!hzgkCxLtCOTmFXGauj:nixos.org

NixOS Gaming

453 Members
Gaming things, my hands are typing words.129 Servers

Load older messages


SenderMessageTime
17 Jul 2025
@olivia:computer.surgeryoliviaI wouldn't expect it to, nobody likes it lol02:58:25
@olivia:computer.surgeryoliviaI believe the intention is to remove it eventually, but I'm not very aware of what's going on with amd stuff so take that with a grain of salt02:58:51
@olivia:computer.surgeryolivia https://docs.mesa3d.org/drivers/radv.html#shader-compilation yeah

We still maintain an LLVM based compiler backend too, which is these days solely used for testing and hardware bringup. Users are recommended NOT to use the LLVM backend.
02:59:54
@woobilicious:matrix.orgwoobiliciousOh really? how is that related to the ACO compiler? doesn't that use llvm?03:00:08
@olivia:computer.surgeryoliviano ACO is NIR only03:00:44
@woobilicious:matrix.orgwoobiliciousahh my memory is wrong then.03:01:31
@olivia:computer.surgeryolivia(I believe the llvm backend in radeonsi predates NIR)03:01:33
@woobilicious:matrix.orgwoobilicious olivia: I think the FSR4 code just converts to fp16 manually instead of them extending the compiler to do the conversion. I guess the code is writting to use cooperative_matrix, so yeah only RDNA3...too bad I wanted slow as dogshit FSR4 haha. 03:33:25
@olivia:computer.surgeryoliviayeah, that seems plausible (I don't know anything about the specifics of FSR4, but almost certainly it will not work without cmats)04:10:31
@olivia:computer.surgeryoliviaoh, the fp8->fp16 lowering is happening in vkd3d-proton04:28:10
@olivia:computer.surgeryoliviahttps://themaister.net/blog/2025/05/09/conquering-fidelityfx-fsr4-enabling-the-pretty-pixels-on-linux-through-maniacal-persistence/04:28:12
@olivia:computer.surgeryoliviahttps://github.com/HansKristian-Work/vkd3d-proton/pull/242704:28:22
@olivia:computer.surgeryoliviakinda cursed04:28:23
@k900:0upti.meK900
In reply to @woobilicious:matrix.org
K900: your post mentions Intel-RT on non-x86, does that mean you can run Arc graphics on Arm?
Always could
04:54:28
@k900:0upti.meK900They're just PCIe04:54:28
@k900:0upti.meK900 The RT code had some handrolled x86_64 SIMD but they got rid of that 04:54:41
@mon:tchncs.deribosomerocker joined the room.23:58:21
18 Jul 2025
@jonhermansen:matrix.orgjonhermansenHoly crap, I've really been missing out. Looking Glass performance is really good. Now if I can only figure out how to correctly isolate and pin my CPU cores...02:40:56
@jonhermansen:matrix.orgjonhermansenIs anyone doing similar things and wanna share their setup?02:42:39
@jonhermansen:matrix.orgjonhermansenI have serious problems with automatic core assignment in virt-manager with a game that is basically single core. Maybe I should simply try reducing the number (I have 16 P-Cores, so I gave 12 to the VM) Maybe I also need hugepages I thought KVM would be maybe 5% to 10% performance hit. The latency on the system becomes really bad and I don't know why.02:47:39
@jonhermansen:matrix.orgjonhermansen* I have serious problems with automatic core assignment in virt-manager with a game that is basically single core. Maybe I should simply try reducing the number (I have 16 P-Cores, so I gave 12 cores to the VM) Maybe I also need hugepages I thought KVM would be maybe 5% to 10% performance hit. The latency on the system becomes really bad and I don't know why.02:49:17
@jonhermansen:matrix.orgjonhermansen I have 144hz monitor and the latency transferring data from the GPU is very low, but I'm still trying to iron out the CPU issues in the guest.

Maybe I should switch to invoking qemu from a script or as a systemd service
03:08:11
@jonhermansen:matrix.orgjonhermansen* I have serious problems with automatic core assignment in virt-manager with a game that is basically single threaded. Maybe I should simply try reducing the number (I have 16 P-Cores, so I gave 12 cores to the VM) Maybe I also need hugepages I thought KVM would be maybe 5% to 10% performance hit. The latency on the system becomes really bad and I don't know why.03:10:08
@jonhermansen:matrix.orgjonhermansen And for some reason, I can not use host-passthrough cpu or else the guest will crash. Even with migratable="off" 03:10:55
@jonhermansen:matrix.orgjonhermansenguest is Windows 1003:11:42
@haauler:matrix.org@haauler:matrix.org joined the room.14:24:28
@gaelle:chat.blahaj.zoneGaƫlle joined the room.14:47:56
@colemickens:matrix.orgcolemickensI feel so lost trying to keep track of FSR3/3.5/4, the new lossless scaling stuff, what exists in Proton/Gamescope/drivers. Ugh.23:27:53
@colemickens:matrix.orgcolemickensI do know Halo has like zero support for FSR so I guess I don't care about that for now.23:28:06
@colemickens:matrix.orgcolemickensSo I guess I just care about whatever Proton has built-in or whatever I can get through Gamescope's upscaling if I choose to use it.23:28:25

Show newer messages


Back to Room ListRoom Version: 10