technos

High Klem, I go by username Raidriar over on Notebookreview, I was wondering if you can make undervolted GTX 980M vBIOS for Alienware 18 using original Dell vBIOS images from here for primary and secondary cards. It MUST use these vBIOS images or else the secondary fan does not activate and the secondary card cooks itself. Please lock maximum voltage to 1.000V (or 0.9870V) https://www.dell.com/support/home/us/en/4/Drivers/DriversDetails?driverId=9W9F7 Thank you

Hi J95, I've been using your drivers for single 980M in the M18x R2 without issue. However, I recently added a second 980M, and now when I enable SLI, I get driver crash every 5-10 seconds and the computer is really unusable. Goes away when I disable SLI. No reported problems in device manager or dell ePSA test

I release here today my personal BIOS for the M18x R2. It is based upon svl7's unlocked A11, Swick's mods to A11, and I have changed back the boot images to default Alienware and Intel logos, as well as reverted the atheros LAN oprom back to the default version that does not take up extra boot time. All other modifications done by swick and svl7 remain intact. Copy files from zip to root of usb in a usb-dos environment, run flash.bat. Simple. Tested by me on two of my M18x R2 systems, works without issue. I take no responsibility if you brick your system, flash at your own risk. Only thing left to do is to somehow enable 2133 for sandy bridge CPU in M18x R2 motherboard. Remember, give thanks to svl7 and swick for their initial work on this. I release this freely to the public, nobody should pay for this. Download: https://drive.google.com/file/d/0B5dGmVHDQ7iuNERzYkMwMW1RMDQ/view?usp=sharing

I have a strange issue as of late. After recovering from soft brick from XTU settings, CPU fan decides not to come on at all...great. CPU is bouncing off thermal limiter. I decide to reflash bios, now CPU fan stuck on low setting. I don't understand how the machine EC has developed a new consciousness for the worse. Any idea how to get the fan to properly step up and step down fan speed? it is bordering on the ridiculous now. No such issues exist with the M18x R1 or M18x R2....sigh. Alienware 18 is such a piece of sh!t.

I'm still waiting on Swick to divulge exactly what he did for the M18x R2 to retrofit Maxwell legacy support. I would like to port this to the m18x R1 if possible.

It is resolved now. Thanks!

I need to change my name over on this board...I made this in 2011 when I was figuring out 6990M undervolting. This is Raidriar over from NBR I was going to run RX 480s in all my systems on a matter of not supporting Nvidia's ridiculous consumer atrocities for the past few years. Now it seems the only answer is the 980M in legacy platforms. On a separate note, I find it slightly amusing how the GTX 980M is the final card for these LVDS systems, kinda how the 9800M GTX (similar names) was the final card for MXM 2.1/HE users before MXM 3.0B was introduced.

No LVDS support RIP all the Alienware guys except M17x R3, R4, and 17 who had the 120Hz panels.

It doesn't happen exclusively with Firefox, it also happens with simply using Windows Explorer, which I assume is also hardware accelerated graphics for aero effect...Will wait for the new drivers and see if I can come up with a workaround. Some more experimentation showed this is only an issue with SLI enabled. When SLI is disabled, this issue goes away. PhysX has no impact on it and neither does the HD audio driver. I would say power states trigger the issue, but I see it changing power states on 353.00 without issue. Aero hardware accelerated graphics and firefox hardware accelerated graphics work fine too on 353.00 without any modification

hey j95, This is raidriar over from NBR. I've been using your modded drivers for a long time with 780M SLI and now I try with the 980M SLI. I have a very weird issue in windows 7 when using the latest drivers. Whenever I am doing light work (browsing, typing, navigating windows, etc), the driver freaks out and crashes randomly (black screen for 3 seconds, revert back, says nvidia kernel driver crashes). I reverted to the 353.00 throttle free driver, and the issue is gone. I've tried several drivers from the 36x.xx and 37x.xx branches, all give me this issue. Any ideas?

Hey svl7, I am a big fan and would like to thank you for all your hard work on behalf of the Alienware scene. I come with a request, would it be possible to have Alienware M18x R2 BIOS A12 unlocked? It did some sort of update to the EC that help fix strange fan behavior on the primary GPU that is absent in BIOS A10, which is your latest unlocked BIOS. Would it be possible to unlock A12 as well? Thank you for your time. BIOS here: Alienware M18xR4 System BIOS for Win8.1 Driver Details | Dell US Filename: M18R2A12WIN.exe

Ah, that sucks. Forgot you had a 120Hz panel.

You should be able to use the 980M just fine in the M17x R4. We support optimus switching, should work just fine.

You are correct, you must forgive me. I was pairing the BIOS deactivation of the iGPU and the MUX as one piece instead of listing them separate.

How would a MUX setup work with optimus then? The definition of a MUX is to disable one GPU and switch to another, unless you have a setup to keep both GPUs enabled and switch betwen the two.

Wrong. Optimus relies on MUXless switching. If you have a MUX, you do not have optimus. You have what is called "fixed switching". End of story. Optimus runs on a muxless dynamic switching scheme, which is impossible to have on an Nvidia card on a MUX design. If you are running optimus, you are dynamically routing the video either to the iGPU or the dGPU, without restarting, though AMD managed to achieve fixed switching without restarts. If you have a MUX, you have to restart to change from iGPU to dGPU if you are running on an nVidia card on the MXM slot. Have a look at the diagram here The Alienware 18 is the only system Alienware has that uses a MUX switch. The Alienware 14 an 17 use a muxless design, which is why they both support optimus.

Then it should work on the M17x R4 correct? It uses MUXless switching, which is optimus capable.

Guys, I think we will all have our answer when Dell decides to release its own 980M. Modifying a VBIOS is great and all, but its not going to compare to Dell's own engineers writing a custom vBIOS for it. I still think switchable graphics is the main culprit here, there may be a different implementation being worked on, which may explain Dell's delay in announcing any notebooks with a 970M/980M. Just wait for the Dell card to be released, you all will be wasting your money until then. SVL7 is a great modder, but there are some things even he can't fix (like retrofitting new nvidia cards while keeping 3D on the M17x series).

A 3820qm is not an unlocked chip, it will still abide by hardware limits in the CPU as defined by intel. A 3920xm/3940xm or ES chips do not follow the limits

Can we have the bios A13 unlocked please? This is probably going to be the last bios they release for the M17x R4 anyway, it has reached end of life.

Hey I downloaded A10 unlocked, but the needed PRRtool is not to be found in the zip, where did it go?

Guess I'm moving up to A09 to give it a shot. Hopefully I won't regret it.

ouch. thats scary, I thought we were able to revive bricked vbios in the past.

be careful guys, it is possible to perma brick, i've done it before and its not pretty.

why not search for references in the m17xr3 bios for 580m device ids and/or strings and replace with 675m or 680m strings/ids