*OFFICIAL* Alienware M18x Benchmark Thread

[Mr. Fox]

Wow, that doesn't sound good at all Mr Fox. Out of curiosity were you overclocked at the time of these issues? And if so was it a light OC or heavy OC?

I have had mine only a few weeks but it has been the most rock solid gaming laptop I've owned. I think I was able to BSOD it by OC the CPU too much but other than that it just seems to do whatever I ask of it.

No, when that happened (as posted in the screen shot above) I just turned on my computer after being off for a while. I don't frequently overclock and when I do so, it's generally just for a benchmark.

[Zero989]

I wonder if it's the work of Sapphire Trixx. I don't have it installed atm and so far so good at stock speeds.

[dajohu]

I wonder if it's the work of Sapphire Trixx. I don't have it installed atm and so far so good at stock speeds.

I kinda wondered that too. I have only used the built in overclocking with the drivers and have never had any issues. I did have to change the settings in the BIOS to go higher than default would allow. I think I changed the max to something like 900/1200.

It's just like with the NVidia cards I've had. I always use the built in OC tools unless there is some reason not to.

[Mr. Fox]

No, I don't use Trixx much any more. It does not load with Windows and has to be manually launched. I use almost exclusively Overdrive in CCC when I want to benchmark.

However, after this problem occurred I fired up Trixx this morning to help diagnose what was happening. Like GPUZ and Catalyst Control Center, Trixx also saw only one video card. Device Manager saw both, both enabled and no problems indicated by Device Manager.

[Zero989]

I kinda wondered that too. I have only used the built in overclocking with the drivers and have never had any issues. I did have to change the settings in the BIOS to go higher than default would allow. I think I changed the max to something like 900/1200.

It's just like with the NVidia cards I've had. I always use the built in OC tools unless there is some reason not to.

ewww ntune? that program is horrible . msi afterburner is probably the best. it has continuous support.

[daveh98]

No, when that happened (as posted in the screen shot above) I just turned on my computer after being off for a while. I don't frequently overclock and when I do so, it's generally just for a benchmark.

It seems to me that you have the problems when overclocking or after you have overclocked. Maybe the cards have become unstable. When they stabilize, it seems to only have problems after another OC attempt. Maybe your cards, at the end of the day, were not good samples to OC. When you continually used Furmark and other stress testing resources, coupled with continued OC attempts, you may have damaged your cards.

Depending on your driver, it can cause the crossfire to be grayed out. The preview drivers did that for me.

I don't know what to say, but hopefully other people take note to some of the actual problems and risks associated with overclocking and the conseqeunt headaches that pursue.

Hopefully you can find a good solution. I only have the 6970s right now but I have been very happy with the performance and the drivers from AMD. I was die hard Nvida years ago and have to say AMD and their catalyst suites have really impressed me. Outside of a few games that definitely tip their hat to Nvidia (through optimizations and gimping of AMD) I have to say they are very solid cards. The 6990s came out much faster than the 6970's and outside of the extra shaders, they are clocked higher but based on the same architecture. Getting stable and high overclocks, outside of a few outliers seems more difficult on these cards.

It furthers the real need a lot of notebook members have to understand when it comes to overclocking notebook GPU's you run much greater risks given thermal and design sacrifices to even getting these cards to work at spec. A benchmark really is, at it's core a measure of system performance at spec. hardcore benching on laptops is relatively novel and while I am not against it, it does show bring with it much more considerations and risk.

For people, like yourself, that rarely overclock as you say, it would seem futile to even engage in it for a sole purpose of a higher bench as benching is something, on the desktop front, a passion and very purposeful. But for casual overclocking it really does not serve any point other than, "how far can I push my cards." But without a lot of dedication and reason and passion to take those calculated risks, you end up with scenarios like this in which the costs much outweighed the benefits for a "quick bench here and there."

IMO of course but hopefully some people can see that benching is a very risky process and really only a select few on the forums should be doing this; especially considering the price of notebook Vs desktop parts, thermal constraints, etc.

[johnksss]

the money i spent for just this one notebook....i could bench quad sli 580s under 4 ln2 pots + dual 590's to the 8 580's i already binned.

or about 10 highend amd gpus under ln2.

or bin about 40 2600k cpus next to the 15 i already did.

or 4 990x or 4 980x to the 3 i binned there.

[Xeneize]

the money i spent for just this one notebook....i could bench quad sli 580s under 4 ln2 pots + dual 590's to the 8 580's i already binned.

or about 10 highend amd gpus under ln2.

or bin about 40 2600k cpus next to the 15 i already did.

or 4 990x or 4 980x to the 3 i binned there.

Something tells me you will be benching those in a couple of weeks lol

[johnksss]

the only ones i didnt bench were the 590's...they were not over clock friendly till tin went to work on them...

[Zero989]

the only ones i didnt bench were the 590's...they were not over clock friendly till tin went to work on them...

damn youre so lucky lol. hopefully some nvidia users or yourself post arkham city benchmarks w/ physx running. i always wondered how well 580m sli handled physx.