Hi,
I am using the following example and I don’t want to run inference, i just want frames and frame meta data from multiple sources. hence i removed nvinfer/pgie from the pipeline still i don’t see my fps increasing. what could be the reason?
in c after stopping pgie i got 60fps and in python with inference fps was 25 fps and without inference it is still 25 fps
What kind of sources are you using? Can you change “streammux.set_property(‘batched-push-timeout’, 4000000)” to “streammux.set_property(‘batched-push-timeout’, 40000)”?
I am using a video file 1080p. ill change the property and let you know @Fiona.Chen
I changed the property, still the fps is same. i am attaching the file for your reference.v2.py (16.2 KB)
Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
Hardware - Jetson Nano
Deepstream - 5.0.1
Jetpack - 4.3
Tensor RT -7+
reproduce the issue - Running imagedata python deepstream example with nvinfer in the pipeline gets 25 fps without pgie in the pipeline it is still 25 fps @Fiona.Chen
Please change “sink.set_property(‘sync’, 1)” to “sink.set_property(‘sync’, 0)” if you want to test performance.
what does this property do?
Sink will not check timestamp if “sync=0” or else the buffer is rendered according to timestamp, so the FPS will be limited by the timestamp.
okay, I’ll check and let you know
I’ve tried your script with my NX board, with ‘sync=1’, fps is 30, with ‘sync=0’, fps is 185. This property should work.
I will try once again and update the result.