*OFFICIAL* M11x Benchmark Thread

Welcome to the first installment of the Alienware M11x Benchmark thread here on NBR. Please note, this thread will cover 11xR1 and 11xR2. Like the 17x thread, we will stick with one for both releases.

DR650SE has volunteered to track scores. Thanks DR650SE!

There are several apps which can be used to benchmark your system. Following is a short list of some which are typically used. There are others and the core benchmarking members will have plenty of suggestions - so ask away.

• 
  FutureMark Benchmark Apps
• PCMark Vantage
• 3DMark Vantage
• 3DMark 06
  
  Component Benchmark Tools
• POVRay v3.7 (multi-threaded version)
• Fritz 9 Chess Benchmark - standalone
• SuperPi Mod v1.5
• HDTune
• Everest Ultimate - Trial Version
• HWiNFO32
• MaxxPI²
  
  Game Benchmark Tools
• S.T.A.L.K.E.R.: Call of Pripyat benchmark
• Resident Evil 5 Benchmark
• Unigine Heaven
• Crysis Benchmark Tool v1.05

A word of caution concerning benchmarking... While benchmarking / number chasing can be quite a bit of fun, please be aware that benchmarking your system is pushing your system to its limits. It can and will stress your components. Do so at your own risk - nobody is responsible for this but you.

For those of you concerned with overclocking via the BIOS, NBR's Jerrry Jackson was able to verify with Dell/AW:

As with all threads here on NBR, please remember the forum rules when posting. One specific forum rule which I request everyone pay special attention to is:

This is done to streamline the thread and not disrupt formatting. Thanks for your understanding and cooperation with this.

Finally, this thread is specifically for benchmark related discussion. No Chit Chat on your funky mouse pad you just ordered or anything unrelated to benchmarking. Benchmark questions, results are fine. We are going to be very strict with this given the amount of activity on the M11x. Also, let's keep this to M11x benchmarks - we have other threads for M17x, M15x benchmarks. Post those there and not here. If you want to reference one of the other systems, link to the post.

Have fun folks and let’s see how far you can push the M11x!

 

Welcome to the M11X Benchmark Thread
Last updated on: 11 JANUARY 2011
----------------------------------------------------
NOTE: When posting your benchmarks, please include a screenshot of your final score, along with CPU-Z and GPU-Z in your screenshot alongside your score. This is to verify your information. Without screenshots with the information, scores will not be included in the top 5. If you need an example please reference benchmark links in my sig. If you have questions how to get the screenshots, please PM me and I will explain it. This is not the case with gaming benchmarks that have in game benchmark tools. Please refrain from attaching large photos (greater then 500x500) causing slow load times and excessive scrolling. You can use links to your photos from sites such as imageshack.us. If you have qualified into the top scores, please PM me with a link to your post so that I may link your score in this post. I will try my best to update this post on a weekly basis. Please PM me if you wish to add something to the list and we can discuss it. Now let's have fun and push the system!

** **PLEASE READ THIS LINK BEFORE POSTING** **

All runs done at default settings, native resolution, 1280 x 768.
All 3DMark Vantage runs done with PhysX OFF
----------------------------------------------------
Synthetic Benchmarks
* Denotes SU7300 CPU
* Denotes SU4100 CPU
---
3DMark06: Top 5

(1) DavyGT - 7897 *
(2) DR650SE - 7129 *
(3) Mandrake - 7098 *
(4) orangeclockwork - 7046 *
(5) jesmith8 - 7036 *




3DMark Vantage: Top 5
*Vantage runs done at default 1280x1024 to recieve an overall score (Use an external monitor)

(1) m11xx - P 3331 *
(2) DR650SE - P 3308 *
(3) DR650SE - P 3278 *
(4) Fuzzyhead - P 3192 *
(5) Chewietobbacca - P 3011 *





PCMark Vantage: Top 5

(1) Chewietobbacca - 3084 *
(2)
(3)
(4)
(5)

----------------------------------------------------
Top Gaming Benchmarks
---
BioShock:

Company Of Heros: aznguyen316 - 31.1 FPS* / All High settings

Crysis: unreal25 - 31.145 FPS* / All Med settings, No AA | aznguyen316 - 30.3 FPS* / All Med settings, No AA, DOX drivers 195.62 | icarus_II - 28.08 FPS* / All Med settings, No AA

Crysis Warhead:

Devil May Cry 4: natbyte - 86.75 FPS* Rating: A | tuxblondinet: - 70.53 FPS* Rating: A

Fallout 3: unreal25 - 44.4 FPS* / Settings High, 4X AA

Far Cry 2: Fuzzyhead - 35.99 FPS*

Grand Theft Auto IV: icarus_II - 25.92 FPS*

Left 4 Dead 2:

Resident Evil 5: natbyte - 49.3 FPS* Rating B | DR650SE - 48.1 FPS* Rating B | KoukFC3S - 43.2 FPS* Rating B

S.T.A.L.K.E.R. Call of Pripyat: kitir - *AVG: Day 38.0, Night 37.0, Rain 40.0 | unreal25 - *AVG: Day 36.6, Night 33.8, Rain 37.1 | aznguyen316 - *AVG: Day 34.1, Night 31.6, Rain 34.4

Street Fighter IV: DR650SE - 89.85 FPS* | diabiosx - 75.83 FPS* | aznguyen316 - 71.11 FPS*

Team Fortress 2: YodaGoneMad - 30+ FPS* / 36 Player Server

Tom Clancy Hawx: DR650SE - 47 FPS*

Half Life 2: KoukFC3S: - 95.82 FPS* /Max settings No AA

World in Conflict: DR650SE - 83 FPS* / Very Low Settings

----------------------------------------------------
**NOTE**: All gaming discussion not specifically related to benchmarking of built in benchmarking tools will be removed from this thread. Gaming discussion can be found in the **Official M11X Gaming Discussion Thread**. DO NOT make FPS requests please.

 

Dox 3DMark06 run, Crysis, Street Fighter IV, Resident Evil 5, Stalker CoP benchmarks

OC'd GPU 3Dmark06, Starcraft II beta, Company of Heroes benchmarks

SSD benchmark via Crystal Disk Mark

 

I'll test BC2, I already have a key.

 

If anyone could post some empire total war numbers that'd be awesome

 

just a very premature update on very brief benchmarks as of right now

overclocked at 1.73, got the upgrade on cpu, 500 gb, 4gb ram

Crysis - All high with anti turned off. Very beginning of the first stage fps was around 16-20. I played up until the first battle.

All medium with anti turned off same situation fps was around 26-40.

LFD2 - All high with anti turned to 2x i believe ran 30

Mass Effect - Max settings with anti turned off ran in the 30s

The Sims 3 - Max Settings was in the 40s, the game felt slowish tho

Dragon Age Origins - Max Settings was in the 20s to low 30s ( was slow whenever you initiated and got out of a convo)

Borderlands on all high with glare, shadowing turned off ran 35-50

just tried out batman AA as well and on max settings with AA off the beginning sequencd got me fps in the 20s although in the first battle i got in the 30s and low 40s

l4d2 again with multicore on, AA off, sync off, settings on max
Dark Carnival 2nd Stage
min 18, max 93, avg 44.4333

just ran the re5 benchmark on 1280X720 with aa off, blur off
settings on high
directx 10
variable scenes
42.7 fps with a grade of B
63.3 scene 1, 37.9, 37.4, 39.3 for the other scenes

 

Alright so I didn't have enough time today to do anything beyond synthetic benchmarks. That is, no real-world in-game benchmarks.

First of all, I'd like to preface this since I know there will be users and/or various trolls out there who love to bring up numbers as a point of comparison.

Let me state the fact that I hate 3dMark06. Yes, it is the most common benchmarking tool out there, but that doesn't mean it has a serious flaw. And that serious flaw is that it is way too CPU intense. This is in fact why it is still used as a benchmark tool for the extreme desktop overclockers out there - because it isn't completely GPU reliant, and thus fast GPU + fast CPU is required. Of course, those guys also put dry ice on over-volted computer parts to obtain those world records.

To illustrate this situation, I took 3dMark06 runs on the desktop. The desktop is running a QX9650 Core 2 Quad and an ATI 4870 X2 (overclocked at 800mhz core). These runs are all at the default 3dMark06 settings (1280x1024 resolution).

With the QX9650 overclocked at 3.6 GHz, and the ATI 4870X2 clocked at 800Mhz, I scored 19,402 marks.


With the ATI 4870X2 downlocked 25% (from 800Mhz to 600Mhz core), I scored 18,768 marks.


Now I downclocked my CPU by 25% (to 2.7GHz from 3.6Ghz) whereas I left my 4870X2 at 800Mhz. Lo and behold, my score dropped to.. 15,078 marks.


Of course, if I were actually using this to play a game, that 20% drop in GPU is going to hurt my frame rates a lot more than that 20% drop in CPU. I hope this illustrates how futile it is to compare actual gaming performance across different notebooks with different CPU's, since 3dMark06 is so incredibly CPU dependent. Vantage is going to be a better indicator of real world performance since it is far less CPU intensive.

Anyways, I didn't have a lot of time to do benches, but this is what I do have for today. I will do more tomorrow when I get a chance. Please note the max temperatures of the CPU and the GPU after each run... this thing is cooled nicely! The Core 2 doesn't really reach above 60C and the GPU barely crosses it as well! Very nice

My configuration:

SU7300
4GB DDR3 RAM
250GB 7200RPM HDD

First off, a PCMarkVantage run at default settings. That's right, no CPU overclock (I will run this while I sleep with the CPU OC'd since PCMark takes forever). Score of 2,389.


First 3dMark06 run was everything at default. That is, 3dMark06 with no CPU overclock, no GPU overclock, resolution of 1280 x 768. Scored 5352 marks.


Next, I attached this to my 20" Dell monitor to run 1280x1024 resolution. This is 3dMark06 at the full default - no overclocks, resolution of 1280x1024. Scored 5159 marks. Interesting to note that despite cranking up the resolution, the performance didn't suffer *that* much.


Next, 3dMarkVantage at default settings for performance preset. That means 1280x1024 resolution. Scored P2541.


Next, I did the BIOS overclock to 1.6GHz. Note that for some reason, 3dMark06 thinks its a 1.73GHz.

Anywho, this is 3dMark06 at 1280 x 768. Score of 6208


Now at 1280 x 1024 resolution. Score of 5670 marks!


And then I ran 3dMarkVantage again. Same settings for Performance, 1280x1024 resolution, etc. but this time OC'd. Scored P2664


As you can tell, the CPU OC wasn't that big of a deal for Vantage, which is a better indicator of ingame performance since its score is much more GPU based than 3dMark06.

Now to see what happens when the resolution is increased, especially for you guys who like to run things on external monitors. I ran this at 1680 x 1050, and scored 5213. Needless to say, out of a ultraportable, that's some awesome power and ought to show you that the GT335M is quite capable.


I also ran the feature tests on 3dMark06. According to my math, the GT335M's 72SP's are paired with 24 TMUs, and 8-12 ROPs. Big difference over the GT330M which probably has 16 TMUs and 8ROPs.


And finally, I didn't get a whole lot of chances to play around with OC'ing the GPU... but here goes. I overclocked the core to 525MHz, and Shader to 1260 MHz... and at 1280x1024 resolution, I got 6066 marks!


Alright, going to sleep. The sun's already up hopefully I don't have to catch up with 20 more pages tomorrow

 

If you want to play the comparison game, the 6066 3dmark06 score is close to my Acer AS5740G's score at the same resolution (it has a i5-520M and ATI 5650 for comparison). Then again, I think those kinds of comparisons are meaningless

Oh and a little teaser of what may be to come!

 

Be sure to disable PhysX in the NV Control Panel prior to doing your runs. While it will impact 3DMark06 (not for the better either-lower scores in my experience) for Vantage it will inflate your CPU score.

When benching, its best if it is disabled for a true hardware result. Also, as I am sure you are aware, driver versions will all produce somewhat different results. Some bench better while others are worse.

Have fun - thanks for posting all of this. Nice to see some actual screenshots!

 

THANKS CHEWIETOBBACCA for the awesome benchmarks and the comparisons ^^

 

Yes PhysX is disabled.

Actually for Vantage, they apparently changed the CPU scoring so that PhysX is no longer counted. I did a run with PhysX on just to see what happened, and while I got 367+ fps in the physics test, it didn't change my score by much (well within the range of error)... so apparently Futuremark decided to remove the physics aspect.

I'll put up a pic for it tomorrow. The physics thing is actually cool to watch when it's not at < 10 fps

 

awesome man thank you! +rep, now I want to see Vantage 330m scores and see how much better the 335m is =) Great to see the GPU OC's well too. Never seen that program, only ever used nVidia System Tools to OC, but I like the look for your teaser pic program

 

Great job chewie! Accept 3DMark06 for what it is and you don't have to hate it. It a great test for when you are CPU OC'ing and it's free for people to run over and over again.

 

Hater... j/k, Nice job, I'll try and getting around to updating Post #2 tonight when I get off work. I have a filter at work that blocks image hosting sites. How far did you OC the GPU? Get the display drivers to crash yet?

+1, tommorrow I've been too generous

 

Interesting, looks like there is a new patch for Vantage but I don't have a nVidia card to test.

 

See Drake - I guess I did miss it... LOL. I'll let you know what I find when I have time to check it...

 

Nice job on the benches.

As for WOW, I had mentioned before I will be running our weekly 25 man ICC tomorrow at 6 server and will try to get some video during some of the more graphic intensive fights (ie. Marrowgar). I use an incredible amount of addons being RL, so dont be shocked when you see my screen while tanking!

+repped you as well.

 

Great job!
It's good to finally see some numbers and backup AW stated 6100 in 3DMark06.It might not be the best program to run benchmarks but at least it gives something to go by.
Will be waiting for more, so far it looks like it can game really well
+ rep

 

as for the 3DMarks06 (free trial version): mine's got 6320 OC'd
not doing other benchings...

 

I just ran mine and got 6382 in OC mode at the default resolution (1280x768)

 

WEI for OC w/ max specs (8gb SSD su7300)

4.6
5.4
6.4
6.4
7.0

 

Well, I have been in the process of doing some OC of the GPU to see how much extra we can get out of 3dmark 06, and the bad news is things seems very much CPU bound.

I am still testing, but even 10% bumps in core/shader are only yielding 30-40 points in the bench.

Still working will report more as I find out.

 

Yes, OCing your gpu will do nothing in 3dmark 06 scores. You would have to test Vantage to see the difference.

 

Well 3dmark06 is more CPU intensive. Why not run Vantage instead for the GPU. I'm not even planning on running 3dmark06 when I get mine.

EDIT: Impruv beat me to it...

 

Guys let me know if you want these benchmarks at naitive resolution of 1280 X 768, or 1280 X 1024. As of now the top score(s) are at native resolution. Thats can be adjusted if you want to bench at the standard 1280 x 1024

 

Thanks guys. I'll check out more things after I take care of some work today

Yep hehe, I don't have a problem with it when CPU OC'ing is around. To test GPU performance though, not as much!

 

Guys what about some su4100 vs su7300 bechmarks .

 

SU4100vsSU7300 benchmarks has been beaten to death in the acer 1410/1810 thread

 

Hey, i know at least two members would like to see benchmarks done on GTA IV.... anyone up for that, or am i going to have to wait till i get mine to do it?

 

Not sure if anyone posted this yet but a member sneakerhead at SD Forums posted his benchmarks for SU4100:

SU4100 at 1.3 not overclocked
3d marks score: 5420
sm 2.0 score: 2385
sm 3.0 score: 2752
cpu score: 1151

SU4100 at 1.73 overclocked
3d marks score: 6246
sm 2.0 score: 2893
sm 3.0 score: 3003
cpu score: 1341

Not bad for the base CPU, kinda makes me feel better I didn't shell out $100 extra for the SU7300 I'm cheap...

 

Dell claims it will run at 30fps, not sure at what settings, but it was one of the games they listed... not sure if I fully believe it, but it should per Dell.

 

Here is PCMarkVantage run at an OC of 1.6 GHz. Score went from 2389 -> 3084

So I was mistaken, but PhysX drivers are actually not installed by default on the M11X. So none of the tests are done with PhysX. I did install the latest PhysX drivers and turned it on to see if anything would happen, and it did indeed change the CPU score as expected. The score at default clocks + OC'd cpu was bumped to: P3261




GPU OC

I haven't tried to find the max overclocking on the GPU but I did run it at 550/1320/870 fine.

I've also made a run with 3dMarkVantage with PhysX off, but the GPU overclocked at Core/Shader/Memory frequency of 550/1320/870 MHz. Scored P3011


For comparison, my Acer AS5740G 15.6" notebook with a Core i5-520M and ATI 5650 at stock scored ~P3200 That's incredible performance out of the GT 335M!

And one more, for 3dMark06 at 1280x768 at 550/1320/870. Scored 6454 marks!

 

@chewie

Those are nice numbers there. very impressive for only 1.6GHz

 

I just did a stock run with just the basic OC on the CPU and got P2690.

PhysX is disabled.

 

@chewie

"So I was mistaken, but PhysX drivers are actually not installed by default on the M11X. So none of the tests are done with PhysX. I did install the latest PhysX drivers and turned it on to see if anything would happen, and it did indeed change the CPU score as expected. The score at default clocks + OC'd cpu was bumped to: P3261"

That seems odd. Thought that was standard as part of 330 and 335m level drivers? We all need to install from Nvidia site? Quite a few games appear to be beginning to use the feature.

 

Yes, you'll have to go online to download the actual PhysX drivers, as it is not pre-installed

 

I was able to enable the PHYSX in the nvidia panel. I broke 3K without it enabled. I enabled now and will see what I get.

 

I'm not trying to be nit picky cause I really do appreciate everyone's hard work, but there is a total lack on screenshots in a benchmark thread, agreed?

 

I understand that, but really at this point its not a competition. I have been running the benchmarks to see what I can set my system up for during my regular gaming.


The good from what I see:

System runs cool even at pretty high overclocks.

The bad:

From my tests we are still cpu bound at around 3k (without physx).

It will be interesting to see what the more extreme overclockers will be able to do.... Maybe a PLL CPU unlock mod. That will probably be needed to break 3500 even with Physx

 

yeah it's pretty cool although 3Dmark06 doesn't get the GPU to as high as many games I noticed. WHen I had a Sager w/ GTX 260m it got upwards to 79C highest after an hour of Crysis but it never got that high on 3DMark06 runs.

 

Very impressive with the M11x. I think we will see someone breaking the 7k mark soon.

 

I honestly don't think that will happen untill that rumored BIOS update comes around. If it does, the CPU can be overclocked to 1.73.

 

True. I wonder if AW/Dell are holding back and we will get more then 1.73Ghz OC out of the BIOS update?

 

^^ j1ngles has the 4100 - posted a few pics of the benchmarks in the other thread here: http://forum.notebookreview.com/showpost.php?p=5918100&postcount=2139

Though he hasn't done exactly what you're looking for (in system properties - it's showing 1.73 - but he didn't provide screenshots of CPU-Z, etc.) - I'm sure if you ask him - he'll be glad to test it out ..

 

so, according to Chewie's and J1ngles' benchmarks, there's only a 17 point diffrence between the the su4100 and the su7300. I know it's only a synthetic benchmark, but it's still interesting.

 

My scores fluctuate a bit... my latest 3dMark06 score was 50 points higher than the one I posted, but yeah, it is indeed negligible

 

I'd wager to bet that, while overall, the synthetic benchmarks aren't off by too much between the 2 processors - you'll find that the 7300 actually performs better at multitasking, etc.

I used to have an older laptop, and due to a screen issue with it, got an updated "newer" laptop - same processor "family" in similar fashion as the 4100 vs. 7300 - in that it was a minor step up with additional cache.

In my day-to-day tasks, (coding, compiling, email, browsing, *games*, etc.), the "newer" processor was much "snappier" than the other version - even though they were technically the same processor, etc.

A 1MB cache may not seem that different - but it does make an overall difference / impact to the machine. Add to that an SSD, and the additional RAM (or even bumping up to quicker RAM) - I think you'd be surprised at the difference.

Once many people have their systems in - I'm sure we'll start to see several benchmarks and comparisons that indicate that "bump" in overall system responsiveness.

Additionally, while I see everyone clamoring for a Core series in this laptop - I don't think the technology is there yet - either from a low-wattage/performance perspective - nor from a production line standpoint. Will we see Core processors in this unit? Most definitely - but I wouldn't expect to see them before the holiday season later this year. And even then, it'll be at a *much* higher initial price point.

 

I agree with you jsgiv. BTW here's a cache size comparison benchmarks via Tom's Hardware I posted this in the Owner's lounge but I guess it's more relevent here.

Cache size benchmarks

 

i dont understand this, some people are saying BFBC2 is unplayable, others say its fine. some people say they get 3500 3d mark 06 scores and others get 6500's.

am i missing something or do these conflicting figures just not make any sence.

 

The guy with the 3500 3dMark06 scores was using different settings then everyone else. He changed and retested and it came out to the norm everyone else has been getting.

As for BFBC2, maybe people did not disable the shadows?