You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think the performance while using eCAL with foxglove would prove significantly better if the bridge was implemented in C++, similar to ecal-rerun-bridge.
The text was updated successfully, but these errors were encountered:
Do you encounter specific problems with the bridge?
So far, I haven't seen any performance problems. We have recently fixed some internal memory leaks for the Python binding, which we will be releasing bugfix releases for (eCAL 5.13.3, eCAL 5.12.7).
When I initially implemented this bridge, I also tried implementing it in C++, but it was a lot more difficult to build, and I didn't see any benefits over the python version (latency, throughput, ...), so I dropped the C++ implementation.
I did notice the FPS drops with rerun while using python. Hence I believe while using high FPS topic visualisation, writing it in C++ would make it more resilient.
I am interested in contributing to the C++ implementation if there is any initiated. Python while having a GIL might cause bottlenecks when handling more topics at a higher FPS.
I think the performance while using eCAL with foxglove would prove significantly better if the bridge was implemented in C++, similar to ecal-rerun-bridge.
The text was updated successfully, but these errors were encountered: