Summer last year (2001) I got the idea to measure real world acceleration times (0-100 kph) for my Boxster by taking video shots of the speedo. The digital video camera provides 25 frames per second and the time code can be displayed for each frame. By recording the speedo all the way during acceleration in fact a complete v-t (velocity versus time) curve can be generated.

In the beginning I started taking video shots when acceleration from a stand still. It appeared that a lot of spread was introduced because of slippage at the start when releasing the clutch at relatively high rpm. To get rid of this start effect and at the same time spare the clutch for too much wear, it was Joris' idea to start accelerating at a low stable speed in first gear (~ 15 kph) and extrapolate the curve to zero kph. So Joris was also interested which was a great opportunity to also compare his 2.5l Tiptronic with my 2.5l manual. We took the recordings at my favourite stretch between two roundabouts. The Tiptronic recordings were done starting in first gear and in automatic transmission mode which shifts exactly at red line.

As we all know the Boxster's speedo is not very accurate. In order to translate the speedo readings (from video) into actual speed, we needed to know the speedo error at each velocity. We did put a lot of effort into the speedo calibration which has been posted before. If you are interested in the calibration results and how this was done then look at the next link:

The video revealed that the Boxster's speedo updates the display every 0.5s . For each run every refreshed speedo display value including time code was read from video and put into an Excel sheet for further calculations. First all speedo readings were converted into actual speed. The next step was to extrapolate the curves to 0 kph which was done by a linear fit through three data points at the lower end of the curve. Now the curves were completed and represent actual speed versus time. The nice thing is that a certain delay of the speedo display does not effect the shape of the curve and therefore also not the time interval between two points on this curve like 0-100 kph.

The graph below shows three recordings for the manual and two for the Tip. I have to say that in two previous sessions with my manual Zoef, I recorded somewhat slower times and learned from these recordings that I shifted too early. For the best times I needed to really shift at red line which needed some practice. As for my shifting I drive the stick for almost 30 years and like to do it in a sporty manner so I do have some experience but probably it can still be done better.

Important to know is that for all runs we drove with two people aboard (driver and the video shooter) and to this respect we expect no significant weight difference between our two cars apart from the typical weight increase of the Tiptronic transmission. Because of extra weight (two persons) a straight forward comparison with factory figures is not possible though.

What can be concluded:

1) for 0-100 kph in these runs my manual beats the Tip with only 0.1-0.2 s !
2) the manual is faster between the shifts (steeper curve) but definitely loses during the shifts
3) the manual shift-dips show up clearly in the graph
4) there is virtually no acceleration loss during shifts for the Tip !
5) with the Tip, acceleration is very reproducible as was expected
6) although the 0-100 time is almost equal, the covered distance for the manual (which in fact is represented by the area underneath each curve) will be larger.
7) from these curves the next 0-100kph (0-62mph) times can be derived:

8.1s for 2.5l Tip
7.9s for 2.5l manual

Btw. I'm still during break in for my new Boxster S but looking forward to add curves (both road and v-t) for the S :-)

Rene

Btw. this was posted on Porsche Pete's Boxster Board: