CarTest results and comparisons vs. MB AMG

[swamp2]

OK so I got lazy writting a numerical integrator for my acceleration spreadsheets so I bought the CarTest software. Not a very cool web site but again fantastic software. I could go on and on about the level of detail in the software, the included physics and its validation etc. In place of that let me just post a few details specifically about the E92 M3 6MT. BTW all results figures are in seconds and mph (obviously).

spec/CarTest/reported:

0-60 4.47 4.40
1/4 mi 12.9@110.5 12.9@111
0-200 km/h 16.6 15.8
0-1000 m 23.6 23.3

This was a totally "blind" (non-tweaked) prediction using exact available specs for the car. Pretty amazing I'd say. Next I looked at the C63 AMG with its "monstrous" hp and torque "advantage"....

0-60 4.28
1/4 mi 12.82@110.7
0-200 km/h 16.1
0-1000 m 23.6

Like most of us predicted, M3 just a hair slower in most contests. Most of the huge torque and hp advantage is made up for by less power to the wheels, more weight, BER on the M3 (which I took into account in the software!) and the high redline of the M3. By the way the default values and differences in total drivetrain losses between MT and automatic are exactly as I estimated in my other post - 20% vs. 15%.

Let's see how well CarTest does on the potent CLK63 AMG Black Series as we have a vareity of test results on this beast. No matter what I tried, triple checking and re-entering the cars data I just could not get anywhere close the the reported results. The only possible conclusion is a MASSIVE under rating of hp and tq. With some trial and error I got fairly close the the cars reported numbers with 550 hp and 530 ft lb (vs. 500/465 reported).

0-60 4.28 4.0 - 4.4
1/4 mi 12.51@114.7 12.4@116 - 12.6@115
0-200 km/h 14.6
0-1000 m 22.8

OK enough blah, blah, blah about MB. Here is the E92 M3 with M-DCT transmission. The only changes to the base 6MT simulation was 40 lb extra for the tranny itself, adjusting the shift times as low as possible and doing some basic optimization of the gear ratios along with the extra gear. Even given urealistically short shift times in the software the acceleration curves still showed about .1 second of deceleration for the shift so many of these numbers may be about .1 s too slow per shift (again depending on the rumors of .008s shift times).

0-60 4.22
1/4 mi 12.7@111
0-200 km/h 15.0
0-1000 m 23.3



The advantages of near instant shifts, 7 gears and closer gear spacing is obvious. The attached text file is a tab delimited text file with a standard CarTest comparison report between the two M3s and the C63 AMG .

In regards to raw speed contests E92 M3 vs. C63 AMG v. E92 M3 M-DCT are all so close that it is clearly a drivers race. However, I am ready to go out on a limb and predict that when looking at best reported times.

E92 M3 M-DCT > C63 AMG > E92 M3 6MT

And of course by ">" I mean better not more time....

Next even if MB has really cleaned up and improved their handling, which has been happening, becuase of the weight advantage and just because track handling is a hallmark strength for the M3, I'll also predict on most tracks:

E92 M3 M-DCT > E92 M3 6MT > C63 AMG

[Wally330]

Hi,

Thanks for this. One thing though - your 0-100 km/h is indicating what? Those times are about as fast as a Yugo. You mean 0-200 km/h right?

Cheers

[swamp2]

Originally Posted by Wally330

Hi,

Thanks for this. One thing though - your 0-100 km/h is indicating what? Those times are about as fast as a Yugo. You mean 0-200 km/h right?

Cheers

Yes, thanks, since 200 km/h is in the 100 mph range and I was reading the CarTest results off in mph that was my demise. Corrected in the original post. Cheers.

[m_bazeepaymon]

Originally Posted by swamp2

OK so I got lazy writting a numerical integrator for my acceleration spreadsheets so I bought the CarTest software. Not a very cool web site but again fantastic software. I could go on and on about the level of detail in the software, the included physics and its validation etc. In place of that let me just post a few details specifically about the E92 M3 6MT. BTW all results figures are in seconds and mph (obviously).

spec/CarTest/reported:

0-60 4.47 4.40
1/4 mi 12.9@110.5 12.9@111
0-200 km/h 16.6 15.8
0-1000 m 23.6 23.3

This was a totally "blind" (non-tweaked) prediction using exact available specs for the car. Pretty amazing I'd say. Next I looked at the C63 AMG with its "monstrous" hp and torque "advantage"....

0-60 4.28
1/4 mi 12.82@110.7
0-200 km/h 16.1
0-1000 m 23.6

Like most of us predicted, M3 just a hair slower in most contests. Most of the huge torque and hp advantage is made up for by less power to the wheels, more weight, BER on the M3 (which I took into account in the software!) and the high redline of the M3. By the way the default values and differences in total drivetrain losses between MT and automatic are exactly as I estimated in my other post - 20% vs. 15%.

Let's see how well CarTest does on the potent CLK63 AMG Black Series as we have a vareity of test results on this beast. No matter what I tried, triple checking and re-entering the cars data I just could not get anywhere close the the reported results. The only possible conclusion is a MASSIVE under rating of hp and tq. With some trial and error I got fairly close the the cars reported numbers with 550 hp and 530 ft lb (vs. 500/465 reported).

0-60 4.28 4.0 - 4.4
1/4 mi 12.51@114.7 12.4@116 - 12.6@115
0-200 km/h 14.6
0-1000 m 22.8

OK enough blah, blah, blah about MB. Here is the E92 M3 with M-DCT transmission. The only changes to the base 6MT simulation was 40 lb extra for the tranny itself, adjusting the shift times as low as possible and doing some basic optimization of the gear ratios along with the extra gear. Even given urealistically short shift times in the software the acceleration curves still showed about .1 second of deceleration for the shift so many of these numbers may be about .1 s too slow per shift (again depending on the rumors of .008s shift times).

0-60 4.22
1/4 mi 12.7@111
0-200 km/h 15.0
0-1000 m 23.3



The advantages of near instant shifts, 7 gears and closer gear spacing is obvious. The attached text file is a tab delimited text file with a standard CarTest comparison report between the two M3s and the C63 AMG .

In regards to raw speed contests E92 M3 vs. C63 AMG v. E92 M3 M-DCT are all so close that it is clearly a drivers race. However, I am ready to go out on a limb and predict that when looking at best reported times.

E92 M3 M-DCT > C63 AMG > E92 M3 6MT

And of course by ">" I mean better not more time....

Next even if MB has really cleaned up and improved their handling, which has been happening, becuase of the weight advantage and just because track handling is a hallmark strength for the M3, I'll also predict on most tracks:

E92 M3 M-DCT > E92 M3 6MT > C63 AMG

you really believe that 6 speed manual is inferior to the DCT on a track (with a lot of turns no straight aways)?
please do not let your credibility go down the drain with childish claims...

Straight aways are one thing I can understand, but don't start assuming stupid thoughts on other things.

[jworms]

Originally Posted by m_bazeepaymon

you really believe that 6 speed manual is inferior to the DCT on a track (with a lot of turns no straight aways)?
please do not let your credibility go down the drain with childish claims...

Straight aways are one thing I can understand, but don't start assuming stupid thoughts on other things.

DCT > 6spd on a track. hell even SMG > 6spd on a track. race teams have migrated over from the standard manual transmission because the sequential systems proved to be faster around a track. if i remember right the e46 M3s had some issues earlier on with overheating of the smg tranny, but after that was taken care of (i think with just some kind of heat sink-like device) it provided quicker laptimes than its 6spd counterpart. i can only assume bmw's new DCT will be even better.

[swamp2]

Originally Posted by m_bazeepaymon

you really believe that 6 speed manual is inferior to the DCT on a track (with a lot of turns no straight aways)?
please do not let your credibility go down the drain with childish claims...

Straight aways are one thing I can understand, but don't start assuming stupid thoughts on other things.

But you are really making a fool of yourself here. Mixing an insult with your massive misunderstanding really puts the icing on the cake. However, once again I still will not resort to your level with unpleasantries such as "childish" and "stupid".

[Romo]

Great news Swamp, thanks, this is what I call PROGRESS..............wow

I wanted the DCT from the beginning, it is nice to hear it was the right decision.............................spring 2008 is worthwile waiting IMO

[southlight]

Just a question, Swamp: In which gear did you reach 200km/h with the M-DCT. Is it also the 4th like on MT?
I wondered if the DCT will reach 200 km/h in the same gear like on MT cars!?

Best regards, south

[swamp2]

Originally Posted by southlight

Just a question, Swamp: In which gear did you reach 200km/h with the M-DCT. Is it also the 4th like on MT?
I wondered if the DCT will reach 200 km/h in the same gear like on MT cars!?

Best regards, south

Even in metric since you were the one asking. They both reach 200 km/h in 4th gear.

[southlight]

Originally Posted by swamp2

Even in metric since you were the one asking. They both reach 200 km/h in 4th gear.

Thanks for that, but I'm afraid there's anyhing wrong in that graph. The graph says the 6MT M3 needs 5.0 seconds to 100km/h aswell as 16.9 to 200km/h!? Furthermore I don't know if the transmission ratios for the MT are right: Shouldn't the kinks in the graph be where the car reaches 8400rpm and will be shifted, right? But the kinks in this graph are at too high speeds: First shift in the graph is at 70.4 km/h, but in first gear at 8400 we're at 65 km/h. Same goes for the following gears:
2nd 118.5 - 111.2
3rd 178.4 - 166.5
Maybe that's an error in reasoning on my side or the transmission ratio in this graph is too wide!?

Best regards, south

[ILC32]

Very interesting. Nice work.

[Garrett]

Originally Posted by swamp2

Even in metric since you were the one asking. They both reach 200 km/h in 4th gear.

Great, this software is new to me. Is it possible to get a better resolution in the graphes? Meaning having a marker every 1 mph...etc ???

[sajakh]

Originally Posted by swamp2

OK so I got lazy writting a numerical integrator for my acceleration spreadsheets so I bought the CarTest software. Not a very cool web site but again fantastic software. I could go on and on about the level of detail in the software, the included physics and its validation etc. In place of that let me just post a few details specifically about the E92 M3 6MT. BTW all results figures are in seconds and mph (obviously).

spec/CarTest/reported:

0-60 4.47 4.40
1/4 mi 12.9@110.5 12.9@111
0-200 km/h 16.6 15.8
0-1000 m 23.6 23.3

This was a totally "blind" (non-tweaked) prediction using exact available specs for the car. Pretty amazing I'd say. Next I looked at the C63 AMG with its "monstrous" hp and torque "advantage"....

0-60 4.28
1/4 mi 12.82@110.7
0-200 km/h 16.1
0-1000 m 23.6

Like most of us predicted, M3 just a hair slower in most contests. Most of the huge torque and hp advantage is made up for by less power to the wheels, more weight, BER on the M3 (which I took into account in the software!) and the high redline of the M3. By the way the default values and differences in total drivetrain losses between MT and automatic are exactly as I estimated in my other post - 20% vs. 15%.

Let's see how well CarTest does on the potent CLK63 AMG Black Series as we have a vareity of test results on this beast. No matter what I tried, triple checking and re-entering the cars data I just could not get anywhere close the the reported results. The only possible conclusion is a MASSIVE under rating of hp and tq. With some trial and error I got fairly close the the cars reported numbers with 550 hp and 530 ft lb (vs. 500/465 reported).

0-60 4.28 4.0 - 4.4
1/4 mi 12.51@114.7 12.4@116 - 12.6@115
0-200 km/h 14.6
0-1000 m 22.8

OK enough blah, blah, blah about MB. Here is the E92 M3 with M-DCT transmission. The only changes to the base 6MT simulation was 40 lb extra for the tranny itself, adjusting the shift times as low as possible and doing some basic optimization of the gear ratios along with the extra gear. Even given urealistically short shift times in the software the acceleration curves still showed about .1 second of deceleration for the shift so many of these numbers may be about .1 s too slow per shift (again depending on the rumors of .008s shift times).

0-60 4.22
1/4 mi 12.7@111
0-200 km/h 15.0
0-1000 m 23.3



The advantages of near instant shifts, 7 gears and closer gear spacing is obvious. The attached text file is a tab delimited text file with a standard CarTest comparison report between the two M3s and the C63 AMG .

In regards to raw speed contests E92 M3 vs. C63 AMG v. E92 M3 M-DCT are all so close that it is clearly a drivers race. However, I am ready to go out on a limb and predict that when looking at best reported times.

E92 M3 M-DCT > C63 AMG > E92 M3 6MT

And of course by ">" I mean better not more time....

Next even if MB has really cleaned up and improved their handling, which has been happening, becuase of the weight advantage and just because track handling is a hallmark strength for the M3, I'll also predict on most tracks:

E92 M3 M-DCT > E92 M3 6MT > C63 AMG

Extremely informative analysis!

Although I would much rather prefer the involvement of the 6MT, I think the M-DCT will be the M3 that will blow away the competition. It may even be better for BMW to release the M-DCT first to silence the naysayers.

By the way, could you do an analysis of the v2 Procede 335i just for sh##s and giggles.

[Keto]

I was planning on waiting for M-DCT anyway, but this sort of thing confirms my decision.

[Lex]

Just a short note, my service advisor said the new tranny will be 50lbs. lighter than the manual? I don't know if he was thinking 50 Lbs. lighter than the SMG, but that is what he said.

[swamp2]

Originally Posted by southlight

Thanks for that, but I'm afraid there's anyhing wrong in that graph. The graph says the 6MT M3 needs 5.0 seconds to 100km/h aswell as 16.9 to 200km/h!? Furthermore I don't know if the transmission ratios for the MT are right: Shouldn't the kinks in the graph be where the car reaches 8400rpm and will be shifted, right? But the kinks in this graph are at too high speeds: First shift in the graph is at 70.4 km/h, but in first gear at 8400 we're at 65 km/h. Same goes for the following gears:
2nd 118.5 - 111.2
3rd 178.4 - 166.5
Maybe that's an error in reasoning on my side or the transmission ratio in this graph is too wide!?

Best regards, south

Max speed in gears according to my calculations are (km/h):
68
117
232
276
317 (if gear limited, but likely not, drag limited)

I think the problem is simply the fidelity of the graph routines in the software. Looks to me from the graph like 0-62 for the 6MT is 5.0, actual figure from the table report is 4.7. I just don't think you can see 5% or so on the graphs, even the traces are a bit too wide for that. The graphs are better just to see relative performance. Agree the graphing part of the softeware likely could be improved a bit as well...

Here are the numbers in metric, tabular form:

[esquire]

Originally Posted by m_bazeepaymon

you really believe that 6 speed manual is inferior to the DCT on a track (with a lot of turns no straight aways)?
please do not let your credibility go down the drain with childish claims...

Straight aways are one thing I can understand, but don't start assuming stupid thoughts on other things.

maybe you're forgetting that automated clutch transmissions are used in F1 racing. these sorts of transmissions provide quicker shifts than even the most skilled driver, and further provide rev-matched downshifts for proper turn exit-speed. they also eliminate the need for delicate heel-toe downshifting, the driver now being free to brake while the car does it's own shifting. i think maybe you've misunderstood the nature of the beast.

on a different note, your insults were out of line and uncalled for... i don't even see what provoked you to respond the way you did. i'm trying to be objective when i say it was out of line.

[esquire]

swamp,

your simulation doesn't explain car and drivers 4.4 seconds to 60 mph in 6mt. why?

[swamp2]

Originally Posted by esquire

swamp,

your simulation doesn't explain car and drivers 4.4 seconds to 60 mph in 6mt. why?

The simulation figure was 4.47, pretty darn close to 4.4, no?

What it doesn't explain it that some folks are expecting/confident in a better time when the road surface is better so then 4.47 vs. say 4.3 is an OK agreement wetween sim and actual. The simulation of the actual launch (traction, clutch and throtle modulation, wheelspin, etc.) is one of the most difficult parts of such a simulation (not plugging in the numbers, but for the programmer getting the physics right!). Generally the specs that involve little or no wheel spin such as roll ons and acceleration in gear will be more accurate then 0-low speed launches.

Just keep in mind these are engineering simulations, not a crystal ball and can not be expected to be perfect.

[swamp2]

Originally Posted by sajakh

Extremely informative analysis!

Although I would much rather prefer the involvement of the 6MT, I think the M-DCT will be the M3 that will blow away the competition. It may even be better for BMW to release the M-DCT first to silence the naysayers.

By the way, could you do an analysis of the v2 Procede 335i just for sh##s and giggles.

It is easy to be nice and polite when you want something isn't it?

Give me the peak hp, rpm at peak hp, peak tq and rpm at peak tq and I will do some runs...

[RI_RS4]

Swamp,

Maybe this will help you to tweek your simulations. Here are some WOT acceleration curves for my RS4 taken in 3rd gear and 4th gear. Here's the data that you might need on the atmospheric conditions and the car that was tested. These were measured using ECU logs of time and RPM. RPM to MPH conversions were calibrated with a Garmin GPS receiver. And, yes, my dog goes with me when I test the car.

temp=40F
alt=100ft
barometric pressure 30.29
hum=60
mass=4180lbs # car + driver + dog + stuff - 6 gal down from full fuel load
Cd=0.31 # coefficient of drag
FA=2.17 # frontal area in sq meters from RS4 spec sheet
rpm_per_mph=81.01 # 3rd gear, calibrated with Garmin GPS and VAG-COM rpm reading
rpm_per_mph=60.036 # 4th gear


You should be able to use this data to create a simulation for the RS4, and then use it to recalibrate your R8 simulation, since it's engine torque and power curves are nearly identical.

By the way, you've done some good work!


Regards

Scott

[esquire]

Originally Posted by swamp2

The simulation figure was 4.47, pretty darn close to 4.4, no?

What it doesn't explain it that some folks are expecting/confident in a better time when the road surface is better so then 4.47 vs. say 4.3 is an OK agreement wetween sim and actual. The simulation of the actual launch (traction, clutch and throtle modulation, wheelspin, etc.) is one of the most difficult parts of such a simulation (not plugging in the numbers, but for the programmer getting the physics right!). Generally the specs that involve little or no wheel spin such as roll ons and acceleration in gear will be more accurate then 0-low speed launches.

Just keep in mind these are engineering simulations, not a crystal ball and can not be expected to be perfect.

well not exactly, your 4.43 figure is for the M-DCT, whereas car and driver pulled 4.4 with the 6mt. in your analysis, the 6mt hits 60 mph in 4.68 seconds - a significant .28 second departure.