Precision 7560 & 7760 Owners' Thread

Do displays connected to a dock that is connected to the USB-C ports count as connected "via the USB-C ports"?

 

I

For me it is easier to uninstall the reinstall the driver to get the system to automatically use the Intel Card oppose to try and set each app manually. Right now if using with no external monitors, I'd rather apps default to the Intel and then manually set an app to use the Nvidia Card using the Windows Graphics Settings or the Nvidia Control Panel.

I think what you are doing is a work around since there may be a driver hiccup issue.

 

Yes.

 

Thank you.

Can I ask, what is the practical meaning of a particular display running off either the intel or the nvidia graphics?

What I mean is that if for example my internal display is running off the intel graphics, but displaying a program that uses the nvidia card, what does that mean?

 

In this case, the NVIDIA GPU does the 3D rendering and the flushes the output over to the Intel GPU framebuffer for display on the screen. The NVIDIA GPU isn't doing the output directly, it is basically just providing background rendering services. (If there are no apps using it and it has nothing to do, it should be able to power off, which saves you power and battery life — the whole point of this system.)

 

Interesting, thanks a lot.

I have a USB-C dock with 3 monitors attached running off the Nvidia GPU, and the internal display running off the intGPU. (Hybrid on, and Discrete Graphics... on)

Testing with a particular program, I have set it with the Nvidia control panel to use the integrated GPU.
If I use the program on the internal display, the Nvidia GPU is not used at all.
If I use it on an external display, the integrated GPU does the bulk of the work, but I can see the Nvidia GPU is doing a bit. Presume the intGPU is doing the rendering, and just passing the output via the Nvidia GPU to display.

Is my understanding correct?

 

Yes, this rendering passthrough can go both ways.

Apps that do GPU accelerated work will "attach" to a particular GPU. The options in the NVIDIA control panel or in Windows Settings "advanced graphics" area can be used to "guide" it to your GPU of preference but depending on how the app works, that "guidance" may not apply and it could attach to whichever GPU was driving the display that it first opened up on. That can "stick" after it is moved to a different display — some apps may be sophisticated enough to switch GPUs but I'd say most are not and Windows does not do any enforcement of this. So, that could lead to what you are seeing, an app "using" the Intel GPU but running on an NVIDIA GPU-powered display, or the reverse, an app "using" the NVIDIA GPU but running on an Intel GPU-powered display (even after the NVIDIA GPU-powered display has been disconnected).

 

I was looking at the keyboard for unused keys and some options occurred to me.

CE and +/- are obvious candidates but turns out they can't be remapped because these are really just "Delete" and "F9" keys, and they can't be mapped separately.
Numpad "*" and "/" seem like candidates for remapping if you don't use them for the calculator functions very often. (I'd say that I use them often enough that it'd be annoying to have them missing.)
...Then, I spotted the right "alt" and "ctrl" keys. I never use these. They can be mapped separately from the ones on the left. I think that I'm going to try mapping them to "Home" and "End" and see how I like that.

 

Thanks. Really appreciate the explanation!

So this is an example of the guidance not working?

 

So, the NVIDIA control panel can set what the app should use as the "default" GPU. Apps aren't required to use the "default" GPU, they can pick a specific GPU, and there's nothing that you can do to override that other than outright disabling the NVIDIA GPU through Device Manager or something.