*OFFICIAL* M11x Benchmark Thread

people should hook up an external and do 1280x1024....that is the standard that the rest of the world uses...you are just citing the highest res that the screen on the M11x can support

That makes the 9000 scores seem much less significant to me...

 

this is true, but not all of us use external screens, i run all games at 720p.
but the GT335 should only go down maybe 100-150 points at 1280x1024. so if someone is getting 9100, they could still be very close to 9k

 

How are you guys getting your memory clocks so high? I cant get over 948. I tried both msi afterburner and EVGA precision.

It works fine at 948 so I wanna see if I could go higher, but whenever i type in anything over 948, it just reverts it back to that.

 

I still don't see how you guys even get over 6300! I can barely beat that score. I'll pinmod when my warranty is FULLY useless (R1). I'll also try to close down a crapload of junk. What are a good STABLE clocks of the GPU?

 

my highest stable right now are: 560/911/1440.
also you need to turn off ALL background processes and unneeded services before running benchmarks.
you should see an increase in your score.

 

my clocks are in my sig. i dont do any speacial tweaking, i just only install what i need (dont have comand center installed) and have my start up programs pretty well kept.

im pretty sure you can get around 7k with a nice STABLE GPU OC.

 

My R2 is not stable at 166, hence I run 164. I expect I need to mod the memory timings to push that a bit higher.

I run throttlestop, that seems to make the biggest difference with the i7.

The A03 bios mod that ups the voltage on the GPU allows me to clock it higher than I ever have before. The settings for the 9k runs do not translate into stable game play, I have not tested lower values yet.

I have used Gamebooster to stop processes but tbh that only seems to make a slight difference.

Other than that I am running full fat WIn7 ultimate.

 

9120 once i set intel HD to max battery

 

You've pinmodded your system. Which I will do in a couple of months ! I'll rerun the 3DMark06. My nVidia drivers crashed today because I tried to play Crysis again with unstable clocks. Got me a nice lockup.

 

Un-pause your monitoring. The clock speed in the monitoring window shows at which clock your system is.

 

That wouldn't matter. It would show idle clocks just like in your non-OC proof. Need GPUZ.

 

8639

i5 @175BLK with only 2gb memory. 335m at 550/1335/948.

Any idea how I can get my memory clock higher? Whenever I try to input anything over 948, (950 for example). It goes back down to 948 automatically. I tried both evga and msi afterburner.

 

Just try a higher jump, sometimes it the card cannot do the resolution that the program does for frequencies

 

i think the memory slider stops moving past 948mhz

 

Try nVidia Inspector. It supports a much higher limit with the Unlock Max option.

 

You really need to be at my side to believe my OC. When I first OC, my shader hits 1540, then USUALLY my GPU crashes and fails. Then I restart and reapply then if I have luck the OC will work.

 

I don't have to. Your proof shows you are running at stock clocks. What kind of "STABLE" OC relies on luck?

 

which version of 3dmark06 are people using? the paid or free version?

 

doesn't matter

 

ok. earlier i ran the cinebench R10 tests and the i7 R2 scored much better than the i5.

single CPU rendering: 2768
multi CPU rendering: 5701

will install 3dmark06 and see how the stock settings score.

EDIT: 3DMark06 score = 7180 for stock settings.

 

Which i5?

Single CPU: 3274
Multi CPU: 6653

 

a stock i5.

Alienware M11x R2 Review - Direct Comparions R1 vs R2

site is probably bias anyways

 

That site doesn't use Throttlestop.
At stock, I get:
Single CPU: 2550
Multi CPU: 4967

Those results are much better, showing the i7 having a 16% lead instead of a 31% lead, correlating to it's 16% higher dual core clock speed.

Hardware Heaven. Not surprised, they probably spent more time putting the graphs together then actually testing the M11x'es. Too much fluff.

 

i agree very much so...

wasnt a very good review for benchmark sake. but then again there isnt many out at all really.

 

This Anandtech review is better with more R1 comparisons and doesn't drag out it's (lack of) content across many pages but still flawed without Throttlestop. It already has a point headstart by LISTING the specs of the product being reviewed. Oh wells, stock vs stock is good enough I guess.

 

i remember reading that, that was a very good review for the time. not only shown the benchmarks of the R2 against the R1, but many onther laptops as well.

 

Unfortunately NO review will ever be done at anything other than stock speeds, the reviewers do have to represent what a user would get out of the box afterall.

This is something that always ticks me off with Tom's Hardware laptop reviews as the laptops have so much more potential than what they show

 

and not all CPUs will OC to the same settings either!

 

Not comparing OC'ed machines. Only Throttlestop.

 

so stock + TS? not OC +TS?

 

Stock +TS yes.

 

Actually it relies on cooling . But with a fan controller it helps. What kind of proof do you need? List it and I will prove (not including 3DMark06 because it crashes on the "water dragon" scene".

 

A GPUZ screenshot of the sensors page showing the 335 running at 3D clocks would be a start.
Like this one:

 

and do you have any idea why 3Dmark06 is crashing?

ill give you one guess....

 

alright i ran Heaven DX11 Benchmark 2.5 on my R1 at 1.73ghz (modded) and my gpu at 605/911/1510.

on DX9 with quality on high, im able to get 27.5 fps on average on the benchmark.

BEAT THAT! just kidding.

 

i noticed that Heaven cares way more about the GPU than the CPU. so the person with the highest GPU OC is likely to have the highest score regardless of R1 or R2.

 

hey everyone i thought i would share my game benchmark results. these are with the CPU at 1.73ghz, and the GPU at 605/911/1510. and also with the stock gpu clocks.

first off, Lead and Gold - Gangs Of the Wild West. highest settings. low shadows. 1280x720 (fraps tested)

with no gpu oc, at a specific spot, 74fps.
gpu OC'd. 94fps at the same spot.

Burnout Paradise. highest settings, low shadows. 2x AA 1280x720
stays at a constant 60 fps regardless of clocks.

Crossfire FPS game. highest settings, 1024x768.

120-235fps with the gpu OC'd
110-206fps with no OC.

these are the only games i have installed as of right now, but if you would like me to test more i could, this just proves that a good gpu overclock can make a very big difference.

edit: i failed to mention i used Nvidia mobility modder to modify the 263 drivers that dell provided. i did this because i heard it can increase performance a lot.

 

Can you run time demos/ingame benchmarks of popular games so people can better compare performance? Such as Resident Evil 5, Street Fighter 4, Far Cry 2, Crysis, GTA4 ingame benchmarks.

 

yep, i will install resident evil 5 as soon as possible. and maybe far cry 2.
(only other 2 games i own at the moment)
with that high of fps on the games i play, i dont see any reason to upgrade to the GT540 anytime soon.
even better when my 2ghz mod is done tomorrow.

 

alright,
i tried the resident evil 5 benchmark.

stock, dx10, shadows low, everything else high, no motion blur. i got 51.2 fps.

overclocked, 57.1 fps.

good improvement if you ask me.

 

How about Metro 2033 with max out settings at 1366x768???

 

That wont play! I'm sure you could get something pretty decent looking if you mess around a bit with the settings.

 

il try metro 2033 when i get the chance, im downloading far cry 2 right now, and il post my results tomorrow.

 

LOL, I doubt a slide show will be very enjoyable.

Metro 2033 using a custom config (mixture or low and normal) is "playable" @ 1366x768 but only with a hefty GPU overclock (585/1404/1760 in my case).

I say "playable" as this was tested during the beginning portions of the game which are less demanding than latter sections (especially outside).

As for the benchmarks in general we really need to see framerates plotted against time to see where the real benefits are. This is easy enough to do using Fraps and Excel (open office). Do one run with GPU at stock and one run with it overclocked and have both runs showing on one graph.

This style of results are far more indicative of real world gameplay than just a absolute average number.

Games like Crysis and Metro 2033 should show a nice boost across the whole graph but there may be sections that don't show much of a boost at all.

It might also be useful on a couple of games that show a real boost to dig deeper and compare the specific "sections" of the GPU stock and overclocked. This would show us where, in a real GPU limited game, the real benefits are (Core clock and thus Textures & ROPS, Shaders or memory bandwidth).

 

I didn't know that Metro 2033 is even more hardware-killer than Cryis xD

 

http://forum.notebookreview.com/ali...alienware-m11x-benchmark-thread-part-2-a.html