Updated BMW F82 M4 Performance Simulations-Predictions - Even Faster...

[trey100]

For the people arguing over the trap speeds - these are theoretical calculations but are fun nonetheless. I assume it has to also guess at the torque curves which adds to the potential inaccuracy. Should be interesting to compare real world results to these once real dynos are available. One place I think the theoretical calcs are off is the 5-60 runs. In the calculations it shows all 3 cars with a lower 5-60 than their respective 0-60. Mid 3's 5-60 is not going to happen.

[swamp2]

Originally Posted by flinchy

You got at least one key thing wrong

F10 m5 not shifting at redline?

You dont shift at peak tq you shift at the point where tq effective (counting final drive and gear ratio) is higher in higher gear

M5 you ALWAYS shift at redline

According to my work on the subject (I can post the force in gear vs. speed graphs if you like) the new F10 M5 should not be shifted at redline in all gears to obtain maximum acceleration. It is quite a fine line but IIRC in the M5 shifts into 5th through 7th should be at a bit before redline. As is the case with this simulation for the M4, the differences between shifting or not at redline in the M5 would be quite subtle, with only a very small bit to gain by shifting lower. I've seen the reports on m5board.com claiming that shifts are always at redline but I'm quite sure they are incorrect.

When to shift is quite simple, it is not about average power nor about torque, torque is irrelevant. One shifts where the accelerative force is higher in the next gear than if the shift was not made. In other words if lines on the force vs. speed in each gear cross that shift should be made at the vehicle speed (can back calculate rpm) where they cross. Simple.

[Bowser330]

Thanks for the data swamp2, I really hope we can revisit this thread when the first car hits the dyno to see just how close you were to the real thing.

Originally Posted by BigMacSmallFries

Thanks for posting, but 8.3 seconds to 100mph?! An LP560 4 wheel drive/550hp/3400lbs does it in 8. I hope you're right!

Where did you get your data?

6.9 seconds
http://fastestlaps.com/cars/lamborgh...o_lp560-4.html

Further proof your number may be incorrect is that the GTR 530hp/3900lbs 0-100mph in 7.4 seconds
http://www.motortrend.com/roadtests/...t/viewall.html

[swamp2]

Originally Posted by trey100

For the people arguing over the trap speeds - these are theoretical calculations but are fun nonetheless. I assume it has to also guess at the torque curves which adds to the potential inaccuracy. Should be interesting to compare real world results to these once real dynos are available. One place I think the theoretical calcs are off is the 5-60 runs. In the calculations it shows all 3 cars with a lower 5-60 than their respective 0-60. Mid 3's 5-60 is not going to happen.

Not a guess as the the curve an exact torque curve vs rpm is used in the simulation and I posted it. It is shaped just like the F10 M5 curve but has the rumored 395 ft lb torque and my assumed redline of 7600. Indeed those are assumptions. Again, please recall that the current car has been clocked by a magazine 0-60 in 3.9 seconds. You can extrapolate with more power, less weight, wider and perhaps better tire, stiffer chassis, etc. what result is or is not reasonable.

[BigMacSmallFries]

Originally Posted by Bowser330

Thanks for the data swamp2, I really hope we can revisit this thread when the first car hits the dyno to see just how close you were to the real thing.



Where did you get your data?

6.9 seconds
http://fastestlaps.com/cars/lamborgh...o_lp560-4.html

Further proof your number may be incorrect is that the GTR 530hp/3900lbs 0-100mph in 7.4 seconds
http://www.motortrend.com/roadtests/...t/viewall.html

Oh really?? Thanks bud! That takes care of it all. Just ignore all the 7.6-8.4 0-160 times that have been posted for the LP560, and ONLY post that of the PRESS car. Thank you, you're clearly correct, all those other publications weren't flooring the pedal all the way.

Type in "lamborghini lp560-4 0-160 seconds" into google and see the times you get on the first couple of pages. Hint: They average out close to 8. Finding the fastest time on a press car isn't called 'proof', it just means you might be naive.

[CanAutM3]

Originally Posted by swamp2

When to shift is quite simple, it is not about average power nor about torque, torque is irrelevant. One shifts where the accelerative force is higher in the next gear than if the shift was not made. In other words if lines on the force vs. speed in each gear cross that shift should be made at the vehicle speed (can back calculate rpm) where they cross. Simple.

I beg to differ. Maximizing average power generated over an acceleration run is what needs to be done to optimize acceleration. Now there are several ways to calculate and figure this out and I agree that the method you suggest is one of the most straight forward ones (it is also the one I use).

The only difference between propulsive force and wheel torque is the wheel radius (this is pretty much a constant on a street cars, maybe not on a top fuel dragster though ). You need the engine torque curve and gear ratios to plot propulsive force vs road speed. Saying that torque is irrelevant is not really accurate.

You will find that when the force lines of two gears cross, the power generated in the lower gear is equal to the power generated at the higher gear at that speed. Thus maximizing average power.

[See5]

If BMW gets down to 3450 I'll be extremely impressed. The C7 Corvette has an aluminum chassis and they are barely squeaking below 3300lbs(3298) for the base model. A Z51 with the track stuff will probably add about 50-75 pounds to that.

[solstice]

Theory aside, in daily driving I will see little to no incentive to take this type of engine to the redline. Keeping it at full boil in the midrange will be more rewarding than extending it to a flat rpm peak. That you save both some fuel and engine component life doesn't hurt either. The S65 is such a savage towards the peak in both symphony and power that you run it up there all day long, or as often as you can. I think no matter how we dice it, there will be a difference in how we enjoy these cars unless BMW pull another rabbit out of the hat. I will not bet against them even if I chalk it down as unlikely.

[Sedan_Clan]

Nice!

[swamp2]

Originally Posted by CanAutM3

I beg to differ. Maximizing average power generated over an acceleration run is what needs to be done to optimize acceleration. Now there are several ways to calculate and figure this out and I agree that the method you suggest is one of the most straight forward ones.

The only difference between propulsive force and wheel torque is the wheel radius (this is pretty much a constant on a street cars, mybe not on a top fuel dragster though ). You need the engine torque curve and gear ratios to plot propulsive force vs road speed. Saying that torque is irrelevant is not really accurate.

You will find that when the force lines of two gears cross, the power generated in the lower gear is equal to the power generated at the higher gear at that speed. Thus maximizing average power.

I don't disagree with what you've said. It is a matter of simplicity and clarity.

Certainly wheel torque and propulsive force are related by wheel radius, which less dragsters remains close to constant. However, the further back we go in the drive train; wheel, driveshaft, transmission, crankshaft, etc. and the more you speak about torque (unless you are concerned about the strength and stiffness requirements of these components) the more torque becomes a bit meaningless. Clearly we agree that torque at the crank (without taking into account all gearing in the drivetrain) really tells you NOTHING about vehicle performance. Thus why stop at the axle! The simplest and must fundamental concept is indeed propulsive force.

Next, why worry about averaging when you can focus more simply about maximizing instantaneous propulsive force? It just adds conceptual and mathematical complexity that is not required to understand the issue to worry about averaging and/or power, the former technically requires numerical integration of a curve...

No worries, we're pretty well on the same page here.

[swamp2]

Originally Posted by See5

If BMW gets down to 3450 I'll be extremely impressed. The C7 Corvette has an aluminum chassis and they are barely squeaking below 3300lbs(3298) for the base model. A Z51 with the track stuff will probably add about 50-75 pounds to that.

I think you already are aware of this but 3450 is wet, no driver. If that weight was realized for any car, the manufacturer would report ~>3600 curb weight. As mentioned prior these simulations are absolutely including a 160 lb driver in addition to the weight shown in the dialog box input.

[swamp2]

Originally Posted by solstice

Theory aside, in daily driving I will see little to no incentive to take this type of engine to the redline.

I disagree. If this simulation is correct on this point, and I believe it to be almost for sure so, redline shifts will be required in gears 1-3 or 1-4 in the new car for maximum acceleration, which is identical to the existing car. If you want the maximum performance, redline will be required, simple. Of course you could avoid some of the odd feeling associated with the tapering off of torque and acceleration at high rpms by shifting a bit early. Ultimately though that is like hoping you earn less money so you can pay less taxes!

[solstice]

Originally Posted by swamp2

I disagree. If this simulation is correct on this point, and I believe it to be almost for sure so, redline shifts will be required in gears 1-3 or 1-4 in the new car for maximum acceleration, which is identical to the existing car. If you want the maximum performance, redline will be required, simple. Of course you could avoid some of the odd feeling associated with the tapering off of torque and acceleration at high rpms by shifting a bit early. Ultimately though that is like hoping you earn less money so you can pay less taxes!

Not really. I don't race in daily driving so maximum performace is not really important. It's the feel that counts for me in daily driving and as you guessed, the tapering or flattening of increased power output is not really rewarding for feel. The S65 however feels like it's just building power all the time and the feel for the last 2000 rpms is epic. Not all of us buy cars to be "fastest" at the stop lights but for the sensations they generate . I don't really care if it pulls mid 4s or mid 3s to 60mph or if the trap speed at the quarter is 110 or 130. I do care about the steering feel though and I don't want massive turbo lag or abrupt power bumps, don't mess it up BMW!

[swamp2]

Here are the abbreviated M5 results to address the ongoing questions as to ideal shift points in that car.

Only the first three shifts (into 4th gear) are to be made at redline for maximum performance. Sorry the curve traces are so wide, for some odd reason I could not get them narrow as before which is the default. Either way one can still fairly see the curves crossing.




The software automatically determines the ideal shift points based on the force curves crossing and reports them in tabular form as well. 7th gear is not shown in the table (minor bug):




Also, for clarity I am not just using peak torque and hp values but curves almost exactly as supplied by BMW.

[CanAutM3]

Swamp is correct. The M5 does not need to be shifted at red line in the higher gears as clearly indicated in the graph. This is also true for the 1M. I came to the same conclusion with a similar analysis previously.

To me, this is also an indication that the gear ratios on the M5 were not optimized for best overall acceleration (maybe a compromise for MPG and/or comfort ). 1st, 2nd and 3rd are too far from each other while 4th, 5th, and 6th are too close. I would keep 7th with a taller gear for MPG. Choosing gear ratios that would allow the force line of every gear to touch the force line of the next higher gear without crossing would yield better acceleration over the entire range.

My hope is that the F8X M3/M4 DCT ratios are better optimized to take full advantage of the new engine.

[OrangeCrush]

Thanks Swamp...this is awesome as always....now the real question....are you going to trade your M3 for an M4?

[flinchy]

Originally Posted by swamp2

According to my work on the subject (I can post the force in gear vs. speed graphs if you like) the new F10 M5 should not be shifted at redline in all gears to obtain maximum acceleration. It is quite a fine line but IIRC in the M5 shifts into 5th through 7th should be at a bit before redline. As is the case with this simulation for the M4, the differences between shifting or not at redline in the M5 would be quite subtle, with only a very small bit to gain by shifting lower. I've seen the reports on m5board.com claiming that shifts are always at redline but I'm quite sure they are incorrect.

When to shift is quite simple, it is not about average power nor about torque, torque is irrelevant. One shifts where the accelerative force is higher in the next gear than if the shift was not made. In other words if lines on the force vs. speed in each gear cross that shift should be made at the vehicle speed (can back calculate rpm) where they cross. Simple.

found quickly through google, many more that all look the same

http://ecutuninggroup.com/files/2013...ning-Group.jpg

if you shift before redline there, you lose acceleration... well except like 6-7 since it's a super long gear to a slightly slightly longer gear, not enough of a difference.

same for any i can find on google

i used to use:
http://vlsicad.ucsd.edu/~sharma/Potpourri/perf_est.html

VERY good tool there... but it seems to be down now? sad... had to use http://glennmessersmith.com/shiftpt.html#data

but it IS about average power (multiplied out) therefore average torque man... what are you talking about?

ok, we're kinda saying the same thing though - yes it's about average power, which comes from torque, but if your actual at a certain RPM geared out is lower... you'll simply be putting less to the ground in the longer gear

it's not JUST about torque, it's EVERYTHING about torque, torque at the fly then multiplied through the gearbox, then the diff... if the next gear is a certain amount longer, AND the torque drop is too sharp, then yes you can shift early


ED: ah graphs above, and 7 speed, i was working off 6mt- every gear up to 5th was redline and even then it was only 100something off.

not sure on why you're getting 7080 in 4th and i'm not though, very strange

[flinchy]

Originally Posted by CanAutM3

Swamp is correct. The M5 does not need to be shifted at red line in the higher gears as clearly indicated in the graph. This is also true for the 1M. I came to the same conclusion with a similar analysis previously.

To me, this is also an indication that the gear ratios on the M5 were not optimized for best overall acceleration (maybe a compromise for MPG and/or comfort ). 1st, 2nd and 3rd are too far from each other while 4th, 5th, and 6th are too close. I would keep 7th with a taller gear for MPG. Choosing gear ratios that would allow the force line of every gear to touch the force line of the next higher gear without crossing would yield better acceleration over the entire range.

My hope is that the F8X M3/M4 DCT ratios are better optimized to take full advantage of the new engine.

1m/n54 is totally different, they make peak power in the midrange (5500-6000rpm, and taper HARD to redline, WAY more than the M5/M6 taper.

1m/135i you shift at 7000 in 1-2 then after that it drops to about 6900, 6750, 6600 or something from memory.. certain people's tunes let them go to 7000 in 3rd as well, but that's a case by case basis.

even with the (somewhat unrealistic) torque curve provided in the OP, with the gearing also provided (at possibly off ratios)... using a calc, it's shift to 4th at redline, then the taper i put in is a bit higher than the OP, so 7,7, 7,6 as the ratios get REALLY close together with a REALLY steep taper.

pretty poor optimising of ratios, if they were a bit further apart, the whole rev range could be happily used AND it'd rev lower in top gear on the highway lol.

[swamp2]

Originally Posted by CanAutM3

To me, this is also an indication that the gear ratios on the M5 were not optimized for best overall acceleration (maybe a compromise for MPG and/or comfort ). 1st, 2nd and 3rd are too far from each other while 4th, 5th, and 6th are too close. I would keep 7th with a taller gear for MPG. Choosing gear ratios that would allow the force line of every gear to touch the force line of the next higher gear without crossing would yield better acceleration over the entire range.

My hope is that the F8X M3/M4 DCT ratios are better optimized to take full advantage of the new engine.

Again, when examining contests that use multiple gears, the performance is relatively insensitive to the individual ratios (within reason of course). I believe with the M5 there absolutely is some compromise with the ratios for mpg. You know BMW and "efficient dynamics"... I played around a lot with the gear ratios and because of the torque falling at high rpm and due to the number of gears, it is nearly impossible to have all redline shifts without artificially lowering the redline in the first place. More or less a NA vs. FI occurrence.

I did note that the only shift at redline in your set of suggested gears was 1->2.

[swamp2]

flinchy: The links you provided are entirely irrelevant. The torque curve I am using for the new M4 is nearly exactly like the FACTORY DYNO from BMW for the M5, just adjusted for a bit higher redline. The method to calculate shift points is precisely as I described earlier and this is precisely what CarTest is doing. Until you can offer some proof of your proposition on the F10 M5 gearing/shift points, with a clear description of how the calculation is being performed, you're not positively contributing to the discussion here.

Last but not least you should also do some reading on some of the (epic) debates here on the forum about power vs. torque. I used to think about vehicle performance a bit like you are. It is kind of a crutch for truly understanding the power (no pun intended) and simplicity of power (over torque).

[Bowser330]

Originally Posted by BigMacSmallFries

Oh really?? Thanks bud! That takes care of it all. Just ignore all the 7.6-8.4 0-160 times that have been posted for the LP560, and ONLY post that of the PRESS car. Thank you, you're clearly correct, all those other publications weren't flooring the pedal all the way.

Type in "lamborghini lp560-4 0-160 seconds" into google and see the times you get on the first couple of pages. Hint: They average out close to 8. Finding the fastest time on a press car isn't called 'proof', it just means you might be naive.

I sense a little sarcasm...

Dude don't get your panties all in a bunch, i just wanted to know where you got your data, I see now, from "the first couple of pages" from google comment that you've got about as much "proof" as what i offered, at least I had the decency to link the damn sites.

Oh and thanks for saying a might be naive, you're a real class act.

[BigMacSmallFries]

Originally Posted by Bowser330

I sense a little sarcasm...

Dude don't get your panties all in a bunch, i just wanted to know where you got your data, I see now, from "the first couple of pages" from google comment that you've got about as much "proof" as what i offered, at least I had the decency to link the damn sites.

Oh and thanks for saying a might be naive, you're a real class act.

Sorry, just telling it like it is. I'm a rather classy fellow. lol