Discussion in 'Videocards - AMD Radeon Drivers Section' started by Hilbert Hagedoorn, Dec 8, 2020.
60 fps stutters and buffers
It did work before?
Massive frame skips here also - even if I let it buffer. Bandwidth issue with the video connection maybe? DisplayPort 1.2 here...or is VSR supposed to make things work?
as in you tried it and they no longer miss graphics or crash? or????
Since yesterday was the monthly update routine this should be part of the OS now (.685) though I wasn't aware of any D3D9 specific fixes, best I can tell Microsoft has made a number of updates for DXGI and D3D11/D3D12 throughout a few of these preview updates and then once more for the move to the final build.
Changelog for this latest cumulative isn't very in-depth though one would see it as some sort of Microsoft Edge update and a few security enhancements whereas the actual notes from the preview blog post has a lot more about specific fixes in it.
EDIT: Actually the preview update (.662) did apply something for the D3D9 components after a quick check.
Something for the game bar.
And more, especially compared to this.
Bit of a difference in detail here around what's actually included.
(Cumulative update after all.)
Well unless you defer updates, system is quite aggressive without some tweaks on making sure you have the latest or what the OS perceives as the latest otherwise ha ha.
(No Windows 10 you can't handle the drivers leave those alone heh.)
Sigh yeah it could be a bit better but at least it's fast when it comes to updating at the lack of user control and customization plus details on what these updates it sorta forces actually contain and eventual known issues of which there's a few.
(At least for the moment you don't need Enterprise tier access or whatever to call it to get at the advanced customization though reigning in this behavior isn't the most straightforward. )
well installed this one and back to shut down problem again so back to 20.11.1 and working fine
So, did you checked logs for a service/whatever which is blocking shutdown as been suggested to you in another thread?
It's usually under Event ID 203 in "Applications and Services Logs\Microsoft\Windows\Diagnostics-Performance\Operational"
A least my screens got to sleep after 9 month. When my HDMI and my DP was together I've never have a sleep of them only when I had the HDMI only and now it works... but I'm searching a 3080 now.
perfect so far little lower on benchmarks but not by a great deal
i did find event 200 but and right times for a normal shut down but no events for failed shut down at times it happens as i have recoeded times to chk
Weird thing.. My TimeSpy is not utilizing my 6800XT after updating the driver and installing 5900X. I refuse to believe it's the 5900X vs 5800X. All other benchmarks are in order. Just TimeSpy is bugging out on me.
Superposition is fine (not the 5800X->5900X and 1080p Extreme preset for even more CPU bottleneck than 1440p TimeSpy)
Even Port Royal within the same 3DMark suite is fine...
Port Royal fine also...
I think there might be a bug in the timespy benchmark related to the amount of CPU cores (everything over 8+ causes a score degradation)...
I am getting REALLY weird results on my 3960x --- basically around 12900-13000 total score with a 6800xt (+OC the 6800xt a bit).
Or it might be also a bug in the driver... but... only in timespy?? weird....
Unfortunately I am not able to test this ATM, but it would be interesting to find out --- if disabling half of the cores (maybe one whole CCD?) in RyZEN Master would yield a positive effect on the timespy total score.
I also have DP i HMDI monitors connected, no sleep issues ever on 5600xt or 5700xt.
just in case.. i am reading elsewhere that this driver on old chips like rx580 caused 3 cards to DIE
weird but.. you know..
Hopefully those reports you saw is just spreading FUD... else you know.... bad coding ist the actual cause of earthquakes...
I honestly cannot believe such information as it would require.... dunno.... ignoring all the clock limits in the cards bios, turning off the fans completely and letting the cards fry.... just as one of the the NV drivers did a long time ago....
card failure is most likely to occur across driver updates and power cycles.
Hi all ,
Yesterday i got my hands on Cyberpunk2077 but its so intensive that my vega56 struggles with FPS .
min 20...max 38fps all shadow options on Medium and everything else on high. Running on a 1440p 144hz Monitor. Excellent and intense game ,remembers playing me Hl2 from 2004.
Maybe i get a 6800xt in february 2021 when msrp is fine. cheers
GPU is not the only thing that matters...whats the rest of the system.
True... this being rather a general rule for all the electronic devices... no matter if it is AMD or NV, Hitachi/Western Digital or Seagate or any other brand...
Might have also been a "local" issue e.g. cheap/bad PSU that doesn't filter out the spikes that good etc....
IMHO not related to some "faulty drivers" tinfoil hat theories, except the drivers would really turn off the cooling completely...