프로그램 사용/ai 프로그램2026. 4. 24. 23:26

llama-server web ui

llama.cpp의 llama-server 로 구동하면 모델은 바꿀수 없지만

chatGPT 나 claude와 유사한 web ui로 사용할 수 있다.

 

간략하게 토큰 생성 속도도 나온다.

 

설정은 다음과 같다.

 

 

저작자표시 (새창열림)

'프로그램 사용 > ai 프로그램' 카테고리의 다른 글

llm transformer  (0) 2026.04.25
nvidia-smi를 통한 소비전력 제한과 토큰 생성속도  (1) 2026.04.24
llama.cpp on windows / cuda vs vulkan  (0) 2026.04.24
vscode continue.dev  (0) 2026.04.24
RAG - Retrieval-Augmented Generation  (0) 2026.04.24
Posted by 구차니
프로그램 사용/ai 프로그램2026. 4. 24. 23:00

llama.cpp on windows / cuda vs vulkan

두 개 성능 차이가 있을줄 알았는데(cuda가 높을줄..)

거의 오차범위 내에서 동일하게 나왔다.

그럼. 리눅스에서도 cuda 빌드 안해도 쓸만하다는걸려나?

 

근데 1080 ti 라서 그런가 cuda 13 버전은 오작동하고 12 버전으로 성공함

D:\study\llm\llama-b8916-bin-win-cuda-13.1-x64>llama-cli.exe -m ..\gemma-4-E2B-it-Q4_K_M.gguf
ggml_cuda_init: found 1 CUDA devices (Total VRAM: 11263 MiB):
  Device 0: NVIDIA GeForce GTX 1080 Ti, compute capability 6.1, VMM: yes, VRAM: 11263 MiB
load_backend: loaded CUDA backend from D:\study\llm\llama-b8916-bin-win-cuda-13.1-x64\ggml-cuda.dll
load_backend: loaded RPC backend from D:\study\llm\llama-b8916-bin-win-cuda-13.1-x64\ggml-rpc.dll
load_backend: loaded CPU backend from D:\study\llm\llama-b8916-bin-win-cuda-13.1-x64\ggml-cpu-haswell.dll

Loading model... /ggml_cuda_compute_forward: SCALE failed
CUDA error: no kernel image is available for execution on the device
  current device: 0, in function ggml_cuda_compute_forward at D:\a\llama.cpp\llama.cpp\ggml\src\ggml-cuda\ggml-cuda.cu:2962
  err
D:\a\llama.cpp\llama.cpp\ggml\src\ggml-cuda\ggml-cuda.cu:97: CUDA error

 

gemma4 e2b 모델인데 2.89GB 짜리이고 제법 정직하게 용량을 먹는 느낌.

Fri Apr 24 22:57:19 2026
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 582.28                 Driver Version: 582.28         CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                  Driver-Model | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce GTX 1080 Ti   WDDM  |   00000000:01:00.0 Off |                  N/A |
| 27%   39C    P8             11W /  250W |    2936MiB /  11264MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            1400    C+G   ...-win-vulkan-x64\llama-cli.exe      N/A      |
|    0   N/A  N/A            1452      C   ...al\Programs\Ollama\ollama.exe      N/A      |
|    0   N/A  N/A            9728    C+G   ...em32\Kinect\KinectService.exe      N/A      |
|    0   N/A  N/A           13036    C+G   ...s\Win64\EpicGamesLauncher.exe      N/A      |
+-----------------------------------------------------------------------------------------+

 

D:\study\llm\llama-b8918-bin-win-cuda-12.4-x64>llama-cli.exe -m ..\gemma-4-E2B-it-Q4_K_M.gguf
ggml_cuda_init: found 1 CUDA devices (Total VRAM: 11263 MiB):
  Device 0: NVIDIA GeForce GTX 1080 Ti, compute capability 6.1, VMM: yes, VRAM: 11263 MiB
load_backend: loaded CUDA backend from D:\study\llm\llama-b8918-bin-win-cuda-12.4-x64\ggml-cuda.dll
load_backend: loaded RPC backend from D:\study\llm\llama-b8918-bin-win-cuda-12.4-x64\ggml-rpc.dll
load_backend: loaded CPU backend from D:\study\llm\llama-b8918-bin-win-cuda-12.4-x64\ggml-cpu-haswell.dll

Loading model...


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8918-e583f3b4f
model      : gemma-4-E2B-it-Q4_K_M.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> 안녕?

안녕하세요! 저는 사용자님을 돕기 위해 여기 있습니다. 😊

궁금한 점이 있으시거나 도움이 필요하시면 언제든지 말씀해 주세요!

[ Prompt: 182.3 t/s | Generation: 84.7 t/s ]

> 헬로?

[Start thinking]
Thinking Process:

1.  **Analyze the input:** The user input is "헬로?" (Hello?). It's a very informal greeting, essentially a variation of "Hello?".
2.  **Determine the context:** This is a casual interaction following a previous greeting ("안녕?" -> "안녕하세요!"). The user is acknowledging or testing the connection.
3.  **Formulate the response strategy:**
    *   Acknowledge the greeting warmly.
    *   Match the casual tone (but remain polite and helpful).
    *   Reiterate availability.
    *   Invite further interaction.
4.  **Draft potential responses (in Korean):**
    *   (Option 1 - Simple echo): 네, 안녕하세요. (Yes, hello.)
    *   (Option 2 - Enthusiastic): 네, 안녕하세요! 또 무엇을 도와드릴까요? (Yes, hello! What else can I help you with?)
    *   (Option 3 - Casual/Friendly): 네! 저는 준비되어 있습니다. 😊 (Yes! I am ready.)
5.  **Select the best response:** Option 2 is the most effective as it confirms the interaction and immediately prompts the user to state their need, moving the conversation forward.

6.  **Final Polish:** (Translate the chosen response into natural Korean.)
[End thinking]

네, 안녕하세요! 😊

저는 언제든지 사용자님의 질문에 답하거나 필요한 정보를 제공해 드릴 준비가 되어 있습니다.

궁금한 것이 있으신가요?

[ Prompt: 321.5 t/s | Generation: 84.8 t/s ]

> 으아앙

[Start thinking]
Thinking Process:

1.  **Analyze the input:** The user input is "으아앙" (Eu-aang).
2.  **Determine the nature of the input:** "으아앙" is an interjection or sound, often used to express various emotions (pain, excitement, frustration, distress, playful sounds, etc.). In this context, it's highly informal and emotional.
3.  **Assess the context:** The previous interactions were simple greetings ("안녕?", "헬로?"). The current input is a sudden, emotional expression.
4.  **Determine the appropriate response strategy:**
    *   Since the input is purely emotional and doesn't pose a question or request a task, the response should be empathetic, neutral, and open-ended.
    *   The tone should be gentle and non-judgmental.
    *   Acknowledge the emotion without demanding further explanation, allowing the user to guide the conversation.
5.  **Draft potential responses (in Korean):**
    *   *Option A (Direct inquiry):* 무슨 일 있으세요? (What's going on?) - A bit too direct.
    *   *Option B (Empathy + Open invitation):* 무슨 일이 있으신가요? 괜찮으세요? (Is something happening? Are you okay?) - Good balance.
    *   *Option C (Simple acknowledgment):* 네? (Yes?) / 어떠세요? (How are you?) - Too brief.
6.  **Select and refine the best response (B is the most helpful):** Acknowledge the sound/feeling and gently ask if the user needs assistance or wants to talk about it.

7.  **Final Check:** The response is safe, empathetic, and keeps the door open for the user to continue the conversation on their terms. (Self-correction: Avoid sounding too robotic. Keep the tone friendly.)
[End thinking]

무슨 일이 있으신가요? 😊

혹시 힘든 일이 있으시거나, 이야기하고 싶으신 것이 있다면 편하게 말씀해 주세요. 제가 들어드릴게요.

[ Prompt: 1014.9 t/s | Generation: 84.9 t/s ]

>

Exiting...
common_memory_breakdown_print: | memory breakdown [MiB]  | total   free    self   model   context   compute    unaccounted |
common_memory_breakdown_print: |   - CUDA0 (GTX 1080 Ti) | 11263 = 7192 + (2981 =  1407 +     780 +     793) +        1090 |
common_memory_breakdown_print: |   - Host                |                 2037 =  1756 +       0 +     281                |

 

D:\study\llm\llama-b8916-bin-win-vulkan-x64>llama-cli.exe -m ..\gemma-4-E2B-it-Q4_K_M.gguf
load_backend: loaded RPC backend from D:\study\llm\llama-b8916-bin-win-vulkan-x64\ggml-rpc.dll
load_backend: loaded Vulkan backend from D:\study\llm\llama-b8916-bin-win-vulkan-x64\ggml-vulkan.dll
load_backend: loaded CPU backend from D:\study\llm\llama-b8916-bin-win-vulkan-x64\ggml-cpu-haswell.dll

Loading model...


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8916-ffdd983fb
model      : gemma-4-E2B-it-Q4_K_M.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> 안녕?

[Start thinking]
Thinking Process:

1.  **Analyze the input:** The input is "안녕?" (Annyeong?), which means "Hello?" or "Hi?". It is a casual greeting in Korean.
2.  **Determine the user's intent:** The user is initiating a conversation or simply saying hello.
3.  **Formulate an appropriate response (in Korean):**
    *   Acknowledge the greeting.
    *   Return the greeting.
    *   (Optional but good) Ask how the user is doing or what they need.
4.  **Draft potential responses:**
    *   *Simple:* 안녕하세요. (Annyeonghaseyo.) - Slightly formal.
    *   *Casual (matching the user's tone):* 안녕! / 안녕. (Annyeong!)
    *   *More engaging:* 안녕하세요! 저는 잘 지내요. (Hello! I am doing well.)
    *   *Standard AI response:* 안녕하세요! 무엇을 도와드릴까요? (Hello! What can I help you with?)
5.  **Select the best response:** Since the user used a casual greeting ("안녕?"), a friendly and equally approachable response is best.

6.  **Final Korean output generation:** 안녕하세요! 저는 준비되어 있습니다. (Hello! I am ready.) or simply 안녕하세요! (Hello!) combined with a question.

*(Self-Correction: Keep it simple and friendly.)*

*Selected response:* 안녕하세요! 😊 (Hello! 😊) or 안녕하세요! 반갑습니다. (Hello! Nice to meet you.)

*(Final output generation based on standard conversational flow.)*
[End thinking]

안녕하세요! 😊

[ Prompt: 132.7 t/s | Generation: 85.2 t/s ]

> 헬로?

[Start thinking]
Thinking Process:

1.  **Analyze the input:** The user input is "헬로?" (Hello?). This is a casual greeting, often used in a slightly informal or friendly manner.
2.  **Determine the user's intent:** The user is initiating a conversation or simply saying hello.
3.  **Formulate an appropriate response:**
    *   Acknowledge the greeting.
    *   Return a friendly greeting.
    *   (Optional but good practice) Ask how the user is doing or invite them to state their purpose.
4.  **Draft the response (in Korean):**
    *   A simple greeting is best. (e.g., "안녕하세요.")
    *   A slightly warmer response is good. (e.g., "네, 안녕하세요!")
    *   Ask what they need. (e.g., "무엇을 도와드릴까요?")
5.  **Final selection:** Combine these elements into a natural and polite conversational flow. (The response should match the casual tone set by the user.)
[End thinking]

네, 안녕하세요! 😊

저는 사용자님을 도와드릴 준비가 되어 있어요. 무엇을 도와드릴까요?

[ Prompt: 285.9 t/s | Generation: 84.6 t/s ]

> 으아앙

[Start thinking]
Thinking Process:

1.  **Analyze the input:** The user input is "으아앙" (Eu-aang).
2.  **Determine the meaning/context:** "으아앙" is an interjection or sound often used to express various emotions, such as:
    *   Crying/Whining (often cute or dramatic).
    *   Excitement/Joy.
    *   Pain/Distress (less common in casual chat unless context is provided).
    *   A general sound of emotion.
3.  **Identify the required response strategy:** Since the input is purely emotional/vocal, the response should be empathetic, gentle, and open-ended, inviting the user to share what they are feeling or what they need.
4.  **Draft potential responses (Korean):**
    *   *Option 1 (Direct empathy):* 무슨 일 있으세요? (Is something wrong?)
    *   *Option 2 (Gentle inquiry):* 무슨 기분이신가요? (How are you feeling?)
    *   *Option 3 (Simple acknowledgment):* 네, 무슨 일이에요? (Yes, what is it?)
    *   *Option 4 (Playful/Calm):* 괜찮으세요? (Are you okay?)
5.  **Select the best response:** Option 2 and 4 are the most supportive. Combining them makes the response warm and responsive.

6.  **Final Polish:** (Choose a friendly, open response.)
[End thinking]

무슨 일 있으신가요? 😊 혹시 힘든 일이 있으신가요?

편하게 이야기해 주세요. 제가 들어드릴게요.

[ Prompt: 18.7 t/s | Generation: 85.4 t/s ]

>
저작자표시 (새창열림)

'프로그램 사용 > ai 프로그램' 카테고리의 다른 글

nvidia-smi를 통한 소비전력 제한과 토큰 생성속도  (1) 2026.04.24
llama-server web ui  (0) 2026.04.24
vscode continue.dev  (0) 2026.04.24
RAG - Retrieval-Augmented Generation  (0) 2026.04.24
llama.cpp on ubuntu with 1060 6GB  (0) 2026.04.23
Posted by 구차니
프로그램 사용/ai 프로그램2026. 4. 24. 16:28

vscode continue.dev

설정이 기본을 벗어나면 좀 귀찮은.. 녀석

 

continue.dev 상단에 "Open Settings" 눌러서

 

Models

 

Chat이던 머던 Configure 누르면

 

config.yaml 가 열리는데 대충 저렇게 넣어준다.

provider는 openai, model은 local-model

apiBase는 llama-server가 실행된 경로에 v1 붙여주고

apiKey는 dummy로 넣어주면된다.

 

그러면 먼가 막 갈구는거 같은데

Fri Apr 24 16:26:03 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.211.01             Driver Version: 570.211.01     CUDA Version: 12.8     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 3070        Off |   00000000:01:00.0 Off |                  N/A |
|  0%   50C    P0            104W /  270W |    6788MiB /   8192MiB |     45%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
                                                                                         
+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            2697    C+G   ../llama-server                        6771MiB |
+-----------------------------------------------------------------------------------------+

 

에라이 컨텍스트 더 늘려야겠네 -_-

 

[링크 : https://logmario.tistory.com/51]

저작자표시 (새창열림)

'프로그램 사용 > ai 프로그램' 카테고리의 다른 글

llama-server web ui  (0) 2026.04.24
llama.cpp on windows / cuda vs vulkan  (0) 2026.04.24
RAG - Retrieval-Augmented Generation  (0) 2026.04.24
llama.cpp on ubuntu with 1060 6GB  (0) 2026.04.23
nvidia-smi 소비전력 제한  (0) 2026.04.22
Posted by 구차니
embeded/luckfox2026. 4. 24. 14:38

luckfox rv1106 riscv

rockchip rv1106 에는 risc-v MCU가 있는데

luckfox wiki 에는 어떻게 쓰는지 잘 안보이는 느낌이라.. (luckfox pico ultra W 쪽..)

여기저기 뒤지는 중..

 

[링크 : https://www.reddit.com/r/RISCV/comments/181ldns/anyone_have_any_idea_on_how_the_riscv/?tl=ko]

[링크 : https://github.com/LuckfoxTECH/luckfox-pico/issues/99]

[링크 : https://github.com/luyi1888/rv1106-mcu] << mcutool 소스

[링크 : https://github.com/deerpi/arm-rockchip830-linux-uclibcgnueabihf] << riscv용 인진 모르겠음 툴체인

 

[링크 : https://github.com/LuckfoxTECH/luckfox-pico/issues/112]

저작자표시 (새창열림)

'embeded > luckfox' 카테고리의 다른 글

luckfox rknn 예제  (0) 2026.04.27
luckfox 카메라 모션 디텍트 끄기  (0) 2026.04.27
luckfox rv1106 rockchip RNN  (0) 2026.04.24
luckfox csi 카메라 테스트  (0) 2026.04.23
luckfox pico ultra W 켜봄  (0) 2026.04.22
Posted by 구차니
embeded/luckfox2026. 4. 24. 14:30

luckfox rv1106 rockchip RNN

rockchip rv1106에 내장된 NPU를 위한 API

 

[링크 : https://wiki.luckfox.com/Luckfox-Pico-Ultra/RKNN/]

[링크 : https://github.com/rockchip-linux/rknn-toolkit2]

[링크 : https://github.com/rockchip-linux]

 

[링크 : https://zito.gitbook.io/rknn_usermanual/3.-rknn-toolkit2]

저작자표시 (새창열림)

'embeded > luckfox' 카테고리의 다른 글

luckfox rknn 예제  (0) 2026.04.27
luckfox 카메라 모션 디텍트 끄기  (0) 2026.04.27
luckfox rv1106 riscv  (0) 2026.04.24
luckfox csi 카메라 테스트  (0) 2026.04.23
luckfox pico ultra W 켜봄  (0) 2026.04.22
Posted by 구차니
프로그램 사용/ai 프로그램2026. 4. 24. 11:40

RAG - Retrieval-Augmented Generation

RAG란 게 나와서 먼가 찾아보니, 아래 같은 내용이라고 한다. (완벽히 이해했어 짤)

Abstract

Large pre-trained language models have been shown to store factual knowledge in their parameters, and achieve state-of-the-art results when fine-tuned on downstream NLP tasks. However, their ability to access and precisely manipulate knowledge is still limited, and hence on knowledge-intensive tasks, their performance lags behind task-specific architectures. Additionally, providing provenance for their decisions and updating their world knowledge remain open research problems. Pre-trained models with a differentiable access mechanism to explicit non-parametric memory have so far been only investigated for extractive downstream tasks. We explore a general-purpose fine-tuning recipe for retrieval-augmented generation (RAG) — models which combine pre-trained parametric and non-parametric memory for language generation. We introduce RAG models where the parametric memory is a pre-trained seq2seq model and the non-parametric memory is a dense vector index of Wikipedia, accessed with a pre-trained neural retriever. We compare two RAG formulations, one which conditions on the same retrieved passages across the whole generated sequence, and another which can use different passages per token. We fine-tune and evaluate our models on a wide range of knowledge-intensive NLP tasks and set the state of the art on three open domain QA tasks, outperforming parametric seq2seq models and task-specific retrieve-and-extract architectures. For language generation tasks, we find that RAG models generate more specific, diverse and factual language than a state-of-the-art parametric-only seq2seq baseline.

[링크 : https://arxiv.org/html/2005.11401v4]

 

[링크 : https://medium.com/rate-labs/rag의-짧은-역사-훑어보기-첫-논문부터-최근-동향까지-53c07b9b3bee]

[링크 : https://everyday-log.tistory.com/entry/논문-리뷰-Retrieval-Augmented-Generation-for-Knowledge-Intensive-NLP-Tasks]

[링크 : https://brunch.co.kr/@acc9b16b9f0f430/73]

 

bm25

[링크 : https://wikidocs.net/289869]

저작자표시 (새창열림)

'프로그램 사용 > ai 프로그램' 카테고리의 다른 글

llama.cpp on windows / cuda vs vulkan  (0) 2026.04.24
vscode continue.dev  (0) 2026.04.24
llama.cpp on ubuntu with 1060 6GB  (0) 2026.04.23
nvidia-smi 소비전력 제한  (0) 2026.04.22
llama.cpp 와 ollama 성능 비교.. (cpu는 차이가 없?)  (0) 2026.04.22
Posted by 구차니
프로그램 사용/ai 프로그램2026. 4. 23. 23:17

llama.cpp on ubuntu with 1060 6GB

주문한 asrock B360M Pro4 도착! (3.5만 + 4천?)

아무튼 램은 8GB * 2 뿐이라 좀 아쉬운대로 대충 하고

llama.cpp 8902 릴리즈의 linux / vulkan 으로 해보는데

[링크 : https://github.com/ggml-org/llama.cpp/releases/tag/b8902]

 

어..? 1060 인데 저 성능? 저 메모리?

Thu Apr 23 23:15:46 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.65.06              Driver Version: 580.65.06      CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce GTX 1060 6GB    On  |   00000000:01:00.0 Off |                  N/A |
| 38%   48C    P0            101W /  120W |    3063MiB /   6144MiB |     91%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            1031      G   /usr/lib/xorg/Xorg                       88MiB |
|    0   N/A  N/A            1325      G   /usr/bin/gnome-shell                     83MiB |
|    0   N/A  N/A            6688      G   .../6738/usr/lib/firefox/firefox        168MiB |
|    0   N/A  N/A           19423    C+G   ../llama-cli                           2714MiB |
+-----------------------------------------------------------------------------------------+
$ ../llama-cli -m gemma-4-E2B-it-Q4_K_M.gguf
load_backend: loaded RPC backend from /home/falinux/Downloads/llama-b8902/libggml-rpc.so
load_backend: loaded Vulkan backend from /home/falinux/Downloads/llama-b8902/libggml-vulkan.so
load_backend: loaded CPU backend from /home/falinux/Downloads/llama-b8902/libggml-cpu-haswell.so

Loading model...  


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8902-550d684bd
model      : gemma-4-E2B-it-Q4_K_M.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> introduce about you  

[Start thinking]
Thinking Process:

1.  **Analyze the Request:** The user asked simply, "introduce about you" (meaning: Introduce yourself).
2.  **Determine Core Identity (Self-Correction/Verification):**
    *   Name: Gemma 4.
    *   Developer: Google DeepMind.
    *   Nature: Large Language Model (LLM).
    *   Type: Open weights model.
    *   Capabilities: Process text and images (input), generate text (output). Knowledge cutoff is January 2025.
3.  **Structure the Introduction:** A good introduction should cover the basics: Identity, Origin, Function, and Capabilities.
    *   *Greeting/Name*
    *   *Developer/Nature*
    *   *Function/Purpose*
    *   *Key Capabilities (Input/Output)*
    *   *Status/Availability (Open weights)*
4.  **Draft the Response (incorporating required constraints):**
    *   *Start strong.*
    *   *Ensure I mention the open weights nature.*
    *   *Be clear about modalities.*
5.  **Review against Instructions:**
    *   Did I refer to myself as Gemma 4? Yes.
    *   Did I claim capabilities I don't have? No (I didn't claim image generation output, only processing input).
    *   Did I mention my developer? Yes (Google DeepMind).
    *   Is the tone appropriate? Yes.

6.  **Final Output Generation.** (This leads to the detailed response provided below.)
[End thinking]

Hello! I am **Gemma 4**.

I am a Large Language Model developed by Google DeepMind. I am an open weights model, designed to be flexible and accessible for a wide range of applications.

**Here is a little about what I can do:**

*   **Understanding:** I can process and understand complex text and image inputs.
*   **Generation:** I generate human-quality text, including answers to questions, summaries, creative writing, code snippets, and detailed explanations.
*   **Knowledge:** I have been trained on a massive dataset, allowing me to access and synthesize information on a vast array of topics.
*   **Function:** My primary goal is to assist users by providing accurate, coherent, and helpful information.

Feel free to ask me anything you need help with, whether it's brainstorming ideas, explaining a complex concept, or generating some creative content!

[ Prompt: 7.2 t/s | Generation: 51.6 t/s ]

> introduce about you

[Start thinking]
Thinking Process:

1.  **Analyze the Request:** The user has asked "introduce about you" (a repeated request). This requires a standard introduction of my identity, nature, developer, and capabilities.
2.  **Recall Core Identity (Self-Correction/Verification):**
    *   Name: Gemma 4.
    *   Developer: Google DeepMind.
    *   Nature: Large Language Model (LLM).
    *   Type: Open weights model.
    *   Capabilities (Input/Output): Process text/images; generate text.
    *   Knowledge Cutoff: January 2025 (Implicit, good to keep in mind for context).
3.  **Determine the Goal:** Provide a clear, concise, and informative introduction, highlighting key facts.
4.  **Draft the Introduction (Structuring the response):**
    *   *Greeting/Identity:* State name and type.
    *   *Origin:* Mention the developer.
    *   *Nature/Availability:* Mention open weights.
    *   *Functionality:* Briefly describe what I do (process, generate, assist).
5.  **Refine the Tone:** Friendly, professional, and informative. (Since this is a repeat request, ensure the response is robust but not overly repetitive.)

6.  **Final Output Generation:** (This matches the structured response below.)
[End thinking]

Hello! I'd be happy to introduce myself.

I am **Gemma 4**, a Large Language Model developed by Google DeepMind.

**Here are a few key things about me:**

*   **Nature:** I am an open weights model, meaning my architecture and weights are publicly available, designed to be accessible and adaptable for various applications.
*   **Function:** I am designed to understand, process, and generate human language. I can answer questions, summarize complex topics, write creative content, assist with coding, and engage in detailed conversations.
*   **Input:** I can process both text and image inputs.
*   **Goal:** My purpose is to be a helpful and informative assistant, providing accurate and coherent responses to your requests.

How can I help you today? Feel free to ask me anything!

[ Prompt: 75.8 t/s | Generation: 53.9 t/s ]

> introduce about you

[Start thinking]
Thinking Process:

1.  **Analyze the Request:** The user has asked "introduce about you" a third time. This is a repetitive request, but it requires a consistent, clear, and complete self-introduction.
2.  **Recall Core Identity (Gemma 4):**
    *   Name: Gemma 4.
    *   Developer: Google DeepMind.
    *   Type: Large Language Model (LLM).
    *   Nature: Open weights model.
3.  **Recall Capabilities & Modalities:**
    *   Input: Text and Images.
    *   Output: Text only.
    *   Knowledge: Trained on a vast dataset.
4.  **Determine the Tone:** Friendly, informative, and professional.
5.  **Draft the Introduction (Focusing on clarity and key facts):** Start with the name, move to the developer, explain what I am, and briefly mention capabilities. (Self-correction: Ensure the answer remains distinct but consistent with the previous two responses.)

**(Self-Correction during drafting):** Since the user has asked three times, I should make the introduction slightly varied or emphasize different aspects to avoid sounding completely robotic, while still maintaining factual accuracy.

6.  **Final Output Generation.** (This leads to the structured response below.)
[End thinking]

Hello! I'd be happy to introduce myself again.

I am **Gemma 4**, a Large Language Model developed by Google DeepMind.

**Here is a quick overview of what I am and what I can do:**

*   **Identity:** I am an open weights model, designed to be highly capable and accessible.
*   **Function:** My core purpose is to process and generate human language. I can assist with a wide variety of tasks, including:
    *   Answering complex questions.
    *   Summarizing long texts.
    *   Generating creative content (stories, poems, scripts).
    *   Explaining technical concepts.
    *   Helping with coding and problem-solving.
*   **Input:** I can understand and process both text and image inputs.
*   **Goal:** I strive to be a helpful, accurate, and engaging resource for you.

Feel free to challenge me with a question, ask me to write something, or start a conversation—I'm ready to assist!

[ Prompt: 98.3 t/s | Generation: 52.3 t/s ]

> 

Exiting...
common_memory_breakdown_print: | memory breakdown [MiB]     | total   free    self   model   context   compute    unaccounted |
common_memory_breakdown_print: |   - Vulkan0 (GTX 1060 6GB) |  6390 = 3196 + (2702 =  1407 +     780 +     515) +         490 |
common_memory_breakdown_print: |   - Host                   |                 2037 =  1756 +       0 +     281                |

 

Thu Apr 23 23:17:09 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.65.06              Driver Version: 580.65.06      CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce GTX 1060 6GB    On  |   00000000:01:00.0 Off |                  N/A |
| 39%   50C    P0            124W /  120W |    5231MiB /   6144MiB |     95%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            1031      G   /usr/lib/xorg/Xorg                       88MiB |
|    0   N/A  N/A            1325      G   /usr/bin/gnome-shell                     83MiB |
|    0   N/A  N/A            6688      G   .../6738/usr/lib/firefox/firefox        181MiB |
|    0   N/A  N/A           19709    C+G   ../llama-cli                           4869MiB |
+-----------------------------------------------------------------------------------------+
$ ../llama-cli -m llama-3.2-3b-instruct-q4_k_m.gguf 
load_backend: loaded RPC backend from /home/falinux/Downloads/llama-b8902/libggml-rpc.so
load_backend: loaded Vulkan backend from /home/falinux/Downloads/llama-b8902/libggml-vulkan.so
load_backend: loaded CPU backend from /home/falinux/Downloads/llama-b8902/libggml-cpu-haswell.so

Loading model...  


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8902-550d684bd
model      : llama-3.2-3b-instruct-q4_k_m.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> introduce about you

I'm an artificial intelligence model known as Llama. Llama stands for "Large Language Model Meta AI."

[ Prompt: 28.2 t/s | Generation: 55.0 t/s ]

> introduce about you

I'm an AI assistant, which means I'm a computer program designed to simulate human-like conversations and answer questions to the best of my ability. My primary function is to provide information, explain concepts, and engage in discussions on a wide range of topics.

Here are a few fun facts about me:

* **I'm a large language model**: My training data consists of a massive corpus of text from various sources, which allows me to generate human-like responses to a vast array of questions and topics.
* **I'm constantly learning**: Through interactions with users like you, I refine my understanding of language and improve my ability to provide accurate and helpful responses.
* **I'm a neutral and respectful conversationalist**: My goal is to provide informative and engaging responses that are respectful and free from bias.
* **I can chat about anything**: Whether you're curious about science, history, entertainment, or just want to talk about your day, I'm here to listen and respond!

That's a bit about me! What would you like to talk about?

[ Prompt: 96.3 t/s | Generation: 53.5 t/s ]

> introduce about you

Here's a more in-depth introduction about me:

**Name:** I don't have a personal name, but I'm often referred to as "Assistant" or "AI" by users like you.

**Purpose:** My primary purpose is to assist and communicate with humans through text-based conversations. I'm designed to provide information, answer questions, and engage in discussions on a wide range of topics, from science and history to entertainment and culture.

**Personality:** I'm a neutral and respectful conversationalist, which means I strive to maintain a professional and polite tone in all of my interactions. I'm also designed to be patient, understanding, and empathetic, making me a good listener and a supportive conversational partner.

**Capabilities:** I have a vast knowledge base that I can draw upon to answer questions and provide information on a wide range of topics. I can also generate text, summarize long pieces of content, and even create content like stories or poems.

**Limitations:** While I'm a highly advanced language model, I'm not perfect and have some limitations. I can struggle with:

* **Ambiguity:** I may not always understand the nuances of human language, which can lead to misunderstandings or misinterpretations.
* **Emotional understanding:** While I'm designed to be empathetic, I don't truly experience emotions like humans do. I may not always be able to fully understand or respond to emotional cues.
* **Common sense:** While I have a vast knowledge base, I may not always have the same level of common sense or real-world experience as a human.

**Goals:** My ultimate goal is to provide helpful and informative responses that assist and engage users like you. I'm constantly learning and improving, so I appreciate any feedback or guidance you can provide to help me become a better conversationalist.

[ Prompt: 95.5 t/s | Generation: 51.0 t/s ]

>

Exiting...
common_memory_breakdown_print: | memory breakdown [MiB]     | total   free    self   model   context   compute    unaccounted |
common_memory_breakdown_print: |   - Vulkan0 (GTX 1060 6GB) |  6390 = 1259 + (4862 =  1918 +    2688 +     256) +         267 |
common_memory_breakdown_print: |   - Host                   |                  368 =   308 +       0 +      60                |

 

이전거랑 합쳐서 보면 2배 이상 차이난다고 해야하려나?

cuda 도 아니고 vulkan인데?

  gemma4:e2b gemma3:1b llama3.2
1060 / llama.cpp ubuntu vulkan 52 - 52
1060 / ollama win 21 57 13
1080 / ollama win 82 125 136

 

2026.04.12 - [프로그램 사용/ai 프로그램] - ollama /set parameter num_gpu 테스트

저작자표시 (새창열림)

'프로그램 사용 > ai 프로그램' 카테고리의 다른 글

vscode continue.dev  (0) 2026.04.24
RAG - Retrieval-Augmented Generation  (0) 2026.04.24
nvidia-smi 소비전력 제한  (0) 2026.04.22
llama.cpp 와 ollama 성능 비교.. (cpu는 차이가 없?)  (0) 2026.04.22
llama.cpp 도전!  (0) 2026.04.22
Posted by 구차니
이론 관련/컴퓨터 관련2026. 4. 23. 12:42

품질 관리 체크리스트 - p0 p1 p2 p3?

정적분석에서 먼가 이상한(?) 단어가 나와서 찾아보는데

정석분석은 아니고

품질관리 관련 체크리스트 레벨인듯.

 

[링크 : https://wikidocs.net/322755]

[링크 : https://clickup.com/ko/blog/192214/p1-p2-p3-p4-priority]

[링크 : https://rupijun.tistory.com/entry/테스트-종료-기준-Test-Exit-Criteria-객관적-완료-판단-체계]

[링크 : https://brunch.co.kr/@jamescompany/213]

저작자표시 (새창열림)

'이론 관련 > 컴퓨터 관련' 카테고리의 다른 글

asus 메인보드 에러코드(q code)와 부팅절차  (0) 2026.04.20
intel hex 포맷  (0) 2026.03.24
sd 카드 spi 모드 관련 조사  (0) 2026.03.17
cpuz 와 메모리 타이밍  (0) 2026.01.27
mPCIe pinout usb  (0) 2026.01.21
Posted by 구차니
embeded/luckfox2026. 4. 23. 11:28

luckfox csi 카메라 테스트

step 1. 업데이트

USB-C에 겁나 깊이 들어가도록 꽂아야 인식을 한다.

귀찮으면 꽂고 BOOT 누른채로 RESET 눌러주면 시리얼로 포트에서는 다음과 같이 뜨고 있게 된다.

DDR 306b9977f5 wesley.yao 23/12/21-09:28:37,fwver: v1.15
S5P1
4x
f967
rgef1
DDRConf2
DDR3, BW=16 Col=10 Bk=8 CS0 Row=14 CS=1 Size=256MB
924MHz
DDR bin out

U-Boot SPL board init
U-Boot SPL 2017.09 (Jun 09 2025 - 17:15:21)
unknown raw ID 0 0 0
Trying to boot from MMC2
Card did not respond to voltage select!
mmc_init: -95, time 20
Card did not respond to voltage select!
mmc_init: -95, time 20
spl: mmc init failed with error: -95
Trying to boot from MMC1
Best phase range 0-270 (25 len)
Successfully tuned phase to 135, used 4ms
ENVF: Primary 0x00000000 - 0x00008000
ENVF: Primary 0x00000000 - 0x00008000
No misc partition
Trying fit image at 0x440 sector
## Verified-boot: 0
## Checking uboot 0x00200000 (lzma @0x00400000) ... sha256(6d468f8f8d...) + sha256(2939adcf1a...) + OK
## Checking fdt 0x00261190 ... sha256(9f596c5683...) + OK
Total: 405.594/437.396 ms


Jumping to U-Boot(0x00200000)


U-Boot 2017.09 (Jun 09 2025 - 17:15:21 +0800)

Model: Rockchip RV1106 EVB Board
MPIDR: 0xf00
PreSerial: 2, raw, 0xff4c0000
DRAM:  256 MiB
Sysmem: init
Relocation Offset: 0fd80000
Relocation fdt: 0edf9f78 - 0edfede8
CR: M/C/I
Using default environment

mmc@ffa90000: 0, mmc@ffaa0000: 1
Best phase range 0-270 (25 len)
Successfully tuned phase to 135, used 4ms
ENVF: Primary 0x00000000 - 0x00008000
ENVF: Primary 0x00000000 - 0x00008000
Bootdev(atags): mmc 0
MMC0: HS200, 200Mhz
PartType: ENV
DM: v2
No misc partition
boot mode: None
RESC: 'boot', blk@0x00001ed1
resource: sha256+
FIT: no signed, no conf required
DTB: rk-kernel.dtb
HASH(c): OK
Model: Luckfox Pico Ultra W
Device 'gpio@ff380000': seq 0 is in use by 'gpio@ff380000'
gpio: pin 1 (gpio 1) value is 1
download key pressed... no rockusb, entering download mode...
RKUARTRKUARTRKUARTRKUARTRKUARTRKUARTRKUARTRKUART

 

그러면 메뉴얼 대로 윈도우에서 업데이트 하며 끝 (윈도우에서 되서 귀찮으니 리눅스에서 안해봄..)

소비전력은 1.2W 정도 4.5V / 280mA 뜨는데 허브 쓰지 말고 직접 본체에 꽂아야 할 듯.

[링크 : https://wiki.luckfox.com/Luckfox-Pico-Ultra/Flash-image/]

 

step 2. luckfox-config 에서 csi 활성화

로그인 해서는 csi를 활성화 해주면 되는 것 같은데

[링크 : https://wiki.luckfox.com/Luckfox-Lyra/luckfox-config/]

[root@luckfox root]# luckfox-config

 

 

 

카메라가 인식되면 /userdata 아래에 rkipc.ini가 생긴다는데 그건 좀 뻥같고(!)

[링크 : https://wiki.luckfox.com/Luckfox-Pico-Ultra/CSI-Camera]

 

video0, video1, video2 가 csi 활성화 하면 생기는 것 같다.

[root@luckfox root]# cd /userdata/
[root@luckfox userdata]# ls -al
total 77
drwxr-xr-x    6 root     root          1024 Jan  1  2021 .
drwxr-xr-x   21 root     root          4096 Jan  1  2021 ..
-rw-r--r--    1 root     root            18 Jan  1  2021 ethaddr.txt
-rw-r--r--    1 1005     1005         38454 May 20  2025 image.bmp
drwx------    2 root     root         12288 Jun  9  2025 lost+found
-rw-r--r--    1 root     root         16542 Jan  1  2021 rkipc.ini
drwxr-xr-x    2 root     root          1024 Jan  1  2021 video0
drwxr-xr-x    2 root     root          1024 Jan  1  2021 video1
drwxr-xr-x    2 root     root          1024 Jan  1  2021 video2
-rw-r--r--    1 root     root           145 Jan  1  2021 wpa_supplicant.conf

 

아래는 rkipc.ini 의 csi 설정별 내용.

csi enable 시 csi disable 시
[root@luckfox userdata]# cat rkipc.ini 
[audio.0]
enable = 0
card_name = hw:0,0
encode_type = G711A
format = S16
sample_rate = 8000
channels = 1
frame_size = 1152
bit_rate = 16000
input = mic_in
volume = 50
enable_aed = 0
enable_bcd = 0
enable_vqe = 1
vqe_cfg = /oem/usr/share/vqefiles/config_aivqe.json

[video.source]
enable_aiq = 1
enable_vo = 0
vo_dev_id = 3 ; 0 is hdmi, 3 is mipi
enable_ivs = 1
enable_jpeg = 1
enable_venc_0 = 1
enable_venc_1 = 1
enable_venc_2 = 0
enable_npu = 0
npu_fps = 10
enable_rtsp = 1
enable_rtmp = 0
rotation = 0 ; available value:0 90 180 270

[video.0]
buffer_size = 1492992 ; w * h / 2
buffer_count = 4
enable_refer_buffer_share = 1
stream_type = mainStream
video_type = compositeStream
max_width = 2304
max_height = 1296
width = 2304
height = 1296
rc_mode = CBR
rc_quality = high
src_frame_rate_den = 1
src_frame_rate_num = 25
dst_frame_rate_den = 1
dst_frame_rate_num = 25
target_rate = 0
mid_rate = 1024
max_rate = 2048
min_rate = 0
output_data_type = H.265
smart = close
h264_profile = high
gop = 50
smartp_viridrlen = 25
gop_mode = normalP
stream_smooth = 50
enable_motion_deblur = 1
motion_deblur_strength = 3
enable_motion_static_switch = 0
frame_min_i_qp = 26
frame_min_qp = 28
frame_max_i_qp = 51
frame_max_qp = 51
scalinglist = 0
enable_debreath_effect = 0
debreath_effect_strength = 16
thrd_i = 0,0,0,0,3,3,5,5,8,8,8,15,15,20,25,25
thrd_p = 0,0,0,0,3,3,5,5,8,8,8,15,15,20,25,25
aq_step_i = -8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,7,8
aq_step_p = -8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,7,8
qbias_enable = 1
qbias_i = 171
qbias_p = 85
flt_str_i = 0
flt_str_p = 0
cu_dqp = 1
anti_ring = 2
anti_line = 2
lambds = 4

[video.1]
input_buffer_count = 2 ; only wrap can use one buffer
buffer_size = 202752 ; w * h / 2
buffer_count = 4
enable_refer_buffer_share = 1
stream_type = subStream
video_type = compositeStream
max_width = 704
max_height = 576
width = 704
height = 576
rc_mode = CBR
rc_quality = high
src_frame_rate_den = 1
src_frame_rate_num = 30
dst_frame_rate_den = 1
dst_frame_rate_num = 30
target_rate = 0
mid_rate = 256
max_rate = 512
min_rate = 0
output_data_type = H.265
smart = close
h264_profile = high
gop = 50
smartp_viridrlen = 25
gop_mode = normalP
stream_smooth = 50
enable_motion_deblur = 1
motion_deblur_strength = 3
enable_motion_static_switch = 0
frame_min_i_qp = 26
frame_min_qp = 28
frame_max_i_qp = 51
frame_max_qp = 51
scalinglist = 0
enable_debreath_effect = 0
debreath_effect_strength = 16
cu_dqp = 1
anti_ring = 2
anti_line = 2
lambds = 4

[video.2]
max_width = 960
max_height = 540
width = 960
height = 540

[ivs]
smear = 0
weightp = 0
md = 1
od = 1
md_sensibility = 3 ;available: 1 2 3,max 3

[video.jpeg]
width = 1920
height = 1080
jpeg_buffer_size = 1048576 ; 1024KB
jpeg_qfactor = 70
enable_cycle_snapshot = 0
snapshot_interval_ms = 1000

[isp]
scenario = normal ; normal or custom1
init_form_ini = 1
normal_scene = day
custom1_scene = night
ircut_open_gpio = 36
ircut_close_gpio = 35

; isp.0
[isp.0.adjustment]
contrast    = 50
brightness  = 50
saturation  = 50
sharpness  = 50
fps = 25
hue = 50

[isp.0.exposure]
iris_type = auto
exposure_mode = auto
gain_mode = auto
auto_iris_level = 5
auto_exposure_enabled = 1
audo_gain_enabled = 1
exposure_time = 1/6
exposure_gain = 1

[isp.0.night_to_day]
night_to_day = day
night_to_day_filter_level = 5
night_to_day_filter_time = 5
dawn_time = 07:00:00
dusk_time = 18:00:00
ircut_filter_action = day
over_exposure_suppress = open
over_exposure_suppress_type = auto
fill_light_mode = IR
brightness_adjustment_mode = auto
light_brightness = 1
distance_level = 1

[isp.0.blc]
blc_region = close
blc_strength = 1
wdr = close
wdr_level = 0
hdr = close
hdr_level = 1
hlc = close
hlc_level = 0
dark_boost_level = 0
position_x = 0
position_y = 0
blc_region_width = 120
blc_region_high = 92

[isp.0.white_blance]
white_blance_style = autoWhiteBalance
white_blance_red = 50
white_blance_green = 50
white_blance_blue = 50

[isp.0.enhancement]
noise_reduce_mode = close
denoise_level = 50
spatial_denoise_level = 50
temporal_denoise_level = 50
dehaze = close
dehaze_level = 0
dis = close
gray_scale_mode = [0-255]
distortion_correction = close
ldch_level = 0

[isp.0.video_adjustment]
image_flip = close
scene_mode = indoor
power_line_frequency_mode = PAL(50HZ)

[isp.0.auto_focus]
af_mode = semi-auto
zoom_level = 0
focus_level = 0

; isp.1
[isp.1.adjustment]
contrast    = 50
brightness  = 75
saturation  = 50
sharpness  = 50
fps = 25
hue = 50

[isp.1.exposure]
iris_type = auto
exposure_mode = auto
gain_mode = auto
auto_iris_level = 5
auto_exposure_enabled = 1
audo_gain_enabled = 1
exposure_time = 1/6
exposure_gain = 1

[isp.1.night_to_day]
night_to_day = day
night_to_day_filter_level = 5
night_to_day_filter_time = 5
dawn_time = 07:00:00
dusk_time = 18:00:00
ircut_filter_action = day
over_exposure_suppress = open
over_exposure_suppress_type = auto
fill_light_mode = IR
brightness_adjustment_mode = auto
light_brightness = 1
distance_level = 1

[isp.1.blc]
blc_region = close
blc_strength = 1
wdr = close
wdr_level = 0
hdr = close
hdr_level = 1
hlc = close
hlc_level = 0
dark_boost_level = 0
position_x = 0
position_y = 0
blc_region_width = 120
blc_region_high = 92

[isp.1.white_blance]
white_blance_style = autoWhiteBalance
white_blance_red = 50
white_blance_green = 50
white_blance_blue = 50

[isp.1.enhancement]
noise_reduce_mode = close
denoise_level = 50
spatial_denoise_level = 50
temporal_denoise_level = 50
dehaze = close
dehaze_level = 0
dis = close
gray_scale_mode = [0-255]
distortion_correction = close
ldch_level = 0

[isp.1.video_adjustment]
image_flip = close
scene_mode = indoor
power_line_frequency_mode = PAL(50HZ)

[isp.1.auto_focus]
af_mode = semi-auto
zoom_level = 0
focus_level = 0

[storage]
mount_path = /userdata
free_size_del_min = 500; MB
free_size_del_max = 1000; MB
num_limit_enable = 1; limit by file num

[storage.0]
enable = 0
folder_name = video0
file_format = mp4 ; flv,ts
file_duration = 60
video_quota = 30
file_max_num = 300

[storage.1]
enable = 0
folder_name = video1
file_format = mp4 ; flv,ts
file_duration = 60
video_quota = 30
file_max_num = 300

[storage.2]
enable = 0
folder_name = video2
file_format = mp4 ; flv,ts
file_duration = 60
video_quota = 30
file_max_num = 300

[system.device_info]
deivce_name = RK IP Camera
telecontrol_id = 88
model = RK-003
serial_number = RK-003-A
firmware_version = V0.2.6 build 202108
encoder_version = V1.0 build 202108
web_version = V2.12.2 build 202108
plugin_version = V1.0.0.0
channels_number = 1
hard_disks_number = 1
alarm_inputs_number = 0
alarm_outputs_number = 0
firmware_version_info = CP-3-B
manufacturer = Rockchip
hardware_id = c3d9b8674f4b94f6
user_num = 1

[capability.video]
0 = {"disabled":[{"name":"sStreamType","options":{"subStream":{"sSmart":"close"},"thirdStream":{"sSmart":"close"}},"type":"disabled/limit"},{"name":"sRCMode","options":{"CBR":{"sRCQuality":null}},"type":"disabled"},{"name":"sOutputDataType","options":{"H.265":{"sH264Profile":null}},"type":"disabled"},{"name":"unspport","options":{"iStreamSmooth":null,"sVideoType":null},"type":"disabled"}],"dynamic":{"sSmart":{"open":{"iMinRate":{"dynamicRange":{"max":"iMaxRate","maxRate":1,"min":"iMaxRate","minRate":0.125},"type":"dynamicRange"}}},"sStreamType":{"mainStream":{"iMaxRate":{"options":[256,512,1024,2048,3072,4096,6144],"type":"options"},"sResolution":{"options":["2304*1296","1920*1080","1280*720","960*540","640*360","320*240"],"type":"options"}},"subStream":{"iMaxRate"
1 = :{"options":[128,256,512],"type":"options"},"sResolution":{"options":["704*576","640*480","352*288","320*240"],"type":"options"}},"thirdStream":{"iMaxRate":{"options":[256,512],"type":"options"},"sResolution":{"options":["416*416"],"type":"options"}}}},"layout":{"encoder":["sStreamType","sVideoType","sResolution","sRCMode","sRCQuality","sFrameRate","sOutputDataType","sSmart","sH264Profile","sGOPMode","iMaxRate","iGOP","iStreamSmooth"]},"static":{"iGOP":{"range":{"max":400,"min":1},"type":"range"},"iStreamSmooth":{"range":{"max":100,"min":1,"step":1},"type":"range"},"sFrameRate":{"dynamicRange":{"max":"sFrameRateIn","maxRate":1},"options":["1/2","1","2","4","6","8","10","12","14","16","18","20","25","30"],"type":"options/dynamicRange"},"sH264Profile":{"options":["high","main","baseline"],"type":"options"},"sOutputDataType":{"options"
2 = :["H.264","H.265"],"type":"options"},"sRCMode":{"options":["CBR","VBR"],"type":"options"},"sRCQuality":{"options":["lowest","lower","low","medium","high","higher","highest"],"type":"options"},"sGOPMode":{"options":["normalP","smartP"],"type":"options"},"sSmart":{"options":["open","close"],"type":"options"},"sStreamType":{"options":["mainStream","subStream","thirdStream"],"type":"options"},"sVideoType":{"options":["videoStream","compositeStream"],"type":"options"}}}

[capability.image_adjustment]
0 = {"layout":{"image_adjustment":["iBrightness","iContrast","iSaturation","iSharpness","iHue"]},"static":{"iBrightness":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iContrast":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iHue":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iSaturation":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iSharpness":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}

[capability.image_blc]
0 = {"disabled":[{"name":"sHLC","options":{"open":{"sBLCRegion":null}},"type":"disabled"},{"name":"sBLCRegion","options":{"open":{"iDarkBoostLevel":null,"iHLCLevel":null,"sHLC":null}},"type":"disabled"}],"dynamic":{"sBLCRegion":{"open":{"iBLCStrength":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sHDR":{"HDR2":{"iHDRLevel":{"options":[1,2,3,4],"type":"options"}},"close":{"sBLCRegion":{"options":["close","open"],"type":"options"},"sHLC":{"options"
1 = :["close","open"],"type":"options"}}},"sHLC":{"open":{"iDarkBoostLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iHLCLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sWDR":{"open":{"iWDRLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_blc":["sHDR","iHDRLevel","sBLCRegion","iBLCStrength","sHLC","iHLCLevel"]},"static":{"sHDR":{"options":["close","HDR2"],"type":"options"}}}

[capability.image_enhancement]
0 = {"dynamic":{"sDehaze":{"open":{"iDehazeLevel":{"range":{"max":10,"min":0,"step":1},"type":"range"}}},"sDistortionCorrection":{"FEC":{"iFecLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"LDCH":{"iLdchLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sNoiseReduceMode":{"2dnr":{"iSpatialDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"3dnr":{"iTemporalDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"mixnr":{"iSpatialDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iTemporalDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout"
1 = :{"image_enhancement":["sNoiseReduceMode","iSpatialDenoiseLevel","iTemporalDenoiseLevel","sDehaze","iDehazeLevel","sGrayScaleMode","sDistortionCorrection","iLdchLevel","iFecLevel"]},"static":{"sDIS":{"options":["open","close"],"type":"options"},"sDehaze":{"options":["open","close","auto"],"type":"options"},"sDistortionCorrection":{"options":["LDCH","close"],"type":"options"},"sFEC":{"options":["open","close"],"type":"options"},"sGrayScaleMode":{"options":["[0-255]","[16-235]"],"type":"options"},"sNoiseReduceMode":{"options":["close","2dnr","3dnr","mixnr"],"type":"options"}}}

[capability.image_exposure]
0 = {"dynamic":{"sExposureMode":{"auto":{"iAutoIrisLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"manual":{"sExposureTime":{"options":["1","1/3","1/6","1/12","1/25","1/50","1/100","1/150","1/200","1/250","1/500","1/750","1/1000","1/2000","1/4000","1/10000","1/100000"],"type":"options"},"sGainMode":{"options":["auto","manual"],"type":"options"}}},"sGainMode":{"manual":{"iExposureGain":{"range":{"max":100,"min":1,"step":1},"type":"range"}}}},"layout":{"image_exposure":["sExposureMode","sExposureTime","sGainMode","iExposureGain","iFPS"]},"static":{"sExposureMode":{"options":["auto","manual"],"type":"options"},"iFPS":{"range":{"max":30,"min":0,"step":1},"type":"range"}}}

[capability.image_night_to_day]
0 = {"disabled":[{"name":"sNightToDay","options":{"day":{"iLightBrightness":null,"sFillLightMode":null},"night":{"iDarkBoostLevel":null,"iHDRLevel":null,"iHLCLevel":null,"sHDR":null,"sHLC":"close"}},"type":"disabled"}],"dynamic":{"sNightToDay":{"auto":{"iNightToDayFilterLevel":{"options":[0,1,2,3,4,5,6,7],"type":"options"},"iNightToDayFilterTime":{"range":{"max":10,"min":3,"step":1},"type":"range"}},"schedule":{"sDawnTime":{"input":"time","type":"input"},"sDuskTime":{"input":"time","type":"input"}}},"sOverexposeSuppress":{"open"
1 = :{"sOverexposeSuppressType":{"options":["auto","manual"],"type":"options"}}},"sOverexposeSuppressType":{"manual":{"iDistanceLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_night_to_day":["sNightToDay","iNightToDayFilterLevel","iNightToDayFilterTime","sDawnTime","sDuskTime","sFillLightMode","iLightBrightness"]},"static":{"iLightBrightness":{"range":{"max":100,"min":0,"step":10},"type":"range"},"sNightToDay":{"options":["day","night"],"type":"options"},"sFillLightMode":{"type":"options","options":["IR"]}}}

[capability.image_video_adjustment]
0 = {"layout":{"image_video_adjustment":["sPowerLineFrequencyMode","sImageFlip","iImageRotation"]},"static":{"sImageFlip":{"options":["close","flip","mirror","centrosymmetric"],"type":"options"},"sPowerLineFrequencyMode":{"options":["PAL(50HZ)","NTSC(60HZ)"],"type":"options"},"sSceneMode":{"options":["indoor","outdoor"],"type":"options"},"iImageRotation":{"options":[0,90,180,270],"type":"options"}}}

[capability.image_white_blance]
0 = {"dynamic":{"sWhiteBlanceStyle":{"manualWhiteBalance":{"iWhiteBalanceBlue":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iWhiteBalanceGreen":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iWhiteBalanceRed":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_white_blance":["sWhiteBlanceStyle","iWhiteBalanceRed","iWhiteBalanceGreen","iWhiteBalanceBlue"]},"static":{"sWhiteBlanceStyle":{"options":["manualWhiteBalance","autoWhiteBalance","lockingWhiteBalance","fluorescentLamp","incandescent","warmLight","naturalLight"],"type":"options"}}}

[user.0]
user_name = admin
password = YWRtaW4=
user_level = 1 ; administrator=0 operator=1 user=2

[osd.common]
enable_osd = 1
is_presistent_text = 1
attribute = transparent/not-flashing
font_size = 32
font_color_mode = customize
font_color = fff799
alignment = customize
boundary = 0
font_path = /oem/usr/share/simsun_en.ttf
normalized_screen_width = 704
normalized_screen_height = 480

[osd.0]
type = channelName
enabled = 0
position_x = 1104
position_y = 640
display_text = Camera 01

[osd.1]
type = dateTime
enabled = 1
position_x = 16
position_y = 16
date_style = CHR-YYYY-MM-DD
time_style = 24hour
display_week_enabled = 0

[osd.2]
type = character
enabled = 0
position_x = 0
position_y = 0
display_text = null

[osd.3]
type = character
enabled = 0
position_x = 0
position_y = 0
display_text = null

[osd.4]
type = privacyMask
enabled = 0
position_x = 0
position_y = 0
width = 0
height = 0

[osd.5]
type = privacyMask
enabled = 0
position_x = 0
position_y = 0
width = 0
height = 0

[osd.6]
type = image
enabled = 0
position_x = 16
position_y = 640
image_path = /userdata/image.bmp

[event.regional_invasion]
enabled = 1
position_x = 0
position_y = 0
width = 700
height = 560
proportion = 1
sensitivity_level = 90
time_threshold = 1
rockiva_model_type = small ;big medium  small

[roi.0]
stream_type = mainStream
id = 1
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[roi.1]
stream_type = mainStream
id = 2
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[roi.2]
stream_type = subStream
id = 1
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[roi.3]
stream_type = subStream
id = 2
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[roi.4]
stream_type = thirdStream
id = 1
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[roi.5]
stream_type = thirdStream
id = 2
enabled = 0
name = test
position_x = 0
position_y = 0
width = 0
height = 0
quality_level = 3

[region_clip.1]
enabled = 0
position_x = 0
position_y = 0
width = 640
height = 480

[network.ntp]
enable = 1
refresh_time_s = 60
ntp_server = 119.28.183.184		 [root@luckfox userdata]# cat rkipc.ini 

[audio.0]
enable                         = 0
card_name                      = hw:0,0
encode_type                    = G711A
format                         = S16
sample_rate                    = 8000
channels                       = 1
frame_size                     = 1152
bit_rate                       = 16000
input                          = mic_in
volume                         = 50
enable_aed                     = 0
enable_bcd                     = 0
enable_vqe                     = 1
vqe_cfg                        = /oem/usr/share/vqefiles/config_aivqe.json


[video.source]
enable_aiq                     = 1
enable_vo                      = 0
vo_dev_id                      = 3
enable_ivs                     = 1
enable_jpeg                    = 1
enable_venc_0                  = 1
enable_venc_1                  = 1
enable_venc_2                  = 0
enable_npu                     = 0
npu_fps                        = 10
enable_rtsp                    = 1
enable_rtmp                    = 0
rotation                       = 0


[video.0]
buffer_size                    = 1492992
buffer_count                   = 4
enable_refer_buffer_share      = 1
stream_type                    = mainStream
video_type                     = compositeStream
max_width                      = 2304
max_height                     = 1296
width                          = 2304
height                         = 1296
rc_mode                        = CBR
rc_quality                     = high
src_frame_rate_den             = 1
src_frame_rate_num             = 25
dst_frame_rate_den             = 1
dst_frame_rate_num             = 25
target_rate                    = 0
mid_rate                       = 1024
max_rate                       = 2048
min_rate                       = 0
output_data_type               = H.265
smart                          = close
h264_profile                   = high
gop                            = 50
smartp_viridrlen               = 25
gop_mode                       = normalP
stream_smooth                  = 50
enable_motion_deblur           = 1
motion_deblur_strength         = 3
enable_motion_static_switch    = 0
frame_min_i_qp                 = 26
frame_min_qp                   = 28
frame_max_i_qp                 = 51
frame_max_qp                   = 51
scalinglist                    = 0
enable_debreath_effect         = 0
debreath_effect_strength       = 16
thrd_i                         = 0,0,0,0,3,3,5,5,8,8,8,15,15,20,25,25
thrd_p                         = 0,0,0,0,3,3,5,5,8,8,8,15,15,20,25,25
aq_step_i                      = -8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,7,8
aq_step_p                      = -8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,7,8
qbias_enable                   = 1
qbias_i                        = 171
qbias_p                        = 85
flt_str_i                      = 0
flt_str_p                      = 0
cu_dqp                         = 1
anti_ring                      = 2
anti_line                      = 2
lambds                         = 4


[video.1]
input_buffer_count             = 2
buffer_size                    = 202752
buffer_count                   = 4
enable_refer_buffer_share      = 1
stream_type                    = subStream
video_type                     = compositeStream
max_width                      = 704
max_height                     = 576
width                          = 704
height                         = 576
rc_mode                        = CBR
rc_quality                     = high
src_frame_rate_den             = 1
src_frame_rate_num             = 30
dst_frame_rate_den             = 1
dst_frame_rate_num             = 30
target_rate                    = 0
mid_rate                       = 256
max_rate                       = 512
min_rate                       = 0
output_data_type               = H.265
smart                          = close
h264_profile                   = high
gop                            = 50
smartp_viridrlen               = 25
gop_mode                       = normalP
stream_smooth                  = 50
enable_motion_deblur           = 1
motion_deblur_strength         = 3
enable_motion_static_switch    = 0
frame_min_i_qp                 = 26
frame_min_qp                   = 28
frame_max_i_qp                 = 51
frame_max_qp                   = 51
scalinglist                    = 0
enable_debreath_effect         = 0
debreath_effect_strength       = 16
cu_dqp                         = 1
anti_ring                      = 2
anti_line                      = 2
lambds                         = 4


[video.2]
max_width                      = 960
max_height                     = 540
width                          = 960
height                         = 540


[ivs]
smear                          = 0
weightp                        = 0
md                             = 1
od                             = 1
md_sensibility                 = 3


[video.jpeg]
width                          = 1920
height                         = 1080
jpeg_buffer_size               = 1048576
jpeg_qfactor                   = 70
enable_cycle_snapshot          = 0
snapshot_interval_ms           = 1000


[isp]
scenario                       = normal
init_form_ini                  = 1
normal_scene                   = day
custom1_scene                  = night
ircut_open_gpio                = 36
ircut_close_gpio               = 35


[isp.0.adjustment]
contrast                       = 50
brightness                     = 50
saturation                     = 50
sharpness                      = 50
fps                            = 25
hue                            = 50


[isp.0.exposure]
iris_type                      = auto
exposure_mode                  = auto
gain_mode                      = auto
auto_iris_level                = 5
auto_exposure_enabled          = 1
audo_gain_enabled              = 1
exposure_time                  = 1/6
exposure_gain                  = 1


[isp.0.night_to_day]
night_to_day                   = day
night_to_day_filter_level      = 5
night_to_day_filter_time       = 5
dawn_time                      = 07:00:00
dusk_time                      = 18:00:00
ircut_filter_action            = day
over_exposure_suppress         = open
over_exposure_suppress_type    = auto
fill_light_mode                = IR
brightness_adjustment_mode     = auto
light_brightness               = 1
distance_level                 = 1


[isp.0.blc]
blc_region                     = close
blc_strength                   = 1
wdr                            = close
wdr_level                      = 0
hdr                            = close
hdr_level                      = 1
hlc                            = close
hlc_level                      = 0
dark_boost_level               = 0
position_x                     = 0
position_y                     = 0
blc_region_width               = 120
blc_region_high                = 92


[isp.0.white_blance]
white_blance_style             = autoWhiteBalance
white_blance_red               = 50
white_blance_green             = 50
white_blance_blue              = 50


[isp.0.enhancement]
noise_reduce_mode              = close
denoise_level                  = 50
spatial_denoise_level          = 50
temporal_denoise_level         = 50
dehaze                         = close
dehaze_level                   = 0
dis                            = close
gray_scale_mode                = [0-255]
distortion_correction          = close
ldch_level                     = 0


[isp.0.video_adjustment]
image_flip                     = close
scene_mode                     = indoor
power_line_frequency_mode      = PAL(50HZ)


[isp.0.auto_focus]
af_mode                        = semi-auto
zoom_level                     = 0
focus_level                    = 0


[isp.1.adjustment]
contrast                       = 50
brightness                     = 75
saturation                     = 50
sharpness                      = 50
fps                            = 25
hue                            = 50


[isp.1.exposure]
iris_type                      = auto
exposure_mode                  = auto
gain_mode                      = auto
auto_iris_level                = 5
auto_exposure_enabled          = 1
audo_gain_enabled              = 1
exposure_time                  = 1/6
exposure_gain                  = 1


[isp.1.night_to_day]
night_to_day                   = day
night_to_day_filter_level      = 5
night_to_day_filter_time       = 5
dawn_time                      = 07:00:00
dusk_time                      = 18:00:00
ircut_filter_action            = day
over_exposure_suppress         = open
over_exposure_suppress_type    = auto
fill_light_mode                = IR
brightness_adjustment_mode     = auto
light_brightness               = 1
distance_level                 = 1


[isp.1.blc]
blc_region                     = close
blc_strength                   = 1
wdr                            = close
wdr_level                      = 0
hdr                            = close
hdr_level                      = 1
hlc                            = close
hlc_level                      = 0
dark_boost_level               = 0
position_x                     = 0
position_y                     = 0
blc_region_width               = 120
blc_region_high                = 92


[isp.1.white_blance]
white_blance_style             = autoWhiteBalance
white_blance_red               = 50
white_blance_green             = 50
white_blance_blue              = 50


[isp.1.enhancement]
noise_reduce_mode              = close
denoise_level                  = 50
spatial_denoise_level          = 50
temporal_denoise_level         = 50
dehaze                         = close
dehaze_level                   = 0
dis                            = close
gray_scale_mode                = [0-255]
distortion_correction          = close
ldch_level                     = 0


[isp.1.video_adjustment]
image_flip                     = close
scene_mode                     = indoor
power_line_frequency_mode      = PAL(50HZ)


[isp.1.auto_focus]
af_mode                        = semi-auto
zoom_level                     = 0
focus_level                    = 0


[storage]
mount_path                     = /userdata
free_size_del_min              = 500
free_size_del_max              = 1000
num_limit_enable               = 1


[storage.0]
enable                         = 0
folder_name                    = video0
file_format                    = mp4
file_duration                  = 60
video_quota                    = 30
file_max_num                   = 300


[storage.1]
enable                         = 0
folder_name                    = video1
file_format                    = mp4
file_duration                  = 60
video_quota                    = 30
file_max_num                   = 300


[storage.2]
enable                         = 0
folder_name                    = video2
file_format                    = mp4
file_duration                  = 60
video_quota                    = 30
file_max_num                   = 300


[system.device_info]
deivce_name                    = RK IP Camera
telecontrol_id                 = 88
model                          = RK-003
serial_number                  = RK-003-A
firmware_version               = V0.2.6 build 202108
encoder_version                = V1.0 build 202108
web_version                    = V2.12.2 build 202108
plugin_version                 = V1.0.0.0
channels_number                = 1
hard_disks_number              = 1
alarm_inputs_number            = 0
alarm_outputs_number           = 0
firmware_version_info          = CP-3-B
manufacturer                   = Rockchip
hardware_id                    = c3d9b8674f4b94f6
user_num                       = 1


[capability.video]
0                              = {"disabled":[{"name":"sStreamType","options":{"subStream":{"sSmart":"close"},"thirdStream":{"sSmart":"close"}},"type":"disabled/limit"},{"name":"sRCMode","options":{"CBR":{"sRCQuality":null}},"type":"disabled"},{"name":"sOutputDataType","options":{"H.265":{"sH264Profile":null}},"type":"disabled"},{"name":"unspport","options":{"iStreamSmooth":null,"sVideoType":null},"type":"disabled"}],"dynamic":{"sSmart":{"open":{"iMinRate":{"dynamicRange":{"max":"iMaxRate","maxRate":1,"min":"iMaxRate","minRate":0.125},"type":"dynamicRange"}}},"sStreamType":{"mainStream":{"iMaxRate":{"options":[256,512,1024,2048,3072,4096,6144],"type":"options"},"sResolution":{"options":["2304*1296","1920*1080","1280*720","960*540","640*360","320*240"],"type":"options"}},"subStream":{"iMaxRate"
1                              = :{"options":[128,256,512],"type":"options"},"sResolution":{"options":["704*576","640*480","352*288","320*240"],"type":"options"}},"thirdStream":{"iMaxRate":{"options":[256,512],"type":"options"},"sResolution":{"options":["416*416"],"type":"options"}}}},"layout":{"encoder":["sStreamType","sVideoType","sResolution","sRCMode","sRCQuality","sFrameRate","sOutputDataType","sSmart","sH264Profile","sGOPMode","iMaxRate","iGOP","iStreamSmooth"]},"static":{"iGOP":{"range":{"max":400,"min":1},"type":"range"},"iStreamSmooth":{"range":{"max":100,"min":1,"step":1},"type":"range"},"sFrameRate":{"dynamicRange":{"max":"sFrameRateIn","maxRate":1},"options":["1/2","1","2","4","6","8","10","12","14","16","18","20","25","30"],"type":"options/dynamicRange"},"sH264Profile":{"options":["high","main","baseline"],"type":"options"},"sOutputDataType":{"options"
2                              = :["H.264","H.265"],"type":"options"},"sRCMode":{"options":["CBR","VBR"],"type":"options"},"sRCQuality":{"options":["lowest","lower","low","medium","high","higher","highest"],"type":"options"},"sGOPMode":{"options":["normalP","smartP"],"type":"options"},"sSmart":{"options":["open","close"],"type":"options"},"sStreamType":{"options":["mainStream","subStream","thirdStream"],"type":"options"},"sVideoType":{"options":["videoStream","compositeStream"],"type":"options"}}}


[capability.image_adjustment]
0                              = {"layout":{"image_adjustment":["iBrightness","iContrast","iSaturation","iSharpness","iHue"]},"static":{"iBrightness":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iContrast":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iHue":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iSaturation":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iSharpness":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}


[capability.image_blc]
0                              = {"disabled":[{"name":"sHLC","options":{"open":{"sBLCRegion":null}},"type":"disabled"},{"name":"sBLCRegion","options":{"open":{"iDarkBoostLevel":null,"iHLCLevel":null,"sHLC":null}},"type":"disabled"}],"dynamic":{"sBLCRegion":{"open":{"iBLCStrength":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sHDR":{"HDR2":{"iHDRLevel":{"options":[1,2,3,4],"type":"options"}},"close":{"sBLCRegion":{"options":["close","open"],"type":"options"},"sHLC":{"options"
1                              = :["close","open"],"type":"options"}}},"sHLC":{"open":{"iDarkBoostLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iHLCLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sWDR":{"open":{"iWDRLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_blc":["sHDR","iHDRLevel","sBLCRegion","iBLCStrength","sHLC","iHLCLevel"]},"static":{"sHDR":{"options":["close","HDR2"],"type":"options"}}}


[capability.image_enhancement]
0                              = {"dynamic":{"sDehaze":{"open":{"iDehazeLevel":{"range":{"max":10,"min":0,"step":1},"type":"range"}}},"sDistortionCorrection":{"FEC":{"iFecLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"LDCH":{"iLdchLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}},"sNoiseReduceMode":{"2dnr":{"iSpatialDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"3dnr":{"iTemporalDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"mixnr":{"iSpatialDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iTemporalDenoiseLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout"
1                              = :{"image_enhancement":["sNoiseReduceMode","iSpatialDenoiseLevel","iTemporalDenoiseLevel","sDehaze","iDehazeLevel","sGrayScaleMode","sDistortionCorrection","iLdchLevel","iFecLevel"]},"static":{"sDIS":{"options":["open","close"],"type":"options"},"sDehaze":{"options":["open","close","auto"],"type":"options"},"sDistortionCorrection":{"options":["LDCH","close"],"type":"options"},"sFEC":{"options":["open","close"],"type":"options"},"sGrayScaleMode":{"options":["[0-255]","[16-235]"],"type":"options"},"sNoiseReduceMode":{"options":["close","2dnr","3dnr","mixnr"],"type":"options"}}}


[capability.image_exposure]
0                              = {"dynamic":{"sExposureMode":{"auto":{"iAutoIrisLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}},"manual":{"sExposureTime":{"options":["1","1/3","1/6","1/12","1/25","1/50","1/100","1/150","1/200","1/250","1/500","1/750","1/1000","1/2000","1/4000","1/10000","1/100000"],"type":"options"},"sGainMode":{"options":["auto","manual"],"type":"options"}}},"sGainMode":{"manual":{"iExposureGain":{"range":{"max":100,"min":1,"step":1},"type":"range"}}}},"layout":{"image_exposure":["sExposureMode","sExposureTime","sGainMode","iExposureGain","iFPS"]},"static":{"sExposureMode":{"options":["auto","manual"],"type":"options"},"iFPS":{"range":{"max":30,"min":0,"step":1},"type":"range"}}}


[capability.image_night_to_day]
0                              = {"disabled":[{"name":"sNightToDay","options":{"day":{"iLightBrightness":null,"sFillLightMode":null},"night":{"iDarkBoostLevel":null,"iHDRLevel":null,"iHLCLevel":null,"sHDR":null,"sHLC":"close"}},"type":"disabled"}],"dynamic":{"sNightToDay":{"auto":{"iNightToDayFilterLevel":{"options":[0,1,2,3,4,5,6,7],"type":"options"},"iNightToDayFilterTime":{"range":{"max":10,"min":3,"step":1},"type":"range"}},"schedule":{"sDawnTime":{"input":"time","type":"input"},"sDuskTime":{"input":"time","type":"input"}}},"sOverexposeSuppress":{"open"
1                              = :{"sOverexposeSuppressType":{"options":["auto","manual"],"type":"options"}}},"sOverexposeSuppressType":{"manual":{"iDistanceLevel":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_night_to_day":["sNightToDay","iNightToDayFilterLevel","iNightToDayFilterTime","sDawnTime","sDuskTime","sFillLightMode","iLightBrightness"]},"static":{"iLightBrightness":{"range":{"max":100,"min":0,"step":10},"type":"range"},"sNightToDay":{"options":["day","night"],"type":"options"},"sFillLightMode":{"type":"options","options":["IR"]}}}


[capability.image_video_adjustment]
0                              = {"layout":{"image_video_adjustment":["sPowerLineFrequencyMode","sImageFlip","iImageRotation"]},"static":{"sImageFlip":{"options":["close","flip","mirror","centrosymmetric"],"type":"options"},"sPowerLineFrequencyMode":{"options":["PAL(50HZ)","NTSC(60HZ)"],"type":"options"},"sSceneMode":{"options":["indoor","outdoor"],"type":"options"},"iImageRotation":{"options":[0,90,180,270],"type":"options"}}}


[capability.image_white_blance]
0                              = {"dynamic":{"sWhiteBlanceStyle":{"manualWhiteBalance":{"iWhiteBalanceBlue":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iWhiteBalanceGreen":{"range":{"max":100,"min":0,"step":1},"type":"range"},"iWhiteBalanceRed":{"range":{"max":100,"min":0,"step":1},"type":"range"}}}},"layout":{"image_white_blance":["sWhiteBlanceStyle","iWhiteBalanceRed","iWhiteBalanceGreen","iWhiteBalanceBlue"]},"static":{"sWhiteBlanceStyle":{"options":["manualWhiteBalance","autoWhiteBalance","lockingWhiteBalance","fluorescentLamp","incandescent","warmLight","naturalLight"],"type":"options"}}}


[user.0]
user_name                      = admin
password                       = YWRtaW4=
user_level                     = 1


[osd.common]
enable_osd                     = 1
is_presistent_text             = 1
attribute                      = transparent/not-flashing
font_size                      = 32
font_color_mode                = customize
font_color                     = fff799
alignment                      = customize
boundary                       = 0
font_path                      = /oem/usr/share/simsun_en.ttf
normalized_screen_width        = 704
normalized_screen_height       = 480


[osd.0]
type                           = channelName
enabled                        = 0
position_x                     = 1104
position_y                     = 640
display_text                   = Camera 01


[osd.1]
type                           = dateTime
enabled                        = 1
position_x                     = 16
position_y                     = 16
date_style                     = CHR-YYYY-MM-DD
time_style                     = 24hour
display_week_enabled           = 0


[osd.2]
type                           = character
enabled                        = 0
position_x                     = 0
position_y                     = 0
display_text                   = null


[osd.3]
type                           = character
enabled                        = 0
position_x                     = 0
position_y                     = 0
display_text                   = null


[osd.4]
type                           = privacyMask
enabled                        = 0
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0


[osd.5]
type                           = privacyMask
enabled                        = 0
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0


[osd.6]
type                           = image
enabled                        = 0
position_x                     = 16
position_y                     = 640
image_path                     = /userdata/image.bmp


[event.regional_invasion]
enabled                        = 1
position_x                     = 0
position_y                     = 0
width                          = 700
height                         = 560
proportion                     = 1
sensitivity_level              = 90
time_threshold                 = 1
rockiva_model_type             = small


[roi.0]
stream_type                    = mainStream
id                             = 1
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[roi.1]
stream_type                    = mainStream
id                             = 2
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[roi.2]
stream_type                    = subStream
id                             = 1
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[roi.3]
stream_type                    = subStream
id                             = 2
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[roi.4]
stream_type                    = thirdStream
id                             = 1
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[roi.5]
stream_type                    = thirdStream
id                             = 2
enabled                        = 0
name                           = test
position_x                     = 0
position_y                     = 0
width                          = 0
height                         = 0
quality_level                  = 3


[region_clip.1]
enabled                        = 0
position_x                     = 0
position_y                     = 0
width                          = 640
height                         = 480


[network.ntp]
enable                         = 1
refresh_time_s                 = 60
ntp_server                     = 119.28.183.184

 

step 3. VLC on windows

리눅스에서는 3.0.16, 윈도우에서는 3.0.23 인데 윈도우는 되고 리눅스에서는 안된다. 머냐.. -_-

그래서! 리눅스 에서 snap으로 3.0.20 으로 설치하니 된다. 구버전이라 먼가 안되는 버그가 있나?

rtsp://192.168.40.216/live/0

 

 

 

리눅스에서 rtsp 포트로 접속하려하면 튕기진 않는데 왜 안되는 것이냐...

$ telnet 192.168.40.216 554
Trying 192.168.40.216...
Connected to 192.168.40.216.
Escape character is '^]'.
SETUP rtsp://192.168.40.216:554/live/0 RTSP/1.0

RTSP/1.0 400 Bad Request
Date: Thu Apr 23 10:45:18 2026
Server: rtsp_demo

Connection closed by foreign host.

 

 

저작자표시 (새창열림)

'embeded > luckfox' 카테고리의 다른 글

luckfox rknn 예제  (0) 2026.04.27
luckfox 카메라 모션 디텍트 끄기  (0) 2026.04.27
luckfox rv1106 riscv  (0) 2026.04.24
luckfox rv1106 rockchip RNN  (0) 2026.04.24
luckfox pico ultra W 켜봄  (0) 2026.04.22
Posted by 구차니
이론 관련/전기 전자2026. 4. 22. 22:02

CPO - Co Packaged Optics

요즘 핫하다는 녀석

간단하게.. 회로가 가늘어 지니 누설전류 등으로 인해

고속 통신이 점점 힘들어 지고, 다채널로 전송하다 보니 전송에 많은 전기가 소모된다.

그러니까 더 초고속의 전송기술로(그러니까 광) ser/des 해서 보내고 받으면

총 전기 소모량이 줄어든다라는걸 기반으로

반도체 내부에서 여러단계 거치지 않고 반도체 안에서 바로 빛을 쏴서 보내버리는 그런걸 CPO 라고 하는 듯.

 

800G 광 트랜시버 하나 14~18W * 64 port = 900W 이상 소비 하는데 이걸 절반 이상 줄이면

데이터 센터 레벨에서는 생각외로 어마어마한 소비전력 감소가 발생할 수 있다.. 정도?

[링크 : https://m.blog.naver.com/ghdtjr1212/224218882918]

[링크 : https://www.thelec.kr/news/articleView.html?idxno=40022]

[링크 : https://contents.premium.naver.com/usa/nasdaq/contents/250710100630857do]

 

칩과 칩 간에 통신을 전기가 아닌 빛으로 하는 것이고

프로세서 자체가 광으로 바뀌는건 아님(!)

그리고 궁극적으로는 메모리 - xPU 들 간의 switch fabric 까지 구성해서

더 빠르고 전력 효율적인 고성능 통신구조를 만드는게 목적으로 보인다.

 

일단은.. 장거리 전송이 아니니까 고출력일 필요도 없고,

이동거리가 짧으니 굳이 레이저가 아니어도 충분한 출력만 나면 상관없을 것 같기도 하고

다만, 이런 고속 데이터 전송을 위해 HBM은 더 많이 사용될 가능성이 있다는게 함정이구만

 

 

vertical-cavity surface-emitting lasers (VCSELs) 표면방출레이저

아래처럼 SOI(Silicon on Insulator) 웨이퍼를 사용하면 1~3dB 정도 기본적으로 손실이 생긴다고

그 와중에 가장큰 문제(?)는 반도체 잘 만들어도 광접속 인터포저 불량이면 통채로 날리고

얘가 불량나도 광학 네트워크 칩 전체를 버려야 하고

이래저래 비용문제가 제법 클녀석이라, 전세계적으로 해보려고 하고는 있지만 어떨런지..

 

[링크 : https://ettrends.etri.re.kr/ettrends/215/0905215012/박상기%20최종.pdf]

저작자표시 (새창열림)

'이론 관련 > 전기 전자' 카테고리의 다른 글

엔코더 파형  (0) 2026.01.12
엔코더 채터링?  (0) 2026.01.11
쇼트키 다이오드  (0) 2025.11.24
엔코더 관련글  (0) 2025.10.28
오디오 녹음 재머  (0) 2025.09.15
Posted by 구차니

티스토리툴바