grow a plant, hug your dog, lift heavy, eat healthy, be a nerd, play a game and help each other out
Huh interesting. Are these displays all the exact same model?
There’s a relatively new feature in the amd gfx display abstraction layer called freesync power optimisation. This feature leverages panel VRR to help mclk idle low at the desktop. This was introduced for single display use with RDNA 2, and expanded to dual display along with RDNA3s MALL advancements. I’m not sure if this is expanded further with RDNA 4 but I can try to find out.
That’s kind of curious. I don’t think 3D_FULLSCREEN should inherently determine idle mclk behaviour in and of itself. Is this just a single 1440p display at 120Hz? Is VRR enabled?
huh, I generally expect Vega10 based GPUs to idle at ~3W TGP (not inclusive of other board power losses like VRM). Can you tell us what you display setup is? Can you get a reading of your idle mclk using something like CoreCtrl?
video playback will likely kick the asic out of idle. What’s your power use at true desktop idle?
Generally speaking, vega 10 (56, 64) and 20 (radeon vii) are able to achieve decently low desktop idle with varied display configs due to the memory technology they employ.
Can you tell us which distro and display config this is with?
I’m also in 3D_FULLSCREEN on NV21XT but my idle mclk is 96 MHz. TGP (not to be confused with TBP which is only communicated on RDNA 3 and up) is 6 watts at idle with my browsers open. GNOME 47 + Wayland, 2560 x 1440 @ 180Hz + 1920 x 1080 @ 60Hz, VRR enabled on both displays. This is with Fedora 41, kernel 6.13.6-200.fc41.x86_64
For context, the GPU index (0,1,2 etc) will depend on the number of video adapters you have in the system. If you have a CPU with integrated graphics, there’s a chance this will be registered as 0000 whereas the dGPU will be 0001.
tech can be tasty too :)
in the UK we call them microcrisps
Signed. Thanks for sharing
is there an article related to this?
The key theme I see associated with UE5 and bundled technogies like nanite is acutely poor in-game performance
I see. You can temporarily edit your grub before the OS loads. This should afford you the opportunity to boot into the system without EDID modifications, though im not sure if your modified EDID will still load under this scenario. If so, you may need to switch into a CLI session to undo your changes.
I’m sorry to hear that. Does this system only have access to this single display? Did you use a kernel command to modify your EDID? If so, are you able to temporarily modify your grub before booting into the OS?
Archwiki references a [@<refresh>] (presumably denoted as [@144] for something like 144Hz) property, hopefully that’s all you should need to define, though I’m not sure if you’ll need to manually recalculate vertical and horz timings or something.
Maybe this can help fill in any gaps
You want to look into modifying your display EDID.
I don’t believe there’s a GUI for this on Linux but this post referencing the Archwiki might come in handy
https://foosel.net/til/how-to-override-the-edid-data-of-a-monitor-under-linux/
it do be that way :(
Their latest microcode update from 7 days ago was declared as final. Time will tell about the efficacy of the most recent fixes.
Unfortunately it may only be available via board partners as beta BIOS updates at present.
I’ve been lucky enough to dumb guy my fedora install since 28, and it’s been pretty decent to me. Granted I’m not using nvidia graphics, and I feel like that could throw a big spanner in the works for regular users. It’s a big enough leap getting into the mindset of installing software from Distro repos rather than directly from the vendor.
I hope the newer nv open kernel modules don’t stay out of tree. Also hope that NVK will give users the ability to just plug and play with mesa drivers in the future.
I’m trying to think of a way to mute the gradient at the bottom edge of a meta box (not sure what it’s referred to in compose) when the content is three lines long or less
This idiotsincars content can’t be two finger scrolled though the gradient sort of implies that it’s longer than three lines
I don’t believe FPO was expanded to >2 displays, though having identical panels would eliminate VBI complexities. Will be interested in understanding how this behaves with rdna 4.