Is the performance bug of nvds_obj_enc_finish() existed on x86 version

[lin.bruno]

**• Hardware Platform (Jetson / GPU) RTX-4090
**• DeepStream Version: 6.2
• JetPack Version (valid for Jetson only)
**• TensorRT Version: 8.5.2.2
**• NVIDIA GPU Driver Version (valid for GPU only): 535.54.03
**• Issue Type( questions, new requirements, bugs): Questions

In this post - 243883. we have performance bug of nvds_obj_enc_finish () on Orin NX platform.
Now we’d like to migrate to x86 platform with RTX-4090.
We want know if the bug is existed on x86 platform? And if yes, please provide solution.

@kpernos9

[Fiona.Chen]

x86 GPU is different(hardware and software). What do you expect the performance for the x86 GPU JPEG encoder is?

[lin.bruno]

Now we are running deepstream-app on x86.
With dual 120 FPS cameras, we get stable 120 FPS performance.

However FPS is dropped, if we save jpeg files by nvds_obj_enc_finish() in some frames,
It seems it costs about 40 ms to do jpegencode process for one image file.

[Fiona.Chen]

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

So your requirement is 1080p 120fps JPEG encoding, right?

Can you send message to tell us your company and project?