HELP ALIENWARE 17 (Ranger) R9 M290x GPU FAILURE? GPU FAN FAILURE?

Hello Laptop experts I'd really need your help with regard to my Alienware 17.

The specs are the ff:
Core i7 4710mq
R9 m290x
128gb SSD M.Sata
Ttb sata hard drive
8gb ddr3 ram
Windows 10 Pro OS


I bought this laptop 2nd hand on Fleabay. When I first used it, it was working for a few days until one day, my r9 m290x couldn't be detected in the BIOS anymore. So basically its showing that I don't have a discrete gpu. I was able to do a quick fix on this by reading various forums and that I needed to re-flash the BIOS. It worked and I was able to play as usual then by the next day, the GPU couldn't be detected again and then I would reflash the BIOS again so it would be detected again. I would do this process a couple of time so that my GPU would be detected.

Soon afterwards, the GPU not being detected problem went by then another problem came in. This time the video card will be detected but its showing Code 43 which I thought was a driver issue. So I did what others would do and uninstall the drivers using DDU then I'd try to reinstall its crimson drivers and mid way during driver installation of the display adapter, my alienware would freeze. I did a reformat of my system but still it would hang midway again while installing the gpu driver. I tried using a different OS, Windows 7 and the driver would be successfully be installed but when I restart the system, its showing code 43 in the hardware device manager.

I knew it was a hardware issue already so I had my r9 m290x reballed. After the GPU being reballed, I was finally able to make it work again and the drivers installed successfully. I was able to play a few games and did some benchmarks testing using furmark. I wanted to make sure that this time it wouldn't fail me so I installed HW INFO and did some custom fan settings for both the gpu and cpu fan to ensure that the fans will spin faster to avoid overheat. After configuring the custom fan settings, I decided to to do a stress test again on my GPU using furmark. I noticed that the CPU fan was blowing hard while the GPU fan was totally silent. After a few minutes, my screen locked up and I had to turn off manually my laptop and turn it on again.

I'm then back to square one and my GPU couldn't be detected anymore and it seems to have totally given up on me.

I'm wondering what could be the issue? Before the GPU's latest failure when I tried to stress test it through furmark after doing some custom fan settings for HW INFO, I believe the GPU fan failed because there was no sound or hot coming out of the GPU vent while the CPU vent was on full speed mode. I also opened my system afterwards and touched my GPU heat sink and it was definitely burning hot.

Now I'm confused on what I'm supposed to do. Is it because my GPU fans aren't working from time to time perhaps why the GPU couldn't be detected in the bios? And it led to the my GPU failing due to overheat? Or is it just my GPU malfunctioning on its own? Or maybe the power socket of my gpu fan in the motherboard is the problem?

I had a long and tiring journey and I'm back to square one and my GPU couldn't be detected any longer even if I try to reflash the bios again. I'm thinking of replacing my gpu fan and the video card but I fear this system has more problems and I might end up wasting my money.

Any advice from you guys would be of great help.

 

The fan isn't broken, when using HWinfo64 for fan control you have to properly choose the settings otherwise GPU fan won't work search Techinferno forums I believe @Mr. Fox has a tutorial there.

I stopped using custom fan profiles a while ago after extensive testing because they yielded the same temps for me in long sessions, it only helped me lower my low usage/idle temps but the noise was unbearable.

 

Hi, thanks for the response buddy. I guess won't be using HW INFO anytime soon. Another weird thing though when I brought my unit to a local repair technician, they tested another video card on it and the fans also didn't work and fried their test video card. Did i do too much bios flashing on this unit? What would be the best BIOS you can recommend?

 

I had the same fan 'stoppage' when messing around with fan control in HWInfo. It was too easy (IMO) to make a mistake so the advice is good to research it. I got lucky in that I was monitoring temps and as soon as I saw 90c I powered it down.
Furmark has also got a lot of bad press in the last 18 months or so and it's generally accepted that it WILL destroy a card that is 'borderline' .

 

Wow, I must have had bad luck :/ After the GPU dying out on me, i was able to revive it through reball and then now it's dead again coz of what I did with HWInfo fan control :/ Anyway thanks for sharing your experience. I badly need a replacement r9 m290x now

 

Maybe the heat damaged the MXM port on the motherboard, can the technicians check it out?

 

Hi there!
Sorry for your trouble
I got the basicly the same Ranger. 16gb ram and win8.1 though...opted out of win10, never even tried it.

Anyhow, I bought it new two years ago and after a month or two had some similar issues.
Happened two times I think - gpu temp went through the roof - 104c - the fans didnt kick in at all it seems. Im using HW for monitoring from day one, never messed with fan profiles.
Noticed the fps dropped down, went back to desktop and the gpu temp graph went off the scale.
My analysis is it had something to do with Desktop mode battery charge, when it was time for it to be enabled it somehow confused the fans. I think so because while I was playing a game Desktop mode becomes enabled, the laptop runs of the battery to drain it down to 75% and then it stays at 75%.
Anyhow, thats what I was able to think of, something stopped the fans from kicking in
It never hapenned again.
I did repaste cpu and gpu with icd7, regularly clean the fans

 

I just bought a replacement r9 m290x and indeed the fans aren't broken at all. My sytem's up and running and I'm currently playing NBA2k17 at 60 FPS. I guess the default settings in HWinfo64 messed up the fan settings and the GPU fan didn't spin at all when I ran Furmark. This is definitely a lesson learned for me.

 

True, I had to research by enabling some settings in HWinfo in order to configure its 2 fans. I have replaced my r9 m290x with same model. Only problem I have now is that there are only 3 speeds for the cpu fan (0, 2000 and 4600 rpm) and gpu fan (0, 2000 and 3400 rpm) Any way to configure the fan speeds?

 

Thanks for sharing your thoughts buddy. The culrprit is HWinfo. It made my gpu fan stop while I was stress testing using Furmark. I am now more careful with HWInfo and I bought a new r9 m290x again.

 

Glad you sorted it out.

How much did you pay for the new r9?
Why didnt you go for 780m or something else?

I had the same experience as I described before just yesterday, almost two years after the previous time it happened.
The laptop is almost always plugged in, high performance plan, desktop battery mode.
I have unplugged it, moved it to my room to watch a movie (plugged it in again) but it stopped the desktop battery mode it seems.
So when I got it back to my desk and plugged it in again it charged the battery to 100%, meaning it started the desktop cycle again (if it stays plugged for few weeks it goes to 75% and stays there).
I went on playing WoT and noticed sluggish fps, imidiately switched to desktop and behold - gpu graph on HWinfo went through the roof! 104c

There you have it, I am convinced that it has something to do with desktop battery mode switch - stops the fans.

 

Glad you sorted your problems out. HWinfo if one is not careful, would do more harm than good.

 

If I had the budget, yeah I would have went for a 780m or even a 980m but I'm too broke lol. I can only shell out $200 max I got my replacement R9 m290x from an Ebay seller from China. I paid $179.00 which was free shipping (Singapore Post w/ tracking). It arrived here in my county, the Philippines after 2 weeks. I read from various forums that desktop battery mode isn't really doing anything good to our laptops. It would be best to just turn it off buddy.

 

Thanks! Yeah, I guess I learned the hard way "If it ain't broke, don't fix it". Too bad I had my dead r9 m290x revived through reball then to die again because of my stupidity in using Hw Info lol.

 

Wow you literally cooked your GPU to death....poor R9 M290X

FYI never use furmark, ever. It is a power hungry program, unrealistically taxes the GPU. Better to run 3dmark11 for stability testing, I do 5 passes before I call something stable.

 

I really did :/ I can still remember taking the GPU out and it was painfully hot. This is duly noted. I'll use 3dmark11 moving forward.

 

So, which app will you use now to control your fans, since hwinfo is not so reliable?

 

I now completely avoid Furmark but I still use hwinfo given that I've done my research and you need to configure it in order to ensure that both fans are being controlled. (I disabled EC support in the settings)

 

you'd be surprised how often that's happened to Alienware users, LOL