Second office graphics card for monitor control alongside RTX2060 for machine learning

Hello everybody,

I am using an RTX2060 Super for trainings of a Tensorflow model for object recognition. My monitor environment has gradually expanded. This means that there are currently two full HD and one widescreen monitor (2560×1080) attached to the graphics card. Therefore, the three monitors occupy graphics memory that I lack for object recognition.

My consideration:

Installing a second graphics card with as little TDP as possible, which only takes care of the output on the monitors. I’ve already thought about a T400, but I’m not sure whether there will be driver problems, as the T400 and the RTX 2060 officially have different drivers. I haven’t found anything about it on the net, not even whether the T400 runs with a GeForce GRD.

What tips do you have for me? Or should I rather combine it with an AMD office card?

Side question: Very rarely do I ever gamble on the computer. Would it work to connect the 2060 to another port on the “big” monitor and switch the input if necessary? Does the monitor then also occupy graphics memory when the input is not active? That would be easier than having to switch cables when necessary.

I am very grateful for any tips and suggestions!