CUDA-Enabled GeForce 1650?

Hi all,

I would like to know if GeForce 1650 is CUDA-Enabled.

In this link https://developer.nvidia.com/cuda-gpus it doesn’t appear but could someone confirm me?

Thank you very much.

Every GPU produced by NVIDIA since about 2008 is CUDA enabled.

While trying to run gromacs on my laptop, the software states that my GTX 1650 is not enabled for computing. nvidia-smi shows the graphics card and the drivers are installed.

Install CUDA. When you install CUDA, select the option to keep your current driver version.(*)

CUDA installers can be gotten here: http://www.nvidia.com/getcuda

Follow the install guide instructions (linked from that page) for your OS carefully.

If GROMACS has specific CUDA version requirements, you will need to check with GROMACS community or rebuild your GROMACS to match your installed CUDA version.

(*) (Note for future readers: this doesn’t necessarily apply to you. Right at the moment, GTX 1650 is a very new GPU, and so any driver that works with GTX 1650 will work with any currently available CUDA toolkit version. This doesn’t apply to every GPU and every CUDA version, and may no longer be valid months or years into the future.)

If you cannot find the answer in the GROMACS documentation, I would suggest asking about GROMACS configuration issues on the official GROMACS mailing list:

http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List

Hi, What is the compute capability of the GTX 1650? I don’t see it here https://developer.nvidia.com/cuda-gpus#compute

All Turing family GPUs currently are compute capability 7.5

https://www.techpowerup.com/gpu-specs/geforce-gtx-1650.c3366

Hi, the compute capability of GTX 1650 is not mentioned. I am a AI researcher and I want to know if this is possible to use this gpu card for training the neural networks or not?

The compute capability is 7.5. It is possible to use it for training neural networks. I won’t be providing tutorials for that here. The topic is covered elsewhere on these forums and other NVIDIA websites, as well as elsewhere on the web.

The GTX 1650 is based on the Turing architecture (specifically, the TU116 GPU) with compute capability 7.5. The minimum requirement for various deep-learning frameworks at this time is typically compute capability 3.5. So yes, you should be able to run those frameworks.

You would want to carefully check the hardware requirements of the particular deep-learning frameworks and applications you intend to run.

Hi, the compute capability of GTX 1650Ti is not mentioned anywhere. Is this card can use for training of neural network or not?

Thank you.

I assume this is actually a GTX 1650Ti Mobile? If so, that is also based on the Turing architecture (specifically, a TU11x chip) with compute capability 7.5, just like the GTX 1650 discussed above.

Again, you would want to carefully check the hardware requirements of the particular deep-learning frameworks and applications you intend to run.