*OFFICIAL* Alienware M18xR1/R2/18 Benchmark Thread - Part 3

here is mz total default score vantage

not bad ?

look at this ...

 

for the maxwell 880m it seems the graphics is rebranded from 780m with just a bit higher clock.. ewww

 

Quick overclock on the 3840qm with modified bios and xtu. Switching over to the 3920xm

http://valid.canardpc.com/ff51bd

 

I can't help but think nVidia will hold back on anything earth shattering on Q1 Maxwell release. I wonder if AMD will have something to show off.

 

you might be a bit disappointed as 4.4 or 4.5ghz is really you'll be able to get out of 3920xm on a daily usage without heat throttling. but if you do it for benchmark sake to push for high end score with some external cooling then you have made the right choice. xm series are awesome in that regards.

they call it maxwell, but it is still 28nm and cause of that they cant fit more config core if its a rebrand. overclocking core and memory clock we can all do that so the real bonus here is the 8gb vram. I certainly won't use 8gb unless i go multi monitor or aa x16 everything on ultra at 1080p, which I don't do lol

cmon AMD! R290M !!!!!

 

I will probably play around with the 3920xm for a month and than probably get rid of it. Not even sure why I picked up the 3920xm when the 3840qm is plenty powerful and hits my magic number of 4.1+ ghz. I don't really care about hitting 4.5ghz-4.75ghz for the additional cost but love to tinker with XM chips. Once I get bored I will probably sell it and break even on the sale.

 

There's no replacement for displacement
Keep your XM!

 

I know I know. I can flip the 3840qm on ebay and break even easily on the sale and get a free 3920xm upgrade. Only issue is when it is time to get warranty support and Alienware wants the OEM cpu back in for exchange. I will have to research and see if I can install the 3840qm in the M14x R2 and overclock it. Not sure if a modified bios is available for it or not.

 

aw dont say that man. the point of having 3920xm is you gotta overclock the heck out of it. right now the bottleneck in my machine is CPU so Im trying to get as high of clock speed as possible lol. even with ramdisk and raid0 array I can tell many of my software isn't much faster than it was on 1 SSD or 1 HDD and only explanation to that is likely the CPU. for example profile of 300 saved tabs in my firefox takes roughly 14 seconds to load up completely in my raided SSD but takes 8 sec in my ramdisk. now ramdisk is at least 7-8 times faster and its latency is multiple times faster too but difference is only 6 seconds, clear sign of my CPU bottlenecking in this area

 

my 2960 was the best cpu ive had to date.. this 3920xm blows major behind(for benching that is) unless i just missed the mark.. but i was schooled by 2 of the best.. it wont go beyond 4.2 4.3 with out blue screening.. going to start looking for a 3940xm i think..

 

What sort of flex were you dialing in Dumb Dumb?

 

did a little playing NVIDIA GeForce GTX 780M video card benchmark result - Intel Core i7-4930MX,Alienware 01W2J2

 

ohhh snap

Newegg.com - CORSAIR Vengeance Performance 16GB (2 x 8G) 204-Pin DDR3 SO-DIMM DDR3 2133 Laptop Memory Model CMSX16GX3M2B2133C11 TIME TO BUY

TIME TO BUY

pretty lame only in the states, I NEED IT IN newegg.ca so i can grab two sets damn it newegg

oh yeah, tri screen gaming on a laptop, two portable USB POWERED display.
ASUS MB168B+ USB-powered 15.6-inch Full HD Portable Monitor Review | TweakTown

too bad its only 15.6 inch

 

the price is to hight

 

so is alienware laptop lol this is the RAM FOR ME PLS WORK IMMA BUY NOW

 

Is there much difference in replacing the cooling paste on M18?
Im trying to make it run on 4,3 ghz but it overheats so very fast its silly.

 

unless you use liquid metal or something similar, no. overheating caused by fan not running until temperature is too high but imho I think theres just not enough fan force or enough area surface overall on the radiator. I could use a compressed gas can and blow at heatsink or cpu all I want and overclock to 5.0ghz in my R2 and not going over 75C in intel burn test. with enough pressure and surface area it will cool.

in other words, theres nothing you can do unless you mod the heatsink, or try and go for 4.2ghz instead =)

 

Nope, it's easy to do the repaste. You don't need to mod the heat sink. What we need is fans that actually run when they need to. Don't even bother with anything except for IC Diamond or the Liquid Ultra stuff that kh90123 just used or you will be wasting your time and money. None of the other pastes do a good enough job. Repasting will help, but until Alienware releases a BIOS/EC update with the fan tables fixed you're going to have a machine that runs much hotter than necessary and with degraded performance. Haswell is already runs way too hot without any help, and the engineers making the fans not work right to try to cut down on noise exacerbates it something fierce. The people that made the decision to do this obviously were not paying close enough attention to what a screwed up product Haswell is. Otherwise, they would have recognized the effort to make the fans too quiet was going to create a far more serious problem that having audible fans. This is what happens when people work in a siloed process and focus on one small piece of a product or project instead of paying attention to the big picture.

 

Something different... I don't think I have seen anyone post an AIDA64 benchmark in a really long time.
The GPUGPU benchmark might be one that some of our AMD brothers with 7970M CF can beat 780M SLI.
(This was run with the GPUs at stock core and memory clock speeds.)

Merry Christmas everyone!

Spoiler

​

 

Ok, couse last time I tried @ 4,3 ghz with Mr. Fox settings it got to 85 in a millisec.. thats how fast. Even fans cannot start that fast..

 

Thought you guys might enjoy this:

NVIDIA GeForce GTX 780M video card benchmark result - Intel Core i7-4930K,Notebook P570WM

Over 18k in 3dmark 11.

4.4ghz 4930k + 780M SLI @ 1151/3500

http://www.3dmark.com/fs/1407575

Gives 6317 in firestrike extreme and 12339 in normal:

http://www.3dmark.com/fs/1407660

 

Wow. That's crazy!

 

Congratulations on an awesome score. And I agree with Brother Bumbo2. "You are Crazy!"

 

One too many crazy there. Bumbo2 should try to hit Mr. Fox's 3Dmark11 scores too, since you have the same CPU and GPUs.

 

Nice job, Brother Meaker. Thanks for sharing. :thumbsup:

 

"oops My Bad" it was a doble post. about benchmark I have no time for that, I have 3 kids and my job that keeps me busy.

 

Added to my index on the first page of this thread...

Spoiler

 

Meaker, those are the highest benches I have every seen! Are you running 2 PSU's?

 

Yes the 9570 requires them with dual gpus, 2×100w gpus +130w cpu would take an entire brick lol.

 

The Alienware M18xR1/R2/18 with an overclocked Extreme CPU and 780M SLI also exhaust the capacity of the single 330W AC adapter just as the 9570/P570WM does. Unfortunately, owners of the new Alienware 18 are unable to utilize a dual adapter setup like the M18xR1/R2 and Clevo P570WM and they are ultimately stuck with a partially usable system due to inadequate power supply.

If you look closer at the results you will see that the overall scores are attributable to the deliciously powerful 4930K 6-core CPU driving the physics scores in that beast. The graphics scores are consistent with what I and Johnksss have seen with 780M SLI in our M18xR2 machines. Due to inadequate power supply as mentioned above, and functional limitations inherent to Haswell processors, the new Alienware 18 is not capable of achieving performance similar to the Clevo P570W or M18xR2.

 

I get about 10% gpu score higher due to a higher core and memory clock.

 

The 780M on the 18 could be pushed a lot further, they just need a lot of power which trips the circuit breaker all the time.

 

for benches, try these TIM Results: Air Cooler; High Mounting Pressure - Thermal Paste Comparison, Part Two: 39 Products Get Tested

read part 1 first or just read that page for results mmmm
also IC Diamond would be roughly in the middle area I guess

 

I have tried a number of those pastes and IC Diamond is much better than somewhere in the middle. I would be much closer to the top end of the scale because it works better for me than MX-4 or PK-2 does, plus it is far more durable than PK-2 was. I never tried PK-3 because I was unimpressed by PK-2. MX-4 was good for only a few days. I repasted twice with MX-4 and threw away the rest of the tube because it performed poorly for me. I wonder why they did not include IC Diamond in their comparison... seems kind of strange that they included so many poor quality pastes and skipped IC Diamond.

 

Popped the 3840qm to mess with it some more and see how much I can extract out of it.

 

Now just hit that benchmark button for us

@ Mr Fox
I notice a lot of benchmarks conveniently leave out ICDiamond in their tests. I also agree IC Diamond wouldn't be far below Liquid Pro on that list. I does smell fishy in such a comprehensive comparison they leave out one of the best TIMs.

This liquid Pro seems awesome. I came so close to ordering some but the whole cleaning process seems like a PITA, plus a little risky using on GPUs. I just picked up more ICD and was done with it.

 

well it is Tomshardware we are talking about. if IC diamond think they are too good to give out sample to them i'd just lol. pk3 is rated at 11.2 while ic diamond is at 8 I think. these guys paste job is no joke, instead of just reading results should go to first part and see the details they cover.

I have pk3 and its roughly 2C better than my ic diamond, I only use ic diamond till mid this year because its about the only cheap thing here in store without ordering online. I wanna go the liquid metal pad but Ive seen its difficulties to remove and clean up and gave up on it.

 

Liquid Ultra is a PITA to remove. It stains the copper base. Luckily on the CPU side things are fine. Just that if you don't remove all of the liquid metal TIM, some tiny droplets of conductive liquid metal will stay behind.

Basically to remove it completely from the copper base one has to lap it. Without lapping it's possible to remove the TIM, but the stain will be there. It seeps into copper. I doubt the performance will be affected much at all, as long as the same TIM is used.

 

Haven't had a chance to benchmark and tweak the settings to get a stable 104.7 BCLK but I have a past run at 104.5 bclk and that is rock solid stable all day long.

XTU - faiz23 - Core i7 3840QM

 

I ran my ES 3820qm at 108.1 which I think 108.3 was highest I was able to get to. problem is at that high of bclk it runs on too much voltage, means higher temperature and shorter cpu life. 107 ish bclk on my 3920xm needed 1.45v to 1.5v to run, but on my es 3820qm only needs at around 1.35v at 108.1.

basically I think multiplier matters too, from 41x onward the voltage required goes up exponentially, which made sense cause my ES was at 39x multiplier.

 

Oh its a Clevo! Wow I thought the M18x was the most powerful notebook in existence. Guess this is that Clevo that can use two GPU's that I've heard of? Really impressive.

So I am a bit confused as to what performs better, 6 or 4 core Intel CPU's, but youre saying that the 6 core is a better performer?

AVADirect Fastest Gaming Notebook - Clevo P570WM

 

Hey all,

I just wanted to post the results of my OC'ing the past few hours.
I have a 4700mq, so I maxed the multi's using Intel's XTU.

I used an unlocked vBios to flash both 770's and upped the voltage to 1.050v.
I was able to achieve the following stable results. Ambient around 45c and max load of 75c.
Great cooling design by Alienware.

I decided to post this thread as there aren't a lot of configs out there with the dual 770m's, and I was having a hard time finding results to make my buying decision.



3DMark11 results a 10150, graphics score of 11400. This is around a desktop GTX 680, however the memory bandwidth is still at 192-bit on the 770m.
To compare, a single 780m benches around a 7200, and the SLI config around 11000.

Definitely not bad.

Thanks.

 

Running at similar clock speeds, an unlocked desktop hex core processor is most definitely better across the board, with 12 versus 8 threads. I would still steer clear of any machine (laptop or desktop) with a Haswell like a plague, but the 4930K is an Ivy-E CPU and it's a killer hex core CPU.

The Clevo P570WM with the 4930K desktop CPU beats a maxed out M18xR2 with XM CPU and 780M SLI, but that same M18xR2 is still the most powerful laptop in existence with a mobile processor. For many folks the Clevo P570WM will not be a good choice. Clevo's warranty is poor in comparison, the chassis is plastic, the aesthetics are very blasé if you like the Alienware look as most of us do, and the battery life is horrible. It requires the dual AC adapter setup to function in a strong performance mode (the M18xR2 only requires the dual AC adapter mod for extreme performance levels). While I am a definitely a PC gaming fan, I am more passionate about overclocked benching than I am playing games, so the P570WM is screaming out to me, LOL. However, for everything other than achieving crazy laptop benchmark scores the M18xR2 is still the best option available anywhere for gamers that don't wish to be tethered to a desk. I wish I could afford to own both machines.

Hi! Welcome to NBR. We are glad to have you in the Alienware Community. Congratulations on your machine. I merged your post with a dedicated M18xR1/R2/18 benchmark thread. This provides a centralized repository to keep track of benchmark performance for systems with all GPU configurations.

It's nice to see your 3DMark11 770M SLI benchmark. We always enjoy seeing our friends push their hardware to greater performance levels. More, please.

Have a look at the opening post and other information on the first page of this thread. Whenever possible, please post the links and screen shots for your results so we can get a detailed look at your achievements.

 

i cant stand to get clevo cause its still on x79. crappy pcie lane, sandy support, hot pcb mobo outdated nm size and of course crappy usb 3.0 support cause its considered a 6 series chipset. 7 series on ivy is so much better in terms of performance and I cant afford to lose that. aside from that m18xr2 also allows me to add storage mod, im unsure if clevo has even enough space for such a mod.

 

another run from raid0 SLC ssd
samsung pro/evo doesnt even come close
running under server 2012 stripe 32K

 

Since the Clevo P570WM is being discussed, these are my latest scores. You can see the physics boost with the hexacore CPU.

3dmark11: P18273
http://forum.notebookreview.com/sager-clevo-reviews-owners-lounges/707507-official-clevo-p570wm-p570wm3-sager-np9570-owners-lounge-21.html#post9507293

Firestrike Extreme: 6348
http://forum.notebookreview.com/sager-clevo-reviews-owners-lounges/707507-official-clevo-p570wm-p570wm3-sager-np9570-owners-lounge-21.html#post9507900

 

NVIDIA GeForce GTX 780M video card benchmark result - Intel Core i7-2920XM Processor,Alienware M18xR1
This is a Firestrike Extreme

anyway here is the pic which follows the criteria

Spoiler

i went ahead a did a stress test on the CPU while taking the pic so the summary in CPU-Z would say 4500mhz cause some people find it hard to believe to get a 2920XM so high, i am proud

if there is anything wrong with this post please inform me and correct me so next time i may not make the same mistakes again

 

Nice benchmark! Great job! You should be very proud. I'm proud of you, kiddo. :thumbsup:

Look at the opening post. Read the information under "Leader Board Requirements" and you will see that your screen shot does not meet the criteria for inclusion in the leaderboard competition. We go by the same rules as HWBOT.org for screen shots, links and tweaks. Futuremark scores with "illegal" cheats (AMD users disabling tessellation and NVIDIA users using PPU) are also excluded, but we don't enforce WHQL driver validation.

*wipes drool from chin* SWEET... I mean, WOOT ! Anyone that loves benching is welcome in this thread. The leaderboard is for Alienware only, but we like seeing your results, too. So, don't be a stranger. I have posted in Clevo benchmark threads as well. When it comes to benchmarks " red and yellow, black and white, they are precious in our sight..."

 

NVIDIA GeForce GTX 780M video card benchmark result - Intel Core i7-4930K,Notebook P570WM

Almost at 6.5k myself with 1170/7000

 

damn son

that doesnt fit your image LOL