[HW Accel Support]: Need to verify if my configuration is setup properly to utilize my N100 iGPU #19311

DashingBuffalo · 2025-07-29T03:29:34Z

DashingBuffalo
Jul 29, 2025

Describe the problem you are having

Hey all! I'm currently seeing my N100 CPU being utilized much more than my iGPU and wanted to make sure that the configuration I have in place is correct settings and just need to deal with the CPU being utilized this much. It typically sits around 75% with about 100ms response time. I only have 2 cameras set up at the moment but would ideally like to have more.

Version

0.15.2-3bda638

Frigate config file

mqtt:
  enabled: false

ffmpeg:
  hwaccel_args: preset-vaapi

go2rtc:
  streams:
    WYZE_1:
      - rtsp://192.168.1.114:8554/1080p?mp4
      - ffmpeg:Wyze#audio=aac
    CAM_1:
      - ffmpeg:rtsp://admin:password@192.168.1.106/cam/realmonitor?channel=1&subtype=0

record:
  enabled: True
  retain:
    days: 7
    mode: motion
  alerts:
    retain:
      days: 7
  detections:
    retain:
      days: 7

objects:
  track:
    - person
    - dog
    - car
    - cat

snapshots:
  enabled: True
  retain:
    default: 30

cameras:
  CAM_1:
    enabled: true
    ffmpeg:
      inputs:
        - path:
            rtsp://admin:password@192.168.1.106/cam/realmonitor?channel=1&subtype=0
          roles:
            - record
            - detect

  WYZE_1:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/WYZE_1
          roles:
            - record
            - detect

version: 0.15-1
camera_groups:
  Garage:
    order: 1
    icon: LuBean
    cameras:
      - CAM_1
      - WYZE_1

docker-compose file or Docker CLI command

services:
  frigate:
    container_name: frigate
    privileged: true # this may not be necessary for all setups
    restart: unless-stopped
    stop_grace_period: 30s # allow enough time to shut down the various services
    image: ghcr.io/blakeblackshear/frigate:stable
    shm_size: "512mb" # update for your cameras based on calculation above
    devices:
      - /dev/bus/usb:/dev/bus/usb # Passes the USB Coral, needs to be modified for other versions
      - /dev/apex_0:/dev/apex_0 # Passes a PCIe Coral, follow driver instructions here https://coral.ai/docs/m2/get-started/#2a-on-linux
      - /dev/video11:/dev/video11 # For Raspberry Pi 4B
      - /dev/dri/renderD128:/dev/dri/renderD128 # For intel hwaccel, needs to be updated for your hardware
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - /mnt/storage/configs:/config
      - /mnt/storage:/media/frigate
      - type: tmpfs # Optional: 1GB of memory, reduces SSD/SD Card wear
        target: /tmp/cache
        tmpfs:
          size: 1000000000
    ports:
      - "8971:8971"
      - "5000:5000" # Internal unauthenticated access. Expose carefully.
      - "8554:8554" # RTSP feeds
      - "8555:8555/tcp" # WebRTC over tcp
      - "8555:8555/udp" # WebRTC over udp
    environment:
      FRIGATE_RTSP_PASSWORD: "password"

Relevant Frigate log output

Not totally sure if this is needed

2025-07-29 03:16:11.951530499  127.0.0.1 - - [29/Jul/2025:03:16:11 +0000] "" 400 0 "-" "-" "-"
2025-07-29 03:16:18.668106155  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668112280  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668115452  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668118111  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668120517  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668122957  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668125339  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668127723  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668130069  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668132202  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668134710  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668136968  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668139122  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668141364  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668143588  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668145792  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668147989  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668150063  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668152161  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.668154355  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: Unable to read frames from ffmpeg process.
2025-07-29 03:16:18.669614767  [2025-07-29 03:16:18] frigate.video                  ERROR   : WYZE_1: ffmpeg process is not running. exiting capture thread...
2025-07-29 03:16:18.962314629  [2025-07-29 03:16:18] watchdog.WYZE_1                ERROR   : Ffmpeg process crashed unexpectedly for WYZE_1.
2025-07-29 03:16:18.962359341  [2025-07-29 03:16:18] watchdog.WYZE_1                ERROR   : The following ffmpeg logs include the last 100 lines prior to exit.
2025-07-29 03:16:18.962576274  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [segment @ 0x5a8a03a71140] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly
2025-07-29 03:16:18.962678179  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vost#0:0/copy @ 0x5a8a0394af80] Non-monotonic DTS; previous: 1621685, current: 1621685; changing to 1621686. This may result in incorrect timestamps in the output file.
2025-07-29 03:16:18.962684637  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vost#0:0/copy @ 0x5a8a0394af80] Non-monotonic DTS; previous: 1924278, current: 1924278; changing to 1924279. This may result in incorrect timestamps in the output file.
2025-07-29 03:16:18.962840147  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vost#0:0/copy @ 0x5a8a0394af80] Non-monotonic DTS; previous: 2423880, current: 2423880; changing to 2423881. This may result in incorrect timestamps in the output file.
2025-07-29 03:16:18.962848764  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vost#0:0/copy @ 0x5a8a0394af80] Non-monotonic DTS; previous: 2834921, current: 2834921; changing to 2834922. This may result in incorrect timestamps in the output file.
2025-07-29 03:16:18.962968476  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [AVHWFramesContext @ 0x7a925c055c40] Failed to sync surface 0x10: 1 (operation failed).
2025-07-29 03:16:18.963060726  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [hwdownload @ 0x7a9260003980] Failed to download frame: -5.
2025-07-29 03:16:18.963084544  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vf#1:0 @ 0x5a8a03939d80] Error while filtering: Input/output error
2025-07-29 03:16:18.963198144  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vf#1:0 @ 0x5a8a03939d80] Task finished with error code: -5 (Input/output error)
2025-07-29 03:16:18.963255064  [2025-07-29 03:16:18] ffmpeg.WYZE_1.detect           ERROR   : [vf#1:0 @ 0x5a8a03939d80] Terminating thread with return code -5 (Input/output error)

Relevant go2rtc log output

info | 2025-07-29 03:05:10 | startup | Preparing new go2rtc config...
info | 2025-07-29 03:05:10 | startup | Starting go2rtc...
info | 2025-07-29 03:05:11 | startup | go2rtc platform=linux/amd64 revision=b2399f3 version=1.9.2
info | 2025-07-29 03:05:11 | startup | config path=/dev/shm/go2rtc.yaml
info | 2025-07-29 03:05:11 | api     | listen addr=:1984
info | 2025-07-29 03:05:11 | rtsp    | listen addr=:8554
info | 2025-07-29 03:05:11 | webrtc  | listen addr=:8555/tcp
info | 2025-07-29 03:05:20 | startup | Starting go2rtc healthcheck service...

FFprobe output from your camera

Unable to run this since my credentials have a character that it can't escape and bash errors out

Install method

Docker Compose

Object Detector

CPU (no coral)

Network connection

Wired

Camera make and model

Amcrest IP5M-T1179EW-AI-V3 | Wyze Cam V3

Screenshots of the Frigate UI's System metrics pages

Any other information that may be helpful

I have found other discussions, and from what I can tell my config should be fine, but there could be something really silly I'm missing that I can't see in other posts such as #9846 for example

Answered by inkeep-ai-assistant[bot]

Jul 29, 2025

Your configuration uses preset-vaapi for hardware acceleration, which is the recommended setting for Intel CPUs including those with integrated GPUs like the N100(1). The device mapping - /dev/dri/renderD128:/dev/dri/renderD128 is also correct for exposing the Intel iGPU to the container(1).

However, high CPU usage (especially for object detection) is expected if you are using the CPU detector rather than a hardware-accelerated detector like OpenVINO or an Edge TPU(2). The preset-vaapi only accelerates video decoding in ffmpeg, not object detection. Your actual detection workload is still on the CPU, which matches what you are seeing in the Frigate UI metrics.

If you want to reduce CPU us…

View full answer

DashingBuffalo · 2025-07-29T03:29:57Z

inkeep-ai-assistant[bot]
Bot Jul 29, 2025

Your configuration uses preset-vaapi for hardware acceleration, which is the recommended setting for Intel CPUs including those with integrated GPUs like the N100(1). The device mapping - /dev/dri/renderD128:/dev/dri/renderD128 is also correct for exposing the Intel iGPU to the container(1).

However, high CPU usage (especially for object detection) is expected if you are using the CPU detector rather than a hardware-accelerated detector like OpenVINO or an Edge TPU(2). The preset-vaapi only accelerates video decoding in ffmpeg, not object detection. Your actual detection workload is still on the CPU, which matches what you are seeing in the Frigate UI metrics.

If you want to reduce CPU usage further, consider using OpenVINO as your detector if supported by your hardware. Example config for OpenVINO:

detectors:
  openvino:
    type: openvino
    device: AUTO

(2)

For ffmpeg video decoding, your current setup is correct for Intel iGPU acceleration(1). For object detection, switching to OpenVINO can provide additional hardware acceleration and reduce CPU load(2).

📚 Sources:

Have another question? Just tag @inkeep.

1 reply

DashingBuffalo Jul 29, 2025
Author

Thank you so much! I appreciate the insight and direction.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[HW Accel Support]: Need to verify if my configuration is setup properly to utilize my N100 iGPU #19311

Uh oh!

{{title}}

Uh oh!

Replies: 1 comment 1 reply

Uh oh!

{{title}}

Uh oh!

Uh oh!

{{title}}

Uh oh!

Select a reply

Uh oh!

Uh oh!

[HW Accel Support]: Need to verify if my configuration is setup properly to utilize my N100 iGPU #19311

Uh oh!

DashingBuffalo Jul 29, 2025

Describe the problem you are having

Version

Frigate config file

docker-compose file or Docker CLI command

Relevant Frigate log output

Relevant go2rtc log output

FFprobe output from your camera

Install method

Object Detector

Network connection

Camera make and model

Screenshots of the Frigate UI's System metrics pages

Any other information that may be helpful

Replies: 1 comment · 1 reply

Uh oh!

inkeep-ai-assistant[bot] Bot Jul 29, 2025

Uh oh!

DashingBuffalo Jul 29, 2025 Author

DashingBuffalo
Jul 29, 2025

Replies: 1 comment 1 reply

inkeep-ai-assistant[bot]
Bot Jul 29, 2025

DashingBuffalo Jul 29, 2025
Author