A couple weeks ago my 'service engine soon' light came on. Almost immediately I also started noticing some erratic engine behavior. When starting cold, the car seemed to operate normally. Normal power, normal clutch actuation, etc. However, after stopping and restarting the engine once normal operating temperature was reached, things were normal no more.
First off, instead of just firing right up, sometimes it would take 2 or 3 seconds of cranking to get the engine going. Then, once on, pressing the throttle caused the engine RPM's to drop several hundred RPM before accelerating. When letting up on the throttle, instead of the engine returning to it's 700 RPM home, it would continue falling, stalling out 4 out of 5 times. Accelerating out of first gear is a challenge. Everytime I begin to let the clutch out, the engine RPM's dive about 4-500 RPMs, often killing the engine. With some gentle feathering of the throttle to get the engine RPM's up around 1300, I can let the clutch out without killing the engine. The RPMs still dive about 500 RPMs, but not enough to stall the engine. Once moving, the engine semes to be cranking at about 60%. Also, 1 out of every 10 times you push the clutch in and let the engine try to return to idle, the engine stalls again.
I connected up the Peake analyzer (for a review of this tool, see the paddock, in the lounge) and it spit out 3 codes: DM-TL leak detected, throttle position sensor 1, and intake camshaft position sensor. I cleared these out, reset the lights, and took it out for a spin until the light came on again. This time, only one code, intake camshaft position sensor.
After doing a little research online, I discovered that this, and it's cousin the exhaust camshaft position sensor, are high failure parts. The people reporting this problem seemed to have similar symptoms, and swapping out this part seemed to cure it. However, for several of these folks, it was only a temporary fix. The engine light would come on again after a couple months and crappy engine behavior would ensue. For one guy, he was on his 4th exhaust sensor, and 2nd input sensor.
Best as I could decipher from the different accounts, there isn't anything explicitly causing these parts to fail. Someone mentioned that a bunch of these sensors (manufactured by Siemens) had left the factory with serious defects, which might explain the recurring problems. If anyone here has access to BMW technical service bulletins, I would be curious if any of them mention the camshaft position sensors. BTW, if anyone has electronic copies of these and wouldn't mind sharing these with a fellow BMW enthusiast, please send me an email or leave me a private message!

I'll be ordering up a new intake CPS on Monday, so I'll let you all know how things work out once I swap it out. Maybe I'll snap a few pictures while I'm at it and throw them up if you guys are interested.