Installed RTX gpu

[DrWH]

Hi the client does not see my new GPU (RTX 3060) on Windows 10 Pro

I deleted the old _internal folder and replaced it with the _internal folder from DeepFaceLab_NVIDIA_RTX3000_series_build_11_20_2021

[Lenny4]

Do you have a message saying "can't detect a GPU", or it's loading forever with the blue spinning circle ?

[DrWH]

On 2024-03-04 12:42, Alexandre Beaujour wrote: Do you have a message saying "can't detect a GPU", or it's loading forever with the blue spinning circle ? -- Reply to this email directly, view it on GitHub [1], or unsubscribe [2]. You are receiving this because you authored the thread.Message ID: ***@***.***> Links: ------ [1] #77 (comment) [2] https://github.com/notifications/unsubscribe-auth/AESPXXQCBLJ7SGR4777LIZTYWRT3JAVCNFSM6AAAAABEC64ZW6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZWGQ4TSNJTGI

A message, "can't detect a GPU"

[Lenny4]

Seems to be related to #59:

@DrWH when you use DeepFaceLab WITHOUT the client and you run data_src faceset extract.bat does DeepFaceLab ask you the question Choose one or several GPU idx (separated with comma) ?

If so does it show you your GPU in the available option ?

[DrWH]

On 2024-03-04 14:25, Alexandre Beaujour wrote: Seems to be related to #59 [1]: @DrWH [2] when you use DeepFaceLab WITHOUT the client and you run data_src faceset extract.bat does DeepFaceLab ask you the question Choose one or several GPU idx (separated with comma) ? If so does it show you your GPU in the available option ? Capture.PNG (view on web) [3] -- Reply to this email directly, view it on GitHub [4], or unsubscribe [5]. You are receiving this because you were mentioned.Message ID: ***@***.***> Links: ------ [1] #59 [2] https://github.com/DrWH [3] https://github.com/Lenny4/DeepFaceLabClient/assets/25544892/08686b40-157e-4278-929a-11cb06666d47 [4] #77 (comment) [5] https://github.com/notifications/unsubscribe-auth/AESPXXQES2VK5ENPBZBGZY3YWR74BAVCNFSM6AAAAABEC64ZW6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZWGY4TONRZGM

I have 2 options [CPU] : CPU and [0] : NVIDIA GeForce RTX 3060. I'm able to use one or both if separated by a comma

[Lenny4]

Alright, I will try to investigate tonight.

But as I don't have the error on my computer it will be hard to fix.

[Lenny4]

@DrWH Do you know how to code in Dart or python ?

[DrWH]

On 2024-03-04 14:32, Alexandre Beaujour wrote: @DrWH [1] Do you know how to code in Dart or python ? -- Reply to this email directly, view it on GitHub [2], or unsubscribe [3]. You are receiving this because you were mentioned.Message ID: ***@***.***> Links: ------ [1] https://github.com/DrWH [2] #77 (comment) [3] https://github.com/notifications/unsubscribe-auth/AESPXXQPYQJNDTJRTXI64KTYWSAZDAVCNFSM6AAAAABEC64ZW6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZWG4YTIMZWHE

No Dart or python. I wrote a botnet with a friend, that was python, but that was years ago. I can't remember I did much python so no experience really

[Lenny4]

Hi @DrWH ,

Can you try the version attached to this message and tell me if the problem is solved.

test-fix-gpu-1.zip

[Lenny4]

@DrWH up

[DrWH]

Thanks for trying. No it did not fix. I can use DeepFaceLab the old school way using bat files but your Client has stopped working for me. I have also recently installed another gpu. I now have an RTX 3090 and RTX 3060. Its a shame that your client does not find either card.

…

On 2024-03-06 01:27, Alexandre Beaujour wrote: Hi @DrWH [1] , Can you try the version attached to this message and tell me if the problem is solved. test-fix-gpu-1.zip [2] -- Reply to this email directly, view it on GitHub [3], or unsubscribe [4]. You are receiving this because you were mentioned.Message ID: ***@***.***> Links: ------ [1] https://github.com/DrWH [2] https://github.com/Lenny4/DeepFaceLabClient/files/14503329/test-fix-gpu-1.zip [3] #77 (comment) [4] https://github.com/notifications/unsubscribe-auth/AESPXXVRF4656UXWPKO5ZZ3YWZWI5AVCNFSM6AAAAABEC64ZW6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZZHEYTKMZXG4

[Lenny4]

Its a shame that your client does not find either card.

Totally agree, but as I can't reproduce the bug on my computer I can't fix it. Maybe one day someone with that problem will try to fix it and make a Pull request.