I have loaded the file as posted in my previous post into the new version (6.5.54) and it immediately recognizes the start codes! great! I will try with some live data later and keep you updated. Thanks.

I have tried with different time base settings and so on, even tried the inverse of he wave forms. The software seems not to recognize the start and stop conditions properly. What I have noticed is that it recognizes a rising clk while data is low as a stop condition, it should be a rising data whil...

You need more packets really, try using a longer timebase. The software needs to see at lease one start and one stop bit. So I will always miss the packet I am interested in? My Picoscope 2205 buffer is only 16kB, not long enough to store data for seconds. I will try whether this is indeed the case...

I tried to use the I2C serial decoding tool, but without any success. The decoder does not seem to recognize a start condition the way it should. What am I doing wrong? Attached a screenshot: clearly the SDA drops before the SCL, and no start condition is decoded, neither is the data properly decode...