M15x & 6970M / 6990M - What you need to know - upgrading, problems, solutions

Blue? You meant cyan，right？That is from Eurocom if ichime is correct.

And what about the black one? What memory chips does it use?

 

@IrateMike -- Try GPU-Z and see whether it is getting readings on the three sensors reported there. Could be a bad sensor or could be a glitch in HWiNFO32/64.

 

Im getting alot of shutdowns now. My 6970m is undervolted to 1v and i cannot seem to handle the heat. Boy are those heatsinks badly designed. Should i undervolt it ever further?

 

What are your max temps?

 

Im always playing a game when it suddenly shutsdown/reboots so i dont know what temps im at when it occurs.. I now updated to the A09 bios but i see in bios i still have 1.16 ECC or whatever it was. Dont know what it is and what its good for but shouldnt it be 1.18.

Ill read the older posts where someone solved that problem. (i believe it was you widezu

 

Yeah your bios only half updated. The newest EC should solve the problem. You should try to flash your bios again.

 

You can also flash the EC manually, see here.

 

But how come that EC 1.18 will fix my problem? Now with 1.18 i always get the unsupported PSU warning on bootup. Had to disable it.

edit: I started Entropia Universe in windowed mode and played it for maybe 15mins and highest temp (MemIO) was about 73 degrees and it shutdown. I have had it at max 82degrees before and no shutdowns.

Think the GPU is giving up on me? The PSU also doesnt feel so hot, its been much hotter before. Maybe the PSU is giving up on me since i get psu warning on bootup.

I restored the GPU voltage to 1.1 in TriXXX to see if the card gets to little cream and shutdowns. Gonna play Entropia a while and bench it.

edit: temps now reached about 78-79 and played for 15mins until it shutdown with 1.1v .

 

Yes, temps are fine. Go back to the standard voltage and see whether it works.
Your PSU behaves strangely... you never had this warning before? Weird... maybe your PSU is failing.

 

Hi, do I have to upgrade power adapter to 240w from m17x to run with i7 920xm and 6970m?

 

One thing to note is that i am always playing without the battery in my computer. Since my computer is always stationary. Could this be the problem of the shutdowns?

edit: nope battery didnt help. Still shutdown

 

I used to have shutdowns when running Vantage. Updating to A09 and EC 1.18 fixed the issue, however if I overclock the card/CPU and run vantage I still get shutdowns. I presume it's normal cause it's too much heat and power. My GPU is also undervolt to 1.0 V.

I never had shutdowns while gaming yet after upgrading to A09/EC 1.18. Fingers crossed!

Do you have a 150W PSU and you get a warning at startup? that's weird, if your laptop is still under warranty you can maybe call dell and ask for a new PSU.

 

im too scared to use RBE to modify my bios. If anyone have the time and knows how to mod my attached bios to 0.950v ill be thankful. Do you guys think it will take of any noticable strain from the psu to downvolt from 1.0 to .950 ?

will these PSU´s work on the M15x?

DELL AC-Adaptor 210W, 19.5V 10.8A (PA-7E)
DELL AC-Adaptor 240W, 19.5V 12.3A (PA-9E)

Or will they toast the system?

 

i did that same mistake mike, its not the gpu temp... that's the "Alienware M15x EC" Temp 0 reading.
not sure about SC2 but.
not sure which one it is though

i'm not, however i get random shutdowns during stress if overclocking the cpu using throttlestop. can anyone confirm if using the 240w stops these shutdowns? basically atm i have my cpu/gpu at stock. attempts to overclock result in random shutdown, issue is not with temps.

 

No, I only have the 150W standard adapter and it worked fine... never had issues while benching, except for driver crashes.
Never had shutdowns while gaming, except for the one time when I forgot to change the fan speed, hahahaha (But thanks to Mumak we have now automatic fan control... he's the hero!)

So it should work fine... maybe if there are fluctuations in the line voltage it could cause problems... that's the only thing I could think about.

Line voltage over here is 230V, but this shouldn't make a difference.


@MrValle: I can mod them for you if you want, but you could also test it with ATI Tray Tools before you flash the vbios, this tool allows you to lower the voltage.

 

I have tried ATI tray tools and it was too complicated to me to understand where i could modify the voltage.

 

Hmm, I haven't got an ATI card in my system at the moment... and ATT is uninstalled.

You need to go to the overclocking settings, then add your new, custom voltages and then you can use these values to set it. It's pretty easy once you found where you have to create the values.
Sorry, can't make a guide atm.
Maybe someone else can make a screenshot.

 

The psu is so hot i almost burn myself on it. I tried playing Entropia Universe with stealth mode on and i could play it without shutdowns. But as soon as i play normaly i get shutdown after only 10-15mins. I even tried running gpu on .950v and it didnt help.

Its this 840QM cpu that is sucking the power out of my psu.

 

I have 2 150w adapters. I have an ES version GPU which sucks all power as well as an over clocked 920XM. Running 1.1v stock gaming for many hours. Have you padded in the correct places? I had to do one small high tech intricate modification using a hammer. I'll post pics of it tomorrow. It allows better contact between heatsink and core as well as the aluminium block and the memory.

 

After I put a huge piece of thermal pad on card's back, no shundown any more.
Have you done like this?

 

The combination of the BIOS warnings and the extreme heat from the unit make it almost certain that the PSU is defective. TDP on an 840QM is only 45W, so the CPU is not the culprit. Chances are you just have a bad power brick, which was not exposed until the higher load from the 6970M was added. I have a 920XM, which can pull twice as much power as a 840QM, and while the PSU would get very warm under bench level overclocking, it was nowhere near the temperature levels you describe. Call Dell and get a replacement.

 

Hey guys, sorry it is the Alienware M15x EC Temp0 reading that is stuck at 90degrees. I did a bit of searching and seems that this is just a bug in the hwinfo program?

In regards to the issues that you're having MrValle, I have the same card as you, and admittedly my card is not overclocked but it is not undervolted either. My power brick doesn't get anywhere near as hot as you describe, and I haven't seen temperatures over 75degrees yet while gaming. Haven't done any benching yet. I did spend quite a while measuring up and cutting out new thermal pads for the memory and the other chips that were able to make contact with the heatsink. I also used the pads I bought from mxmupgrade for padding the chips on the back of the card. So far so good. Fingers crossed!

It's a nice card with great performance, but I must admit I was a little disappointed with the performance in Starcraft 2, the frame rates are good but I have to use vertical sync to prevent screen tearing, and then when scrolling over some terrain it gets choppy. This is with everything on maximum settings. I'm wondering if going to a later video driver might help.. what's the current latest and greatest for this card?

On a side note, I upgraded to the 920xm at the same time as I replaced the card.. big difference on loading times for games, I was quite amazed.

Cheers guys

 

No, it's not a bug in HWiNFO, I asked Mumak, the developer of HWiNFO about this... the system simply reports 90°C for this diode, HWiNFO displays it correctly.
So it's a fault with the diode on the mobo, but it doesn't matter.

That's nothing unusual, I have vsync on in every game, it simply looks better and feels smoother. Don't know why a lot of people want to have vsync off...

 

I called dell and they said that if its a PSU failure i should look at the blue light on the psu cable. If that light goes out during an computer shutdown its a psu failure. And i tried that and no the light stays on.

The computer shuts down even with the battery. If psu fails shouldnt it just jump over to battery mode?

Maybe its ram that shuts computer down? is this possible since i have upgraded ram about a month ago? <--Swapped back original ram and problem persisted.

Dell will now send me a new PSU to try out which was kind even though i dont have any warranty left.

The psu info where it says 150w in bios sometimes says unknown. And also every time i start my computer it starts for a half second or less and instantly dies again, and after about 1-2 seconds it starts again and boots up normaly. Kind of a lagged start. This i have had for quite some time now and dont know if its the psu.

 

There are defects short of total failure. The blue ring light will go off when there is a total failure, i.e., when the brick is fried. Obviously, that does not apply in your case, but equally obvious is the fact that the bios warnings and the extreme heat are telling you that problems exist. The bios warnings occur before any load is applied and are reporting that all is not well with the PSU. The heat and shutdowns confirm the warning. The Dell CSR is being glib or just doesn't know how to respond to a problem that is not defined on his expert tree guide. There may be other problems, it's hard to determine until the power supply is stable.

Edit: My post was written before your message was edited. Glad to hear that Dell is sending a new PSU. That will at least resolve the issue one way or the other. Good luck.

 

Ok. Thank you all for your helpful insights and sorry if i spammed this thread with my problem. I will be receiving the new 150w psu maybe tomorrow he said. But no later than monday and i will report here if it did the trick.

Hopefully other people who has this problem will read here and know what to do and where the problem lies.

Now i play the waiting game

 

MrValle. I had the same thing where when you turn the computer on it can like re-shut down for a couple seconds. For me it could do this over and over infinitely. I eventually found out I'm pretty sure the problem was that in the bios I had overclocked the ram which is impossible so it didn't like it. check if this is it.

 

I run everything on stock. Maybe all of these problems are with the 6970m graphics card. But i doubt it, cant really relate the 6970m with the problems. But who knows.

 

as i suspected, so mine is the PSU since the light goes out. i have to physically unplug the PSU and replug it for it to work again.

depends how it fails.
i'm just glad it's not blowing the system up due to overload.

 

Ok, have you called dell for a replacement PSU?
I dont have any warranty left since 6 months back. But they will replace psu if you call and tell them your light goes out during a shutdown. Im guessing psu is under warranty indefinately. Like its meant to last a lifetime.

My light stays on during shutdown but i described all my symtoms to the dell tech and asked if they could send me a psu so i can try to see if it works.

 

i'll give them a call, i too ran out of warranty.

 

I was thinking to undervolt the 2D and idle voltages 0.05 V less each to get lower temps. Sometimes when watching videos the card gets a little bit hot with the clocks running at 300/900.

Can I brick the card doing this?

 

I tried it with my card and it didn't work, it defaulted to the original voltages. Dunno why.

 

I've had the random shutdown of the PSU also but only while being extremely over locked with a xm940 and over clocked GPU during benching. I confirmed I was pulling more then 200 watts of power with a meter attached to my psu. I'm currently waiting on my 240w psu and will bench at the same settings I had the failures at. It's to bad too cause the bench I had before the failure I had a CPU score in the 21000's but didn't save the screen shot cause I wanted to push alittle harder.

 

Ok, i got a new psu and it still shuts down. And i still have that delayed boot-up of the computer where fans and lights light up when i press powerbutton but only for a few milliseconds and goes all dead. And after maybe 2 seconds the computer autostarts as normal without me pressing power.

I opened up my computer and put a big thermal pad under the x-bracket to shield the components under. And also padded other chips under the card. And still it shuts down but i could run the comp a bit longer now.

I den reflashed the card back to 1.1v to see if it wasnt getting enough power and shut down the computer. But now it shutsdown a bit faster again.

There is some component somewhere that gets too hot and makes the card shutdown. But what component might that be, or maybe my mother board is toasted.

I can play forever when i run in stealth mode no shutdowns happening then.

These are the temps i get just when computer shuts down:

Diode: 69
DispIO: 69
MemIO: 74
Shader: 68

Or maybe the GPU has been damaged and now its more heat sensitive than normaly.

 

@KurtH (and MrValle): What are your specs? 200W is more than I ever read on my kill-o-watt, even with hardcore overclocks.

Maybe having a harddisk and more RAM increases the power draw a bit... but I can hardly believe that it makes such a difference.

 

I thought I should just warn people who buy the Eurocom 6970m card that even if they get the correct screws, you may still have problems.

I was waiting on the screws from Canada, and it ends up they were having a postal strike there. I was sick of waiting, so I decided to drill and tap the mountings so they would accept the M15x heatsink screws. I did this and everything fit perfectly, I applied the thermal paste but noticed that the heatsink was not spreading the paste very well. I fired up the computer and things were fine at idle, but as soon as I started up a game the temperatures skyrocketed over 100degrees! It seems those mountings are too tall to allow a good contact between the M15x heatsink and the die on the 6970m. I decided to drill the mounts completely out like xeroxide, and now no issues.

So the moral of the story is: If you buy a 6970m from upgradeyourlaptop, make sure you get one without the soldered in mounts!

On a side note, can you use the official amd 11.6 drivers on the non-dell cards? And if so, do you uninstall the old drivers first or just install over the top?

Cheers,


Mike

 

11.6 were the last drivers I used and they worked very well. I recommend checking out the leaked 11.7, The Rev created a thread with the links to them.

I usually uninstall the drivers, restart, maybe use CCleaner and install the new drivers.
But I also realized that simply installing the new over the previous one works pretty good too.

 

You might be right about the HD and ram not to mention I was trying to take top benches for my specs I forgot what Throttlestop settings I was using but they were crazy. First few times benching I was able to get on all me leader boards even when my scores were not updated to reflect them. I'll be doing a few more benches soon.

 

i guess it's a habit from the old days, but i always uninstall from control panel, reboot into safe mode and run driver sweeper. this is especially true if mucking around with beta drivers then reverting back to whql. i take it your card's running well now with temps under control?
another success, dells problem with the m15x is that they just made it too damn good

@kurt you're able to draw 200watts without shut down?
you sir have a laptop from the gods. treat that beast well.

 

Oh its shut down on me 2 times now. I was freaked the first time it happened intill i started looking into the PSU problem

 

Ok i padded the hell out of my card. Now i can play a bit longer. But still shutdown. And this last shutdown my temps showed highets temp 78.5 on memIO.

I then immediately opened my computer after the shutdown and felt around on the heatsink and it was too hot to touch. There must be something terribly wrong. Those temp readings i get are totaly wrong, they are giving me wrong numbers because these shutdowns are caused by heat im now sure about that. The whole heatpipe was so hot i couldnt keep my fingers on it for more than 2 seconds.

edit: it was as i suspected.

Temps reported i get with heavy GPU gaming with Dell 6970m bios is:
Diode: 69
DispIO: 69
MemIO: 74
Shader: 68

and i get shutdowns from overheat.

Temps reported i get with original bios that shipped with the card is:

Diode: 77+
DispIO 78+
MemIO 87 and still rising
Shader 81+

Shutdowns alot faster !

(HWiNFO used as temp monitor)

 

Well don't forget, 78C is roughly 170F. That is very hot. I'm not surprised you can't touch that for too long. That's hot.

 

silly question mrvalle but the fans are working right?
and they're blowing air out through the exhausts at the rear of the laptop?

if the heatpipes are hot, it means it's removing heat from the chip which is good, but the fans aren't working fast enough (or at all). only time i've ever seen this happen was in a friends desktop gpu where the fan came factory wired the wrong way. it was spinning, just not in the direction it should have.

 

Yeah, my computer have been working fine until i upgraded to 6970m. The 6970m started shutting down out of the blue. And after i upgraded from i5 to i7 it started shutting down more often. I have no idea what to do anymore. Might me mobo that is tired, or bad gpu

And i can play well in stealth mode, thats weird getting around 65c on memory temp with stealth on.

 

Which brand of 6970 are you using?

 

The one from ebay, eurocomm something.

 

HI guys……I finally upgraded DELL 6970m today, and everything is fine: performance, temp, fan and no shutdown yet.

 

Right so you got an official one?

 

yeah, it's right^^