so yours crashed too when at 670? 660 is stable for me though or was since i am now on win 7 and have to do it again to make sure
-
-
hey guys i was doing comparision between gtx 280m sli and desktop evga gtx 285 ssc edition, to my surprise gtx 285 http://www.youtube.com/watch?v=L0BlCJF0Ke8&feature=fvst only managed 27 fps on very high at 1920*1080 with better cpu where as mandrake benching of m17x shows gtx 280m can pull average of 29 fps. What i don't understand is on the vantage gtx 285 gets 12805 gpu score so in theory gtx 285 should be getting higher fps rite?
-
dondadah88 Notebook Nobel Laureate
I have to remember to do a guide for the ati tweak to squeeze out the max performance out of there card. with that said maybe we can see the 4870 kill farcry2. and stack it against the 280m in vantage.
-
-
Is AA really unnecessary when running at 1920 x 1200?
-
ViciousXUSMC Master Viking NBR Reviewer
Older games need AA more than newer games in general due to the lower poly count and low res mesh's. -
I was worried more about newer games, since the 1200p would put more stress on the laptop (and as I've been told, down scaling LCD's isn't exactly smart).
But running older games with 1200p & AA shouldn't be a problem..I think -
ViciousXUSMC Master Viking NBR Reviewer
Well I ran that X3 benchmark, the results are supprising.
I figured I would top the posted m17x score and the W90 score but I am lower than both.
Using the default settings (includes 4x AA) I only got 68fps average with my GPU's OC and my cpu @ 2.7ghz.
I noticed the score for the M17X is @ 3.2ghz on the cpu.
So my only guess for these results right now is that the higher W90 score was also with a faster cpu. This looks like it could be another of those games that is very cpu heavy.
edit:
Looks like this game does not use quad core and that its cpu speed dependent for its score, that makes it not a very good benchmark. -
dondadah88 Notebook Nobel Laureate
mmm are you sure about that though. i'm going to check my score with the 3870's.
-
ViciousXUSMC Master Viking NBR Reviewer
More testing cant hurt but thats what the evidence points at, I mean why would my W90 score less than somebody else s with a higher gpu overclock?
The difference was his cpu, and his score was in the same range as the M17X with a cpu in the same range.
I can easily turn down my cpu oc some and see if the score gets even lower to confirm it.
Also the fact your older slower cards got identical scores basically to the newer faster cards shows that the gpu is not what is being pushed here, its the cpu. Just like how 3dmark06 acts now.
A SLI 260 and SLI 280 will get the same 3dmark06 score with the same CPU speed. I bet this benchmark will show identical or similar results. -
dondadah88 Notebook Nobel Laureate
that was a heavily tweaked driver and hard optimzations i did. plus don't forget aa was off.
-
ViciousXUSMC Master Viking NBR Reviewer
I'll run it now again @ 2.5ghz isntead of 2.7
If I drop from 67.x FPS then I am pretty sure I can say its a cpu thing.
I'll keep the gpu clocks the same. -
Im sure x3 isnt multithread, i got killed by the duo core machine from MSI as opposed to me using there quad varient (2 ghz as opposed to 3 showed a huge difference
)
The guy was also using full hd res, it still killed me lol. -
ViciousXUSMC Master Viking NBR Reviewer
61.2 FPS @ 2.5ghz vs 67.3 FPS @ 2.7ghz final results.
So that means 200mhz of cpu powerr = 6fps
So I can assume if I had 3.2ghz I would have 15fps more over 67FPS if it scales linerly.
That would put me at 82fps.
The W90 owner that ran stock GPU with a 3.22ghz Dual Core got 78fps.
Consider the few parts of the bench that were not cpu bound due to the heavy gpu processing and that leaves you with only 4fps difference, so seems to fit into place VERY well, and I think we can all conclude that my reasoning and research are correct and that this benchmark is cpu bound and thus a waist of time. -
Would overclocking be a nono if you don't plan to get an external cooler? or alienware's design is proving to be quite resilient to overheating?
-
-
I would say the design is proving more then reslient to the heat, but as tesco would say as opposed to the cooler "Ever Little Helps"
-
6/29/2009 4:30:12 AM - Vista 64
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX10 1920x1080, AA=No AA, Vsync=Disabled, 64 bit test, FullScreen
Demo Loops=1, Time Of Day= 9
Global Game Quality: VeryHigh
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 64.86s, Average FPS: 30.84
Min FPS: 20.96 at frame 1050, Max FPS: 37.83 at frame 994
Average Tri/Sec: -7869602, Tri/Frame: -255203
Recorded/Played Tris ratio: -3.59
TimeDemo Play Ended, (1 Runs Performed)
==============================================================
Completed All Tests
<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>
6/29/2009 4:30:12 AM - Vista 64
Run #1- DX10 1920x1080 AA=No AA, 64 bit test, Quality: VeryHigh ~~ Last Average FPS: 30.84
Edit: Core clock on the GPU was set to 670. -
-
dondadah88 Notebook Nobel Laureate
I ran the benchmark again with the 4x and everything the same when you start it and I got 56fps average. I overclock then it froze so I hae to try lower clocks and see what's up.
-
-
-
No problem, I'm having fun doing them. Still a lot of tweaking to do. 12K Vantage GPU score is around the corner.
-
-
3dmark06 - 16051
Vantage - 11774 -
-
If you want 12k GPU in Vantage, 185.20..
-
-
185.20 adds about 200 more points than 182.46 for me, so it's definitely worth a try.
-
-
-
I know how it looks like since I have the same exact system. It may take you 10 minutes more to figure out where to click.
-
it was a no go on 185.20, distorted screen as with rest of 185 series, kept locking up when trying to run bench too back to 182.46Dox for me
-
try the new 186.21 ore the 186.03 (that are the best for me in 3dmark vantage)
-
I went through some old posts and saw that I had problems downloading the 185.20. I'll try it again.
Edit: Yeah it looks horrible. Trying to run Vantage but it will be impossible if I have to change any settings.
Edit2: Vantage hung during initial load. Going back to the 182.46 drivers. -
EDIT: going to try 186.18 again and run some more tests. -
http://forum.notebookreview.com/showthread.php?t=394775
new driver from dell -
Edit: They're not working for me....on to the next one. -
-
-
ill update my driver thread with it for yall
-
-
it is
it in there so yall don't have to jump through hoops to get it -
Don't forget to test with the 9400M enabled to really make sure they work right.
-
yea cant wait to see if these new drivers have hybrid enabled + push vantage to 12k!!!
-
-
Good news. Also a very good sign that Dell is releasing updated drivers.
-
very good news
+ rep kessciem
-
My gpu score dropped 1500 points. I'll try upping the clocks.
Edit: Not getting any better. Going to retry Dox to make sure I'm still close to 12K.
M17x Benchmarks thread !!!
Discussion in 'Alienware' started by The_Moo?, Jun 22, 2009.