Change.org petition to add an option to disable power saving in GPU settings

[LoadsaMoney]

Afterburner didn't work for me, was still going up and down, up and down.

 

[sammarbella]

Afterburner didn't work for me, was still going up and down, up and down.

Did you open OverDrive in Crimson settings?

If you did, even if you didn't change any settings inside THAT'S the problem.

The only solution is to uninstall drivers and reinstall them and for the God sake...DON'T TOUCH OVERDRIVE ANYMORE!!!



Then AfterBurner will have success locking GPU clocks and timings.

 

[Orangey]

Seems like this thread has run its course and is only riling people up now.

 

[LambChop]

Nothing in common with Nvidia last 0.5 GB VRAM speed affair.Nvidia problem is mainly a hardware thing.

Basically is like you said, since Crimson drivers (November 2015) AMD introduced a more aggresive AND MANDATORY power saving behavior who downclocks GPU at any moment to save power at performance cost.

The only solution right is force GPU clocks and timings via 3rd party tools like ClockBlocker (!) and or AfterBurner.

Did AMD tell people they were doing that ? Cause thats not done by accident.

 

[sammarbella]

Did AMD tell people they were doing that ? Cause thats not done by accident.

Obviously it was not done by accident.

There was no public statement about this new aggresively enforced power saving behavior.

There is a clear shift in AMD GPU top priority from performance to power saving.

I guess this new "green" policy is set to fight the AMD GPUs image as power hungry and high heat hardware generators.

All the marketing around the next AMD GPU arch (ads and PR shows) expose it as a power saving hardware marvel and no word is still said about performance.

It's fine but ppl normally buy dedicated GPUs for their performance not to power saving.

An option to choice between performance and power saving should be the best solution for customers and AMD image as "green" brand.

 

[shankly1985]

It's a bug effecting some user configuration nothing more.. Not all users are effected, so therefore it wasn't amd intension for this.

 

[sammarbella]

It's a bug effecting some user configuration nothing more.. Not all users are effected, so therefore it wasn't amd intension for this.

It's not a bug, it works perfectly for his intended purpose: FORCE power saving at any cost (performance).

Ofc it doesn't affect all user setups, only the users who play games and playback multimedia content are affected.



Users who only use desktop 2D softwares like office or software who only requieres high CPU usage are not affected, the usual lower 2D power state of GPU clocks and voltage is fine for them.

I have nothing against the power saving per se i only ask an option to avoid it when performance must be the furst priority if customer choose it.

I don't like to be FORCED to poer save at all cost without drivers option.

 

[shankly1985]

Out of the Friends I know running GPUs 7950, fury, furyx 290x and 290 non of us are effected by this.. I have asked them to show me there hwinfo64 results from playing a game all have rock solid core and memory clock.

It's a bug...

 

[ltron]

It's not a bug, it works perfectly for his intended purpose: FORCE power saving at any cost (performance).

Ofc it doesn't affect all user setups, only the users who play games and playback multimedia content are affected.



Users who only use desktop 2D softwares like office or software who only requieres high CPU usage are not affected, the usual lower 2D power state of GPU clocks and voltage is fine for them.

I have nothing against the power saving per se i only ask an option to avoid it when performance must be the furst priority if customer choose it.

I don't like to be FORCED to poer save at all cost without drivers option.

You are correct, sammarbella, that AMD changed or tweaked their power management algorithm when Crimson was introduced, but I do not think they intended performance to be affected when compared to the old power management algorithm.

They simply looked at their algorithm and found several areas that they felt could be tweaked in order to save more power without harming performance. See them boasting about that here:https://www.youtube.com/watch?v=AXLXPiJpVGs. However, they must have introduced unintended bugs into the algorithm when they made these changes or were mistaken that they could save power in these areas in the first place.

The fact that they have finally acknowledged this problem as a bug in the latest release notes suggests that it is not working as intended. Also, it seems to only affect a significant minority of games which is why some people may not be affected.

I fully support your petition though and believe that they should at least have given the option to disable power management on a per-game basis since they could not fix the issue in a timely manner and it would enable customers to work around similar bugs if they occur in the future.

You are also quite right to continue the petition as it forces AMD to take notice and shows that customers are unhappy and that more should have been done, much more quickly, to mitigate this problem. As you say, words are just words and we'll only stop badgering AMD once the fix is in our hands and it doesn't break anything else; this is why I signed.

 

[LtMatt]

Out of the Friends I know running GPUs 7950, fury, furyx 290x and 290 non of us are effected by this.. I have asked them to show me there hwinfo64 results from playing a game all have rock solid core and memory clock.

It's a bug...

Yes it is, that's why it's in the known issues of our latest Crimson 16.2 release notes. We're going to fix it.

 

[Dygaza]

*****WARNING, HUGE IMAGE*****

Spoiler

Spoiler

They did announce it. And it's only supposed to be active when gpu power isn't needed (cpu bound or fps capped). I personally haven't ever had problems with stutterting with it. Only problem is that sometimes clocks just don't raise at all, after they have dropped. There are actually games like world of warships where I get exactly 73fps (cpu bound) with clockblocker (1100 core), or normally without (core at about 600). Same game experience, just 3 degrees lower temperatures, and more silent game play.

I think crossfire is where the biggest problems are and that really needs to be taken care for.

I still like the idea of full power performance profile, and that's why I signed. Actually AMD could go one step further, and add a slider that would let user set how aggressive power saving method he wants.

 

[LtMatt]

I've posted this here before, but if anyone is experiencing clock fluctuation then you should be able to resolve the issue by using either Catalyst 15.7.1 WHQL or Catalyst 15.11.1 Beta as a temporary solution.

 

[sammarbella]

Yes it is, that's why it's in the known issues of our latest Crimson 16.2 release notes. We're going to fix it.

And also in the 16.2.1 Crimson drivers released TODAY:

Known Issues

Core clocks may not maintain sustained clock speeds resulting in choppy performance and or screen corruption

http://support.amd.com/en-us/kb-articles/Pages/AMD_Radeon_Software_Crimson_Edition_16.2.1.aspx

That's why i think aknowledgement of a problem means nothing if the next drivers "solution" is to include it again under "know issues" in changelog.

Still waiting for a "a future Radeon Software update".

Maybe next driver? Who knows?

Ofc the use a pre-Crimson driver "solves" the problem introduced in Crimson drivers.

 

[LoadsaMoney]

I've posted this here before, but if anyone is experiencing clock fluctuation then you should be able to resolve the issue by using either Catalyst 15.7.1 WHQL or Catalyst 15.11.1 Beta as a temporary solution.

Telling people to go back to an 8 month old driver, what a fail!.

And also in the 16.2.1 Crimson drivers released TODAY:



http://support.amd.com/en-us/kb-articles/Pages/AMD_Radeon_Software_Crimson_Edition_16.2.1.aspx

That's why i think aknowledgement of a problem means nothing if the next drivers "solution" is to include it again under "know issues" in changelog.

Still waiting for a "a future Radeon Software update".

Maybe next driver? Who knows?

Ofc the use a pre-Crimson driver "solves" the problem introduced in Crimson drivers.

A minor update to Crimson 16.2 has been released, Crimson 16.2.1.

This is identical to 16.2 with the addition of a Far Cry Primal Crossfire Profile.

What a fail!.

 

[Dygaza]

Now things are getting totally out of control. Now even releases of crossfire profiles are getting fail comments?

 

[sammarbella]

Now things are getting totally out of control. Now even releases of crossfire profiles are getting fail comments?

As crossfire user i always welcome working CFX profiles.

AFIK the only difference in this new minor release (16.2.1) is the addition of Far Cry Primal CFX profile.

I think it's good for AMD customers to have game support since launch day.

Speaking of CFX profiles, a completely bug-free Fallout 4 CFX profile is still a dream:

Known Issues

Cannot enable AMD Crossfire with some dual GPU AMD Radeon HD 59xx and HD 79xx series products
Fallout 4 - In game stutter may be experienced if the game is launched with AMD Crossfire enabled

The usual solution for CFX problems is still the same: disable it!

 

[LoadsaMoney]

Its just bloody stupid that you have to download the 16.2s again, and re-install them, again, just for 1 Crossfire profile.

They used to do little CAP files you could just download, couple of ks, that you could just install, now you have to go through the rigmarole of unstalling the drivers, then redownloading them again (300+mb), then, reinstalling them again, just for that couple of ks, what an absolute ******* joke.

 

[Dygaza]

Its just bloody stupid that you have to download the 16.2s again, and re-install them, again, just for 1 Crossfire profile.

They used to do little CAP files you could just download, couple of ks, that you could just install, now you have to go through the rigmarole of redownlaoding the same 300+mb drviers again your already running, just to get that bloody couple of ks.

That is true, those profile downloads were a lot better than whole new driver version. Dunno why AMD can't use some sort of patcher if there is only minor difference. Perhaps this should be asked them, and maybe they could even deliver it.

 

[LoadsaMoney]

That is true, those profile downloads were a lot better than whole new driver version. Dunno why AMD can't use some sort of patcher if there is only minor difference. Perhaps this should be asked them, and maybe they could even deliver it.

Yeah, they started the little CAP files, so you didn't have to wait a month for the next driver that would have it in (when they only did 1 driver a month), as everyone was complaining about it, was ideal that.

How is adding that little profile, into the exact same drivers you already have on, so you have to redownload them again, un-install them, then re-install them again, better ?

No wonder Nvidias smashing them to bloody bits.

Who the hells running them really, is it this guy ?

Did you open OverDrive in Crimson settings?

No, i didn't.

 

[sammarbella]

No, i didn't.

Don't try it, a performance black hole is located right there...



I'm fully aware that this Petition is not going to reach it's goal acting alone.

The only thing making AMD fix this an other things is the public exposition.

Any article on a tech/review site has more force than my rant in some forums or any customer complaint survey filled in AMD site.

Like this one:

XFX Radeon R9 Fury Triple Dissipation Video Card Review

Final Thoughts and Conclusions

After using the XFX Radeon R9 Fury video card for several weeks we ran into some performance issues that were tough to look beyond. We had Rise of the Tomb Raider crashing and then when playing Star Wars Battlefront we had really bad stuttering issues at 2560×1440 and 3840×1440 screen resolutions. Both of these game titles were not available when we published our Radeon R9 Fury launch article, so the gaming experience we had not was not the same as last summer.

AMD has acknowledged both issues that we ran into today with the release of AMD Crimson 16.2 drivers. If you look at the release notes they have listed what we run into earlier this month as known issues.

Core clocks may not maintain sustained clock speeds resulting in choppy performance and or screen corruption
Rise of the Tomb Raider – The game may randomly crash on launch if Tessellation is enabled

We’ve been talking to AMD about our gaming experience for a couple of weeks now, so it looks like the solution is going to take some time to be corrected. That core clock issue is one that is hard for use to overlook on a video card that costs over $500 to buy. If we spent $529.99 on the XFX Radeon R9 Fury video card and then got choppy performance and crashes on some of the more popular game titles we’d be very upset. XFX did a great job on this card with their custom cooler, but there are issues here on the AMD hardware/software side that needs to be worked on and that is obviously beyond XFX’s control. The crashing and stuttering issues likely apply to all brands that have a Radeon R9 Fury card.

Until AMD can provide a fix for the clock rate issues at higher resolutions we can’t suggest to our readers that the AMD Radeon R9 Fury video card is worth buying. The stuttering and crashes were pretty severe and we had a better gaming experience on the XFX Radeon R9 390 8GB video card that we reviewed last month.

So AMD if you don't fix it for current customers you should at least fix it to sell some cards in the future.