M17x R3 6990M Series Massive FPS drops

Hi there.

So I've had this Alienware m17x R3 since June of last year, and during January I started getting these random FPS drops. My FPS will be great (60-80) until I get a few seconds where it drops down to 5-6. This occurs in all my games.

At first I thought it was a heating problem, but I noticed the spikes happen even at 60-70c. I downloaded MSI Afterburner and played a game until I got the spike, and then looked at the monitoring section. During the spikes, my Memory clock (MHz) will plummet from its spot around 680 to 100, and my Core clock (MHz) will plummet from its spot at 900 to 150. From what I can tell, I think the GPU usage might be effected, as it seems to drop to 0 for a bit after those two downward spikes.

I've done literally everything I can think of. I've updated my AMD drivers, set my graphics card Power Play to "High Preformance," lower the FPS to all games to the point where computers much older than this can play it, tried overclocking, made sure Stealth Mode wasn't on, opened up the back to check nothing was out of place, cleared a lot of space on my hard-drive and used CC-Sweeper.

My next step is going to be calling Dell and, subsequently, doing an Alien Respawn. If that doesn't work, then I am going to claim it's hardware malfunctioning.

However, if anyone has any ideas why my clocks are falling from the sky (it happens very briefly), then I'd love to hear it. I'm hoping it's something I can fix on my end.

 

I had that issue when I first installed my 6990m as an upgrade to my 460m. It was throttling for me because the GPU die wasn't making good contact with the GPU die. I'd download hwinfo 64 and look at the sensors. It seems at least with my machine that there is a difference in temps between what the fan EC is reading and what the cards actual tempature is. You could also use HWinfo to turn the fans up to max and see if that helps as well.

 

The EC temperatures and the actual GPU temperatures are only a few percentages apart, with the ECs being around 59c and the actual temp. at around 61-62c.

It doesn't seem like a temperature problem to me. Something is routinely gutting my clocks and forcing them to their absolute minimums for a few seconds. I just can't tell what.

 

hmm sounds like a program could doing it. I'd advise trying doing a respawn and seing if that works. If it doesn't work, have dell come out and replace the video card.

 

You could google around the internet to figure out how to disable the radeon power management stuff. I don't even know what it is called, but I know it can be disabled, that might fix your issue.

 

Hmm. I couldn't really find much information about it.

Just curious, and forgive me if this is a stupid question, but I am still using the A04 BIOS. Should I try upgrading to newer versions? (I think the latest is A09).

Would that possibly help?

As always, appreciated.

 

You can try it, I would upgrade to A08, A09 has some issues.

I probably would not do a restore, since this doesn't sound like a problem that a restore would fix.

Maybe someone who has a 6990m will see this thread and give you advice about the power management stuff. I think for radeon it is actually a driver option, Nvidia hides it in the registry.

Of course, if you can't get it to work on your own calling Dell is often the easiest option. It could be anything from your GPU to motherboard to power supply.

 

^^ Powerplay IS Power Management and OP says it is already set to high performance.

Try downloading: https://www.sapphireselectclub.com/ssc/TriXX/TriXX.aspx

And set the clocks to say, 745/1010. You could try using these drivers too: http://benchmark3d.com/amd-catalyst-8-96-28-february

680/900 sounds like stock clocks for a 6970M; the 6990M runs at 715/900. You must have a 6970M?

Maybe you have some rogue process in the background?

http://www.iobit.com/gamebooster.html Might help..

 

Perhaps it is the 6970M. My dxdiag claims it's "AMD 6970/6990M" (real helpful lol). I'll do some digging to make sure.

I have tried overclocking with sapphire trixx before. I still got the drops. When it's not dropping, the games run fine. It's the 5-6 seconds that my clocks get forced to minimums that gut my gameplay.

I also have gamebooster and have switched off some additional, unneeded processes that doesn't show up in its library. I am also not running any hog-tastic anti-virus programs.

 

Mine would do something similarly odd. Have a 6990m... screen would partially minimize as if I was hitting alt+tab and then pop back up.

I did full uninstall of CCC and the drivers. Then i reinstalled the dell drivers first and the most recent AMD drivers. No more screen glitching but i sometimes have problems with switchable graphics

I am on A08 BIOS btw

 

My temperature still no more than 70, even after play Battlefield 3 for around 5 hours...... idle temperature is around 55-60.....

Since I changed my thermal paste for replace the dell stock, never more than that value....

R3, 6990m, A08.

 

I repasted too and can still hit 80 after a long session...

 

I updated to AO8 and I noticed a pretty big decrease in my FPS issue: I only got one clock-drop during my TF2 session last night. However, I'm in the middle of a few mid-terms so I need to extensively test it. I am probably going to go forward with a factory reset once my Spring break hits so I can rule anything software-related out.

After that, it'll be time to call up Dell.

 

Depending on how much stuff you have installed you might want to do a Windows system backup and make a system image. That way if the restore doesn't fix it, and I will be surprised if it does, you don't have to reinstall everything.

 

Try a power drain as well... remove power cord, battery and hold power button for about 3 minutes. You will be surprised of the changes you could get in your system. I would also recommend removing the coin battery (though you have to reset time / settings in bios).

 

So I have had another night of gaming without any clock drops. Upgrading the BIOS has definitely helped. My laptop also seems to be managing heat a lot better with the upgrade.

The power drain idea is interesting. I'll give it a go if it returns.

Thanks guys.

Bleh. The problem returned tonight. I just can't figure out what it is =X. It's so random. Some days it doesn't come and some days it does.

 

This same issue just happened to me yesterday. My computer is 3 weeks ok, got the M17x r3 6990M.
Here is what I believe triggered my problem though...
I was playing battlefield 3 yesterday and I must have moved so the AC cord came out of the back. I got the FPS hit (as expected) then plugged it back in. FPS still stayed around 15-20. So i closed down everything and restarted. Upon boot up I for a black screen with white text saying something like "Cannot identify power supply or power supply input. Press F1 to continue, F2 for setup etc..."

I pressed F1 to continue and my computer recognizes its plugged in, and I believe I got one game of normal 80+ fps in BF3, then on load up of next map it dropped to 15-20 fps and has been this way ever since (tried Skyrim and Company of Heroes and they all have FPS too).

So a small part of my wants to believe I just get a new or beefier power supply and everything will be fixed, and thats probably wishful thinking. What do you guys think?

(same issue with BIOS AO9 and 8, yes drivers are up to date, and I have played around with the power options. And im not trying to hijack your thread donut, just adding my experiences)

 

hmmmm...sounds like a bad power brick or a bad capacitor on the motherboard. Keep an eye on it and call dell if the issue persists

 

Thanks Slickdude,
I called Dell today and explained the situation to them. They are sending a tech to my house with a new ac adapter, new vid card and new motherboard if thats what needed. Surprising from dell. I'll post my results once its all done. Hope those of you who have this issue find solutions. This really is a pain in the type of issue, not being able to game on a gaming computer. Thank god for minecraft, can still play that game without unbearable fps.

 

So, I called up Alienware support. I did a diagnostics and everything checked out. Then they transferred me over to a "gaming optimization" specialist (yay, not covered by warranty). He basically ran me through tuning my drivers/BIOS to their best performing capacity, cleaning up all my registry errors and optimizing my AMD card. I got a good boost of FPS in benchmarks.

However, the problem didn't go away as I found out tonight. I played TF2 and Skyrim, and I still get these clock-drops. According to MSI Afterburner, they will run at their max capacity (without overclocking) of 680/900 and then spike really quickly to 100/150. That quick spike causes my game to go to crud for 5-6 seconds.

I am running out of ideas =/ I've emailed the tech support agent that assisted me, but I can't seem to get this problem fixed.

 

Forbidden, you have to call them, be stern, and get someone over to your house to replace the card. Do not take no for an answer and get a tech over immediately. At this point, it looks like a hardware issue. Another vid card will probably solve all your issues

 

Alrighty, so here's an update:

I called Alienware support today. This time I got directed to a hardware tech. that ran a few stress tests. They noted that my CPU temperatures were very high and that it was probably the cause of my throttling. They're sending out a dispatch tech. Monday to replace the thermal modulator and the CPU fan.

Thanks for all the help, guys. Especially SlickDude80. Definitely good to have you on these boards.

 

Sounds like that pooey BDPROCHOT thingy. You can disable that in the unlocked bios. Tbh a repaste with PK-1 should lower temps more than anything.

 

Update:

Hey guys, good news i got my problem solved...
First off I called dell and they said theyd send a tech out the next day hich they did. The tech tho sounded like a pothead and ended up postponing twice so I went and took action on my own. I downloaded MSI Afterburner to monitor my gpu. (Im no pro, just casual gamer with minimal tech knowledge), and I see that temp is fine (up to 78 max in game), and everything else looked normal EXCEPT for Memory Clock speed which stayed at 150Mhz out of game (as expected) AND stayed at 150Mhz in game (unexpected, as it was set to 900Mhz).

I had skyrim open, was running at 10-15 fps with this 150Mhz, then when i turned it down to 899mhz, i maximized skyrim and its running a beautiful 75-85 fps, no issues. I almost cried with happiness. I turned it back to 900 Mhz and it went right back down to 10-15fps. Turned it up to 901Mhz and its fine at 75-85fps.... ?

I'm quite happy as I can game again. But am curious as to what is going on. The gpu memory clock speed when set at 900mhz wont go above an idle speed of 150mhz in game. Weird.

Hope this helps someone

 

What's the stock memory clock speed?

 

when I opened afterburner it was on 900Mhz, which I assume is stock as I hadnt adjusted it at all.

 

I have exacly the same problem and i JUST bought my laptop. when gaming (even when playing low-graphics games like LoL), once in a while i get extremely low fps and sound issues, like the machine is running in slow motion for 3-4 seconds.
Any clues whats that?