Alot of jitter in multiple source

[m.haydari]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) T4
• DeepStream Version 7.0
• JetPack Version (valid for Jetson only)
• TensorRT Version deepstream 7.0 installation guide
**• NVIDIA GPU Driver Version (valid for GPU only)**535
• 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)

hi i have alot of jitter in this pipeline can you help me in this case? i alse have to say that the latency of element is too low but frame latency is too high ! why this happend

[fanzh]

1. On which plugin did you add latency_measurement_buf_prob? please add it on nvdsosd.
2. please refer to this link for performance improvement. when running the application, what is the GPU utilization if checking with nvidia-smi? could you share the log of “nvidia-smi dmon”?
3. could you share more latency logs? noticing the frame latency is about 3 seconds, but component latency in the screenshot is not high.

[m.haydari]

@fanzh

latency_measurement_buf_prob? pleas

1- it’s after third nvinfer
2- for batched-push-timeout should i set it to 1’000’000/max_fps_between cameras or 1/max_fps_between cameras ?(for example if i have 3 cameras with 15,15,20 should i do 1’000’000/20 )?all of my cameras has fps 20 so i wrote 40000
i will sent you more logs but i did everything in the link you said before sending previous log

[fanzh]

if you have 3 cameras with 15,15,20 please set 1000000/20=50000. All the cameras has fps 20. so please set 50000

[m.haydari]

i did it but is there anything else that i can do?

[fanzh]

1. you may increase interval of pgie nvinfer and secondary-reinfer-interval of sgie nvinfer to improve inference speed.
2. 1. Is there any improve with the methods above? could you share the new latency log?
3. if the issue remains, to narrow down the issue, could you only test “…->osd->fakesink” pipeline?

[yingliu]

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks