Substential amount of time is lost in the QmitChart widget between the emit signal on the c++ side and the JavaScript-Event, so in the web channel. As each trace/curve triggers its own signal that quickly piles up to a considerably latency.

We should investigate that. Either test the channel blocking/unblocking ability if it helps to communicate blockwise and not for each curve. If that does not help, we might need to rework the data structure and communication style to reduce the number of webchannels interactions.

Minimize web channel usage in the old implementation to reduce the delay to an acceptable amount.
If blocking realy helps must be evaluated.

But I think a rework is needed in the end to get a realy smooth experience. In the end we should update manipulate the complete list of traces on the c++ side with a dirty flags, signal if everyhting is ready and then go through the whole registeredObject in one function instead of many single function calls...