done, there are simply too many variables to consider - and which change based upon other variables, which in turn are affected by others... and it just goes on.
As to the comment that someone made about "boiling oil" they are actually close to the mark - there is a reason why the underside of any car has the gearbox and majority of the driveline open to the air - it makes more sense for economy to enclose the entire underside of a car with aero deflector plates and/or undertrays, as fuel economy and Cd would be greatly reduced - the reasons it isn't done is to allow the drivetain to cool - it does require the ability to dissipate the accumulated heat, else the oil would boil and in some exreme cases, the metals (of the shafts and bell housings, etc) could reach temperatures where they would melt.
I have a great many sensors on my GTR - every component of the driveline (save shafts, because I couldn't work out how to do it... has both fluid temperature and pressure. Every casing has a temperature sensor sender that is also logged. I can tell you without a doubt, that on the dyno, on a DynoDynamics certified 1,000 HP at the wheels run (~745 kW, probably about 1500 DynoJet HP.... Using the aforementioned smaller Amercian horses... Hehe the amount of heat put through various components is as follows, for a 10 second run:
Gearbox was +70 degrees Celcius.
Front diff was +87 degrees Celcius (affected by being in engine bay and attached to the block).
Rear diff was +79 degrees Celcius.
*: These are deltas, as in the additional heat that was added during the run (Ie: Post.Temp - Pre.Temp = Delta.Temp).
Tyre temps were not checked, but they were spinning enough to leave melted rubber onto the underside of the wheel arches. Interesting to note that tyre pressures were 30 psi and after the run, they were 39psi - in 10 seconds... work it out.
Sorry for the rant, one day I'll write up something more scientific, but the bottom line is that a hell of a lot of power is used to drive the drivetrain - and especially the tyres - not the mention the acceleration of shafts and rotating assemblies.
So the cossie 4x4 is more efficient in getting power to the wheels...but the downside appears to be strength..they only good for around 330bhp..whereas the Skyline is around 600bhp.So to sum it up you got a pretty trick transmission system..even though you lose a tad more to the wheels.