Jump to content

OFFICIAL: M18x Benchmarking Thread


StamatisX

Recommended Posts

If I remember correctly XTU 2.1 was the only version that worked correctly with Sandy XM CPUs in Alienware also.

Yup that is the one I have and use. Mr.Fox having I believe ran benchmarks on each M18 what do you think of each of the Extreme Cpus and their over clock potential among the M18 revisions? From Sandy to Haswell... The fabrication size changes effected things but along the way improvements in chipset improved along the way. The ability to use unlocked bios with access to power limits seems to be a factor as well. I believe you have yours to essentially 5ghz. Congrats Mr.Fox very cool. Impressive numbers posted.

Link to comment
Share on other sites

@mw86 - Sandy and Ivy are both outstanding. Haswell, not so much. It runs insanely hot and the inconsistency among processors is terrible. The 4930MX ES that I have is better than most, but many have difficulty getting the MX or 4770K CPU stable above 4.2-4.3GHz. I would discourage anyone looking to buy a new laptop or build a new desktop to avoid Haswell unless they have unlimited funds to keep buying CPUs until they eventually get their hands on a good one. Hopefully, the next generation Intel CPUs will be as consistent and amazing as Sandy and Ivy have been. Moving the voltage regulators back to the mainboard may help.

By the way, we missed you. Nice to have you back with us.

Link to comment
Share on other sites

@mw86 - Sandy and Ivy are both outstanding. Haswell, not so much. It runs insanely hot and the inconsistency among processors is terrible. The 4930MX ES that I have is better than most, but many have difficulty getting the MX or 4770K CPU stable above 4.2-4.3GHz. I would discourage anyone looking to buy a new laptop or build a new desktop to avoid Haswell unless they have unlimited funds to keep buying CPUs until they eventually get their hands on a good one. Hopefully, the next generation Intel CPUs will be as consistent and amazing as Sandy and Ivy have been. Moving the voltage regulators back to the mainboard may help.

By the way, we missed you. Nice to have you back with us.

Well thanks Mr.Fox that has been very informative, much appreciated. I hope the next gen is better as well. Sandy and Ivy has been very amazing.

By the way missed ALL of you guys as well. Cheers my Tech Inferno friends. Glad to be back.

I still have my 4Gb 2133 sodimms of Samsung I flashed with Thaiphoon Burner. The timings were pretty good looking at some of the recent threads. One XMP profile I had at 2133mhz 9-10-10-25 Cr1 timing dual channel at 8gb total. Later on after reading your posts some of the good performance even being such an old chip seems to come from the fact it is 8 banks?... Each side? Looking forward to getting my rig back up and running. Still have 7970m but I want to give a stab at the Physics on Firestrike hehe always loved the cpu side of the benches. Your 12500+ physics was amazing maybe you got higher was it 13k physics on Firestrike?

  • Thumbs Up 1
Link to comment
Share on other sites

Well thanks Mr.Fox that has been very informative, much appreciated. I hope the next gen is better as well. Sandy and Ivy has been very amazing.

By the way missed ALL of you guys as well. Cheers my Tech Inferno friends. Glad to be back.

I still have my 4Gb 2133 sodimms of Samsung I flashed with Thaiphoon Burner. The timings were pretty good looking at some of the recent threads. One XMP profile I had at 2133mhz 9-10-10-25 Cr1 timing dual channel at 8gb total. Later on after reading your posts some of the good performance even being such an old chip seems to come from the fact it is 8 banks?... Each side? Looking forward to getting my rig back up and running. Still have 7970m but I want to give a stab at the Physics on Firestrike hehe always loved the cpu side of the benches. Your 12500+ physics was amazing maybe you got higher was it 13k physics on Firestrike?

You have the magical M471B5273DH0 memory. Is yours stable at 9-10-10-25 CR1, and if so what voltage and date of manufacture? My 2 2GB versions of the same stick can do 10-10-10 fully stable. One of my 4GB can do 10-11-11 fully stable, and the other 4GB sucks and can't do 2133 at any timing. I run the good 4GB and the 2 2GB sticks. They can bench at 9-10-10 but don't last long in stability tests. This is at 1.5V and the 2GB sticks are from 2011, and the 4GB from 2012. It seems that the 2011 might be the best sticks.

  • Thumbs Up 1
Link to comment
Share on other sites

You have the magical M471B5273DH0 memory. Is yours stable at 9-10-10-25 CR1, and if so what voltage and date of manufacture? My 2 2GB versions of the same stick can do 10-10-10 fully stable. One of my 4GB can do 10-11-11 fully stable, and the other 4GB sucks and can't do 2133 at any timing. I run the good 4GB and the 2 2GB sticks. They can bench at 9-10-10 but don't last long in stability tests. This is at 1.5V and the 2GB sticks are from 2011, and the 4GB from 2012. It seems that the 2011 might be the best sticks.

Truly I should retest them when I get my rig all set. I think they were more my bench timings they passed for many hours on mem test back then on 1.55v and we're stable on 1.5v all day at 10-10-10-27. You do some nice work as well by the way :D they may have degraded for all I know but I'll get a fresh bench up that's for sure. Have sometime to wait again and going to get the triple pipe heat sink from the R2.

Link to comment
Share on other sites

  • 2 weeks later...

That's interesting Mr Fox, I watched the video to compare frame rates with you. Of course you stomped all over mine in all the tests, all apart from the 'Combined' test. You were getting in the region of 80fps there. When I run that test I get 55-65fps I'm surprised you didn't get higher than your 80 fps in that particular test - what do you reckon is the cause of that?

Here's my scores for the sky diver run, for a laugh!

NVIDIA GeForce GTX 670MX video card benchmark result - Intel Core i7-2630QM Processor,Alienware M17xR3

  • Thumbs Up 1
Link to comment
Share on other sites

The video is captured at a low GPU clock speed and only 4.3GHz on CPU. Also notice low TDP and low GPU utilization (mostly 50% or less) in the combined test. The main problem is the benchmark is a sloppy piece of software.

Link to comment
Share on other sites

The video is captured at a low GPU clock speed and only 4.3GHz on CPU. Also notice low TDP and low GPU utilization (mostly 50% or less) in the combined test. The main problem is the benchmark is a sloppy piece of software.

Ah, there lies the comparitively low score in the combined test then, praps the CPU is bottlenecking the GPU's in that particular test.

  • Thumbs Up 1
Link to comment
Share on other sites

  • 1 month later...
  • 2 weeks later...
Nice run Deadbydawn awesomeness. How am I doing guys? Intel XTU bench submitted to our HWbot. mw86`s XTU score: 1070 marks with a Core i7 4940MX

Alienware 18

4940mx @ 4.3ghz

3x ssd raid 0

2133mhz ddr3L 16gb cl11-11-11-34 2T (dual channel 2x8gb)

Liquid Pro on the cpu and gpu cores

I just noticed your scores yesterday on the bot. Nice going.

  • Thumbs Up 1
Link to comment
Share on other sites

I just noticed your scores yesterday on the bot. Nice going.

Thanks a lot Johnksss good to be back on the benching scene :D I think I can get this a little higher then Ill try a wprime run I think. Spoke to Mr Fox recent, he helped me get the basics set up for the Alienware 18. Wow Haswell really is a blast furnace.

Link to comment
Share on other sites

It's nice to have you back.

And yes, haswell is a little mini nova waiting to happen. :D

sure is wow, I don't know how these scores compare to others but I got it a bit higher now on 2188mhz for the ram with a baseclock overclock.

http://hwbot.org/submission/2638270_mw86_xtu_core_i7_4940mx_1081_marks

lol i saw your XTU on 5960x, wow you got a beast setup...

Link to comment
Share on other sites

NVIDIA GeForce GTX 675M video card benchmark result - Intel Core i7-3920XM Processor Extreme Edition,Alienware M18xR2

within the results of my current config (3920xm, 675 sli) there were only two results ahead of mine, ones that are non accepted results since a gpu was dedicated to physixs. with my last result i beat one of those two, making me 2nd place behind a non accepted one and 1st place of the accepted, counting results :) cool, still can't wait until i got the money to upgrade the two outdated gpus :)

Link to comment
Share on other sites

I got a new alienware 17 with 4900mq (=m17x R5) a week ago just to see how good it is and how bad its thermal problem is.

BTW my main is m17x R4.

I OCed for 2days and these are my score right now.

-cinebench R15 850

-cinebench R11.5 9.17

-3dmark11 physics score (CPU score) 11728 (35.3 fps)

-xtu 1077

Early thoughts about haswell and new alienware 17 are it is very hot compared to her older IVY sister, m17x R4.

However, with fans kept to run at max RPM, the heat is somewhat controllable.

The better thing about her compared to m17x R4, is that she is "more" unlocked in regard of MSR.

In m17x R4, MSR lock can not be unlocked even with svl7's godly unlocked bios, but alienware 17(=m17x R5) can be unlocked partially with stock bios.

I am keeping saying "partial" or "more" unlocked because the clock of CPU can not be kept at max OCed clock more than 2 min in XTU stress test.

No it is not thermal or TDP throttle.

The only two ways to keep the maximum OCed clock (i.e 4.2-4.2-4.2-4.2 for 4900mq) is to lower cache ratio to x28, which, in my thought, is very silly.

Cache ratio x28 kills performance.

Another way is to lower OC clock to 3.9-3.9-3.9-3.9.

Except for those two ways, whatever you do, the OC clock will not be kept for more than 1 or 2 minutes in XTU stress test.

It will fluctuate between 4.2 and 2.8.

I have purchased 4940mx via Ebay and I hope this can help some of this "partial" locked MSR of alienware 17.

  • Thumbs Up 1
Link to comment
Share on other sites

3DMark - Fire Strike score: 10205 marks with SLI GeForce GTX 880M Alienware 18

screenshot

post-34-14494998512769_thumb.jpg

http://img.hwbot.org/u37675/image_id_1265393.jpeg

Tech Inferno HWBOT submission

mw86`s 3DMark - Fire Strike score: 10205 marks with a GeForce GTX 880M

verification URL

NVIDIA GeForce GTX 880M video card benchmark result - Intel® Core i7-4940MX CPU @ 3.10GHz,Alienware 01W2J2

Alienware 18

17 4940mx @ 4.3ghz

Hyperx Impact 16gb dual channel 1.35v CL11-11-11-34-2T

4x80gb SSD Raid 0

SLI GTX880m @ 1100/1625mhz @ 1.05V

Standard Air Cooling just a raise of the laptop with HWiNFO controlling fans to MAX

No AC

Average Room Temp

Connected to UPS max draw during run 414Watts (using 3dmark Firestrike free version - drew the most power on the demo scene and a bit into gpu test 1) on a Cyberpower1500PFCLCD

Screen set to minimum brightness

Lighting set to Dimmed

Go Dark set except for Status Zones

Webcam disabled in Device Manager

Touchpad disabled on external mouse detection

Free Fall Sensor Disabled in Device Manager

will disable Internal Ethernet via bios for future

Cpu only drew 71watts @4.3ghz @ 69C according to HWiNFO

GPU 1 77c max used 87.9% TDP

GPU 2 83C max used 86% TDP

post-34-14494998513515_thumb.jpg

Hey guys anyone know how this setup comparing to others?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.