I submitted again also. God damn it AMD.

- just installed this one.. - commencing "testing".. 🤓

I wouldn't go that far. 🙂 Forza Horizon 3, Gears of War 4, Dishonored 2, Titanfall 2, Battlefield 1, Civilization 6, Mafia 3, Call of Duty: Infinite warfare and Obduction are all supported by profiles. It is missing a profile for SkyrimSE still though far as I can tell.

Just because there is a CF profile for a game doesn't mean it works in CF; in some cases the profile simply disables CF for the game.

Bleh, I could trigger the idle UVD bug on first try as always. I'll send another report. It's really driver related, as 16.7.2 doesn't trigger it by doing the same.

What exactly is this UVD bug? I've been seeing it mentioned regularly for the last few years but I have yet to see a clear explanation of exactly what it is. My HD7870 and HD7950 both always appeared to function properly. So far, my RX 470 appears to be functioning as expected.

It's circumstance where the card gets locked at UVD clocks, and only uses the lowest UVD in idle (without anything 3D accelerated running), and only its highest UVD clock for higher loads such as 3D (meaning no boost clock for models that have it). It can be triggered for some users while having something running that sets the card in UVD clocks such as a browser and then playing a video with hw accel. After closing everything, the card will be stuck in UVD mode. This probably means idle at 450/1200 on your 7870 and 501/1250 on your HD7950 instead of 300/150 which can mean >30W, and no applied OC via soft (if they are non-boost models, the highest UVD clock will match the stock 3D one).

I just used DDU to uninstall 16.10.2 and install 16.10.3. I still can't crossfire to work with my 2x 290x and Battlefield 1. Anyone else have luck?

I just used DDU to uninstall 16.10.2 and install 16.10.3. I still can't crossfire to work with my 2x 290x and Battlefield 1. Anyone else have luck?

Same here with my two 390s...CFX still not running good in BF1 still huge framedrops and slower overall performance than with just one of the cards. Maybe for that to fix it requiers a patch by DICE, since it doesn`t seem to be an issue with CFX at all since I am not reading a lot about it. SO for me these drivers run the same like th 16.10.2 did.... Cheers! :cheers:

DEL wrong post

Bleh, I could trigger the idle UVD bug on first try as always. I'll send another report. It's really driver related, as 16.7.2 doesn't trigger it by doing the same.

Tell me exactly how you trigger the bug and version of software that you use. And do you have more than one displays connected?

Ok, tested with BF1. I get very high GPU usage 95-99% on both GPUs(like should), but the crazy CPU usage (over 90% sometimes) ruins hole thing(even I have i7 4770k@4.5GHz). I have 120HZ monitor, but my old 7970's cant producy enough frames(high details, AA of, sweetFX 2.0), so I capped fps to 85. Capping the frames decreased CPU usage to 60-80% so GPU's are working now about 80-90% and now everything works fairly smoothly and game is playable.

The clock bug is gone on my r9 280, but not from this driver, ever since i got a update 1 week ago on Adobe Flash it was gone.:)

It's circumstance where the card gets locked at UVD clocks, and only uses the lowest UVD in idle (without anything 3D accelerated running), and only its highest UVD clock for higher loads such as 3D (meaning no boost clock for models that have it). It can be triggered for some users while having something running that sets the card in UVD clocks such as a browser and then playing a video with hw accel. After closing everything, the card will be stuck in UVD mode. This probably means idle at 450/1200 on your 7870 and 501/1250 on your HD7950 instead of 300/150 which can mean >30W, and no applied OC via soft (if they are non-boost models, the highest UVD clock will match the stock 3D one).

Yeah, not something I've experienced with any of my AMD cards. At least not that I'm aware of.

Tell me exactly how you trigger the bug and version of software that you use. And do you have more than one displays connected?

I've explained a few times in previous threads, this procedure will cause it on two PCs I maintain using only single monitor (7870 on dvi, 7970 on DP), and also on a friend's with an Asus 280x with the same software and a single monitor via dvi. This is not the "bug" that causes higher clocks with two displays connected. I'm on W10 x64 AU.351, and background stuff such as Game DVR has been disabled via registry, along anything with an overlay, background apps, etc. 1. Confirm that your card is idling at 300/150, let hwmonitor stay visible. Open Chrome with 3D accel enabled (54.0.2840.71 m (64-bit) ATM), and set to no background running for its apps after closing. Go to facebook, or any other page with lots of images that get loaded on vram, and scroll down a lot until you can check on hwmonitor that the card will switch to UVD clocks and stays there without any browser/scrolling activity (if you can't make this happen, the bug won't be triggered). 450/1200 on yours (exactly same as the 7870 in this PC). 2a. While Chrome is running and the card's clocks are stuck in UVD mode, launch a .264/DXVA video with MPC-HC (1.7.10 using default Custom EVR presenter). At first it should switch to the highest UVD clock, and then if the video isn't quite heavy, it'll return to 450/1200. (some people will trigger it later by playing a .264/DXVA youtube video instead at this point, but the problem here is that it'll depend on your browser, html5, flash, etc) 3. Close MPC-HC. Close Chrome. Clocks will still be stuck a 450/1200, without anything using the GPU. Voilà the bug. It happens on the affected systems on all drivers since and including 16.7.3. 16.7.2 was the last not to be affected. Alternative procedure: 2b. Put your PC on sleep mode while Chrome has fixed 450/1200 clocks. Wake it and close Chrome. Clocks will still be stuck in UVDs lowest clock. I've only tried this last method on 16.8.x, but it might still work. All of my last bug confirmations where using the first procedure.

Even games can get your GPU stuck in UVD clocks if they use hw acceleration. I think that this hits all of 1st GCN cards, because AMD updated encoder on newer cards, but I cant confirm that since I really dont know anyone with weaker 1st gen gcn cards. All 6 pitcairn/tahiti that I could check can reproduce this bug. Considering that I had a little chat with ray_m (product of rejecting my post on his reply on AMD forums) I dont think that this bug will be fixed. In fact, I think that AMD wants to hide it (!?) because most of the people do not use monitoring tools so they cant really detect it or even know how card should work.

hawaii here, not affected by uvd bug still i didnt like the idea of uvd clocks etc. so i "malformed" the uvd table in bios it works pretty well; as in no more uvd clocks while playing dxva2 accelerated video, just normal load dependent clock fluctuations http://imgur.com/a/U0ALv not sure if this could work for other cards

UVD bug (without anything taxing the GPU on single screen) only seems to apply on CGN1.0 Tahiti and Pitcairn for the affected users. Fortunately there seems to be no problem on Hawaii or later.

I have an HD7770, Windows 7, 64-bit. I'm still getting the fan issue. It won't switch off when I wake the computer up from sleep. I have to put it back to sleep and then wake it up again immediately for it to stop. This problem has been happening since 16.10.1.

Tahiti is ancient at this point, you guys are overreacting. Who cares if idle clocks are stuck, performance clocks works just fine and thats what matters. My card is boosting to 1165mhz without any problems and im gaming every day on this system. Still capable tho, Time Spy score, 280X is beating 1050ti : https://s12.postimg.org/g2exdtte5/Untitled.png

Maybe you don't care, but the rest of us do. We clearly aren't overreacting. You flashed a custom BIOS that sets the UVD state 2 to your 24/7 OC clocks, but the rest of us who're using a stock BIOS can't apply OC clocks once the bug shows up. Since it doesn't bother you, kudos to you. But don't tell others not to complain because they have a right to. And that's the purpose of this forum to begin with.

Thats right, Romulus. I did flash custom bios that sets UVD 2 state to my 3d clocks. Good to see you are paying attention. Maybe others should do it as well and avoid some issues. Thats why tools like Vbe7 is there for.