I have an AMD Radeon R9 290 running Moo! Wrapper. It's making the Windows 10 interface very sluggish. I don't want to stop Moo! running when I'm using the computer, just slow it down a bit. Surely there's a way to change the percentage of GPU it's allowed to use?

Please ignore this message, it was in fact my other graphics card (a Nvidia running Collatz) that's slowing things down. No idea why it should slow the display on the AMD monitors, but it does.

It's just the way mixed GPUs and their drivers interact - or the Collaz application was soaking a lot of CPU cycles as well (like FAH does).

I've now got Moo on the AMD running MUCH slower, and using a full CPU core instead of virtually none. All I did was:

Unplug all the monitors (4, 3 on the AMD and 1 on the Nvidea) from the computer, and move the box to another desk, where I opened it and replaced a dodgy CPU cooler.

Plugged it all back in (2 of the monitors may have been plugged in different ports but that's all), and suddenly Moo runs MUCH slower (2 hours instead of 15 minutes per task) and hogs a CPU core.

I upgraded to the latest AMD driver and the latest AMD OpenCL, but it didn't help. MSI afterburner tells me GPU usage is 0%. It used to be 100%.

Working fine on Collatz. Something is up with the Moo GPU tasks.

Collatz can also make video lag unbearable with a config file.

GPU's don't have a way to throttle to something like 50%. It's meant to display images in real time as it gets data. I've seen the lag as well on my iGPU while crunching on a dGPU. It's frustrating. You could set BOINC to not use your GPU while the computer is active.

I've now got Moo on the AMD running MUCH slower, and using a full CPU core instead of virtually none. All I did was:

Unplug all the monitors (4, 3 on the AMD and 1 on the Nvidea) from the computer, and move the box to another desk, where I opened it and replaced a dodgy CPU cooler.

Plugged it all back in (2 of the monitors may have been plugged in different ports but that's all), and suddenly Moo runs MUCH slower (2 hours instead of 15 minutes per task) and hogs a CPU core.

I upgraded to the latest AMD driver and the latest AMD OpenCL, but it didn't help. MSI afterburner tells me GPU usage is 0%. It used to be 100%.

Reboot the pc, Windows has no way to restart the gpu after it does something like that without a full pc restart.

I've now got Moo on the AMD running MUCH slower, and using a full CPU core instead of virtually none. All I did was:

Unplug all the monitors (4, 3 on the AMD and 1 on the Nvidea) from the computer, and move the box to another desk, where I opened it and replaced a dodgy CPU cooler.

Plugged it all back in (2 of the monitors may have been plugged in different ports but that's all), and suddenly Moo runs MUCH slower (2 hours instead of 15 minutes per task) and hogs a CPU core.

I upgraded to the latest AMD driver and the latest AMD OpenCL, but it didn't help. MSI afterburner tells me GPU usage is 0%. It used to be 100%.

Reboot the pc, Windows has no way to restart the gpu after it does something like that without a full pc restart.

The GPU is stopped and started every time BOINC tells it to if I'm interacting with the PC.

I've now got Moo on the AMD running MUCH slower, and using a full CPU core instead of virtually none. All I did was:

Unplug all the monitors (4, 3 on the AMD and 1 on the Nvidea) from the computer, and move the box to another desk, where I opened it and replaced a dodgy CPU cooler.

Plugged it all back in (2 of the monitors may have been plugged in different ports but that's all), and suddenly Moo runs MUCH slower (2 hours instead of 15 minutes per task) and hogs a CPU core.

I upgraded to the latest AMD driver and the latest AMD OpenCL, but it didn't help. MSI afterburner tells me GPU usage is 0%. It used to be 100%.

Reboot the pc, Windows has no way to restart the gpu after it does something like that without a full pc restart.

The GPU is stopped and started every time BOINC tells it to if I'm interacting with the PC.

That's not a reset though, that's a 'take a break, okay resume' thing.

Well if the driver install had asked for a reboot I would have done so. If it didn't, and BOINC didn't realise there was a problem, that's the programmer's fault not mine. And I only installed the driver AFTER there was a problem, so that cannot be blamed anyway. And if you were referring to rebooting after plugging monitors in differently, obviously I did, as I moved the PC physically and obviously had it switched off to install the CPU cooler.

GPU Driver install or upgrade on windows ALWAYS requires a reboot - often 2 or 3 of them - before the new driver starts working correctly if at all.
That's been the case back AT LEAST to Windows For Workgroups 3.10 (BEFORE 95).