logsout.unpack('all') Slow Performance

Hi I'm a bit stuck with an issue I have using logsout.unpack('all') on Matlab R2006b. I've set up a model to log lots of signals in one of our models and I've been using this on a fairly complex model for some time. logsout.unpack('all') runs pretty quickly on my main model (About 10s to unpack 270 signals for a 100s simulation).

However I have another model which is simpler (175 signals) and running for a shorter time (10s). logsout.unpack('all') takes much longer, around 30 minutes for some reason. This is only on the initial unpack after a sim run. If I unpack again without running the simulation it takes around 1s.

I've compared both models Simulation Configuration Parameters, and I can't see any differences.

I'm very confused and would really appreciate some help in identifying the problem as it would help us use this technique for logging Simulink signals.

Thanks

Answer 1

[Dalectric]

According to Mathworks: "it is a reported bug for releases before release R2008b. The issue is it can be slow when unpacking logged timeseries data from a ModelDataLogs object or using the time series tool for timeseries with many events. This effect is only noticeable the first time the timeseries are unpacked. And it typically occurs when the model that generated it contains an enabled subsystem which experiences many enable/disable events.

We have fixed this problem from release of R2008b, and I am afraid that the best solution for this problem is to upgrade your MATLAB to a newer release."

Oh well, at least I know now.