M18X R2 Three Beep Error Code, No Post.

Hello,

A little over a year ago, I was fishing around to upgrade my old Lenovo laptop and took an interest in the old Alienware models and their large “built like a tank” sizes like the M18x R2. I’ve read up on a number of users experiences of folks still using this laptop today and figured a small bid on eBay couldn’t hurt. Even on the off chance that I won and it was rough around the edges, I’m no stranger to working on desktops and laptops. Oh, how naïve I was.

I don’t have a spec list on hand and the seller didn’t exactly save a list of those for easy accessibility. So here’s what I can recall:
Some i7 2.60 GHz
16GB of RAM(4x4GB)
2 GTX 675Ms
1 Samsung 850 EVO
2 500GB HDDs

When I first received it and powered it on, I was greeted by a sea of red pixels. Also at random, the screen appears to shut off as if there is a phantom primary display it switches to. This usually happens during logging into or operating windows but it has happened during boot. I decided to install an SSD to speed up the boot times while troubleshooting, but to my surprise, the HDD flex cable arced burning the cable and the socket. Leaving me to purchase a second motherboard and hard drive cable.

So with those replaced, I was back to functionality on an SSD but still stuck with the previous two issues. I ruled out via the HDMI port to another monitor that the LCD panel was causing the issue and have replaced it; fixing the red pixels from what I could see. I also replaced the flex cable a third time because the second eventually snapped from repetitive handling. Powering up as I said, fixed the red pixel issue, but the screen shut off issues began to happen more frequently and at earlier points in startup. Eventually, the laptop straight up won’t even boot up anymore, returning a three beep error code (beeeep--beep-beep) implying a chipset failure.

So I grumbled and purchased a third motherboard hoping this would be the end of it. But not only does the three beep error code remain but now no lights will turn on. I even questioned if the fans were receiving power until I physically checked them to see they were spinning at a very low RPM. I’ve still have a three-month warranty to return the item but the seller attests there’s nothing wrong with the board saying it passed all their tests (although they will accept my return). What if he’s right? It’s pretty bizarre to have the same error code across two boards and the only parts that have really stayed the same are the GPUs, CPU, RAM, and the media expansion board. I’ve already tested booting with each single RAM chip and the computer seems to be able to detect each individual one, only complaining when there are no RAM chips inserted. Could any of these cause a chipset error?

In summary:

My M18x R2 will not boot regardless of the motherboard so far. It insists with three beeps that there is a chipset error. The seller will take my motherboard back, but if he’s right about it being completely functional, then I’m still back to square one on what could be causing this issue. I’m considering this repair service as opposed to purchasing someone else’s “defective board” for a higher price. They’ll only take the board which should limit what else they could bust up in the process. But I’ll be a little nettled if it turns out said motherboard isn’t the actual problem and something else is the cause.

I also still will probably have the screen shut off issue once I get the thing to boot properly. I have a theory it’s probably a faulty bios setting not staying on the correct display method or if I should straight up not install the GPUs. Not being able to boot prevents me from exploring this further though.

I like to think I’ve learned my lesson in bidding on used laptops, but it sure would be nice to be able to actually use this massive paperweight for even a little while after all the work I’ve put into it. Any aid this forum could provide me would be greatly appreciated.

 

Your GPU(s) is/are dying/dead.

Try to let it run only 1 GPU, see if it works, chances of both your GPU dying are rather small, so I guess you'll have to run single GTX for now.

also:
3 short beeps = Bad chipset
1 long, 2 short = Bad Video

Also, your old LCD Panel might be still fine.

Another good advice:
When purchasing new GPU's / GPU for your notebook, please make sure they are no older than ether the HD 8xxx series or GTX 7xx series. All cards from the fermi / early kepler and the AMD's 6xxx / 7xxxseries are currently all dying.

If you're looking for the best performance per buck, look no further than the HD 8970M

Also your CPU is likely a I7 - 3720QM 2.6ghz / 3.6ghz turbo. (which is kinda weird, normally the M18x do have extreme edition CPU's)

 

Sorry that it took so long to get back to you. I tried as you suggested and I tossed in my second motherboard to see if I could get a stable boot. Never realized there was a difference between 3 beeps and 1 long and 2 short.

While I no longer have three beeps, lights turn back on, and all the usual sounds I expect emit properly; I'm now wrestling with the 8 beeps each of the same length and spacing apart. From what I understand, the left side of the motherboard is the primary GPU, correct? I've swapped heatsinks to move what I'm assuming is the secondary GPU to the primary slot, but no change.

I'm open to running just on integrated graphics for the time being with no connected GPUs but that also produced 8 beeps. Did I get the primary slot wrong or do I need to hook back up my old LCD panel as the beep code suggest?

Thank you for your assistance.

 

The M18x doesn't have integrated graphics. The dGPU direcly talks to the LCD while on optimus systems the Intel graphicscard talks to the Panel.
8 beeps means LCD panel doesn't respond/turn on, which makes sense if you only attached a graphicscard to the secondary slot, try to swap the card onto the other slot and see what happens.

Also try to swap the card with your other graphicscard and try both slots in case the first GPU didn't work out.

 

Back again, this time with good news straight from the once dead laptop as I type. As you expected, I mistook the primary slot and am now running on a single GPU. Stress tests seem to indicate no signs of the display prematurely shutting off, so it looks like that problem's sorted out. Seems the computer was last in dGPU mode and switching back and forth isn't as intuitive as I thought. The latest IGPU drivers will even brick the system when switching back to dGPU mode and requires a CMOS reset to function again. Oh well I'll just have to stick with one and minimize my battery usage.

Two minor issues still remain, but I've never been so giddy before just to be able to move my mouse wildly across a blank desktop. Firstly, my media keys on the keyboard rest do not seem to be active, period. No lights, no feedback, nothing. It seems unlikely its a hardware issue as the cable does not appear to be loose. Software seem to be the more likely culprit, but I haven't had much luck re-installing the OSD drivers.

Secondly, I've noticed this rig runs pretty hot, even on idle despite being on a single card solution. Like 57 to 62 degrees hot. And shutting off my flair programs such as Rainmeter and Wallpaper Engine do not significantly impact this temp reading. I'm suspecting my the fans are on some sort of quiet setting but there doesn't seem to be much I can do to adjust it on the A11 BIOS. I'm considering if one of the custom BIOS offerings here might allow for more control or looking into some software solutions. I've yet to try running a game to see how it fairs just yet.

Because the last user installed his config in RAID mode, I'm going to have to reinstall the OS anyway in AHCI mode in order to truly take advantage of my SDD's speed. I'd like to be able to solve the problem here and now so that I know how to deal with it before I wipe the slate clean, but perhaps the act of wiping may yet be the solution.

At the very least, my year-plus-long struggle nears its close. Thanks for your assistance.

 

Media keys will be active once you installed the proper drivers.

Idle temps are normally rather hot, lapping and proper thermal compound will fix it mostly. You should probably buy dual HD 8970M's, those are by far the best in terms of price per performance ratio.

 

Wrong. M18x has integrated graphics in the form of intel hd 3000 or intel hd 4000 that can be set in such a way where the PCIe slot for the MXM card is completely turned off and only the iGPU is running.

LCD failure is extremely rare on Alienwares in the LED age. 8 beeps tends to mean video card failure 9/10 times on Alienwares.

OP, check your board and make sure your have a PCH cooler present with a thermal pad underneath it (little metal square, two screws holding it down in the middle-left area of the motherboard with the GPU slot side up.

GTX 675M is a garbage rebadge of the garbage GTX 580M, which is known to have frequent core failure. Pull your GPUs out and try running the computer without ANY MXM graphics cards installed, this will force the BIOS to use the iGPU (intel hd 4000 in your case). Before you power on, make sure you disconnect your CMOS battery and main battery, hold the power button for 30 seconds, reconnect your CMOS battery, main battery, and turn on again. You should hear 5 beeps indicating CMOS has been cleared. Shut off, turn back on, and you should be able to POST barring any motherboard or RAM failure.

 

Good Evening,

As you suggested I went ahead and checked the PCH cooler. I even unscrewed it to ensure the thermal pad was present, it all checks out. Running without the GPU didn't go so well as it seems the CMOS reset also sets the system back to dGPU mode which understandably causes problems when one isn't connected. I could verify this theory but I'd like to minimize my attempts of stripping/dressing the laptop before another flex cable decides its at its limit. I'll remain on iGPU mode for the foreseeable future until I at least upgrade the GPUs.

As for the temperature issues, I had a lead referring the proper application of thermal paste. I tend to spread the paste over the entirety of the metal plate when that's supposedly not necessary. So I've tried the line method while wiping any excess near the edges, but the result was only a hotter CPU. Additionally, I've noticed only one of the fans seem to be spinning while in iGPU mode at a quiet setting(The one in the center). I would think 68 degrees Celsius is alarming enough to warrant more fan power or the kicking on of additional fans but to no avail. I made sure to run an ePSA test to ensure a fan didn't die, and all three definitely span at their max 4000+ RPM.

I could repaste the CPU back to what I usually did, but its looking like the fan curve is the real problem to my temps. My Thermal Grizzle Hydronaut paste hasn't failed me yet, I'd be surprised if it is now.

EDIT: I stress tested the CPU for about an hour in Prime95 and the highest temp I got was 81. The second fan also turned on the help with cooling and now two fans are spinning at a much notable RPM. So I guess the thing just runs hot. I'll at least repaste the CPU using my older method I believe that resulted in lower temps anyway. Guess everthings ready to go for a windows re-installation now for AHCI mode.