did you read my post about why you cant get high numbers in 3dmark06?
-
not quite sure, point me to it. -
run only gpu test 1/2 & 5/6 using cores 2/3. then run the test again using all 4 cores. you will see what i mean.(quad core users only) that is why it's so hard for you guys to catch my 9262 score of 16,853 on 3dmakr06 <!-- / message --> <!-- sig -->
also, you will need to keep your laptop cards as cool as possible...so you may want to improvise something for cooling.heck, use a house fan to blow through those vents...the cooler a laptop is while benching, the more you can gain..as the cards heat up, they start to slow down.... -
Didn't Quad score 17K with a Quad on the W90?
-
yes he did, but his cards we're ati's and not nvidia's. it's some wired stuff going on with gtx/3dmark/quad core cpus.
-
-
i maxed out my 9262 when i had it..only thing left to test was a xeon chip@ 3.16 ghz and 95 tdp...which would give it extra room for sure to go higher.
lower watts and the extra 160 mhz -
dondadah88 Notebook Nobel Laureate
that's where I come into play but it won't be for a while lol. john what score do you think 9800m gtx sli and a dual core will get?
-
that i don't remember dadah... 2.0/3.0 scores would be higher than quads though. and i think cpu was in the mid 3's
-
dondadah88 Notebook Nobel Laureate
thanks i'll see how much I can get with my setup now. and do some math and see. but I wonder if it's the wa the data goes from the cpu to sli.(afr) I would have to test it out. but it works with ati that uses crossfire(afr.) I nd more data.
-
Are there any Vantage and 3DMARK06 scores for Q9000 with 260M SLI config?
-
P8700 w/260 SLI LINK
-
If you want t9600+260s in SLI vantage or 3dmark06 I have both.
-
Did you ever submit your scores Clauss? I don't remember seeing them.
-
I wrote them down int he thread somewhere, I remember my scores were a TAD higher. My 3dmark06 was like 14.7k, this one got like 14.47k. Granted that last CPU overclocked a bit better. For vantage...its pretty low, but for some reason it wont let me bring them up again. I bookmarked it and now it just shows a sliver of the page. Odd. But it was like..8.9k GPU score.
-
Very interesting info John , just another reason to not put any stock in 3d06 as a benching tool. -
tis is true...me and ewrecked explored this with the 9262 and the whitebook. the white book had a very very small gain if any while the 9262 gained allot. it would seem that ati could run smooth with dual and quad, while 9800's started having bandwidth issues (just how im gonna some it up)
example.. 9800m gtx
running test 1/2 the max score was 600 if you we're really lucky 6100
run it in dual core mode... 6650 max
3/4 in quad mode.... 7400 or so...in dual core...8000
with 1/2 5/6 in dual core mode and cpu in quad core mode.
dammn, link is having issues at the moment.
and this one ran with all 4 cores enabled
http://service.futuremark.com/compare?3dm06=9365429 -
I really wish futuremark would release a free version for everyone(i have my own key but not everyone does and are willing to buy one either) that can be used over and over
so when people get there system the first thing they dont do is install 06 lol.
Till this happens im afraid 06 is going to remain king of the benchmark apps for many years to come sadly. -
-
That sounds like a good idea but it just stinks because with newer drivers or someone who didnt have there system set up properly in the first place(like im seeing so many M17x owners with issues from Hybrid SLI) then the scores will be all over the place
like the guy who got the 69 score on the GPU lol.
Still sounds like a good idea to have that thread and if for nothing else people that actually have Keys and want to participate . Just afraid its going to be like 5 people who actually participate in the thread lol. -
That's how many we have participating now. Just got gotta hit up the Asus and Clevo crowd to come and participate. If M17x owners want help with their scores then there is always this thread. Then again there's always the gaming forum and this could always go there instead.
-
Probably better off in this forum , the gaming thread forum it would bumped to page 200 after a day or so lol.
We need more Sticky threads for benchmarking and a couple things that come up every hour around here that people wont search for and make new threads about lol. Windows 7 questions for example. -
So sticky a CPU OC'ing thread, a GPU OC'ing thread and a Vantage Benchmark thread?
-
Yes exactly Drake . Those threads are perfect examples of things we need around here that would help a ton of people and prevent Multiple threads popping up every hour.
-
to many stickys are bad to ....
-
oh and don't forget Windows 7. How many is considered too many?
-
Well it doesnt need to be all of them and not to mention im sure some of the older stickys can now be gotten rid to make new ones.
-
I could always post my how to replace your processor guide but it just takes a bunch of time to format it correctly.
-
What you really need for that kind of benchmarking info is a site like Notebookcheck.com that doesn't suck. -
We just need another moderator around here to keep some of the useless drivel out of threads(were all guilty of this to an extent) , so if a Benchmark thread pops up it does not turn into a discussion about everything but benchmarks.
-
If people learned how to use search we would be golden -
Yes exactly like right now but because this topic turned up in the Benchmark thread there is little choice but to carry it on in here .
As for using search ya x10 million. -
as for another moderator ... most forums only have 1
Chaz likes it that way -
what are the benchmarks you can run without the game?
-
You mean like Crysis , Devil May Cry 4 , Streetfighter 4 , Unreal Tournament 3 ? Alot of other demos or games people test dont have built in benchmarks but they used a fixed portion of the game to test average framerates .
-
DMC 4 asked for a game directory .... so i wasn't sure .... man i need DSL
-
If you guys want... I can code a simply system to ppl upload benchmarks with rank by benchmark, etc and host it
-
similar to futuremark? that would be cool.
-
MrButterBiscuits ~Veritas Y Aequitas~
NBR's own hall of fame? for laptop! That's Fantasmagoric!
-
I think it would be a good way to draw people. The few benchmark threads that have popped up haven't had enough interest.
-
Ok... Give me 2 hours
-
I just tried running GPU-Z 0.3.4 on my M17x.
But this versions doesn't show all info. I guess it doesn't support the GTX 260M at the moment.
It only shows the PCI-E 2.0 x16@x8. Shouldn't this be: x16@x16?
It doesn't show Texture Fillrate, Bus Width or Bandwith though.
Is there a more reliable program around at the moment? -
But if anything, the GDDR5 version of the 4870 would yield maybe about 10% on top of the current GDDR3 versions and possibly more factoring in the extra VRAM as well. But I don't expect it to be the one announced @ CES which got a lot of people, including me, very excited as they mentioned the specs of it: a 680mhz core speed and the memory running at 700mhz (GDDR5). Now, the figure has been reduced to 550mhz core speed and 888mhz for memory; still fast, but nowhere near as big as having the core operate at 680mhz. -
550 --> 680 isn't too unreasonable...
I think mobile 4870's w/GDDR5 will give more than a 10% boost in performance. Only time will tell! -
As for the GDDR5 mobile cards, the performance would be around 10% from what I've seen. People have tested overvolted desktop 4850s (so that they can overclock the core to 4870 levels comfortably) and the desktop 4870s they were compared to were around 10% faster on average. -
dondadah88 Notebook Nobel Laureate
well i would like to say some things on this matter.
scook9 has the whitebook and he modded his notebook that's in project freeze. his clocks for his 3870 are over 750/1000 and it's very stable he is also very stable because he is running in 4 place for 3dmark06 and also in vantage.
meaning that the memory heats up quite a bit. i was barely stable over 900. so i think that if the cooling is right on the memory, you may not need a voltage change. (i could be wrong of course.)
ichime i remember the convo that we has about the 10% increase over gddr5 vs gddr3 and i believe it was the same with gddr2 vs gddr3 but the gddr3 excelled in clocks speeds that made the difference. i believe the 4870 will do the same with gddr5 making the clocks higher then gddr3. .
the max i can get with my 8600m gs with gddr2 was 515 with out it going unstable.
my 9800m gtx can hit 9??(i haven't pushed it yet) and that's no where near my old card
mandrakes gddr3 can hit 1,100 on the gddr3(alittle improved memory chips)
nvidia also claims there gddr5 will hit 2000 on the memory on there website but we have to wait for there release
the same with ati. -
cookinwitdiesel Retired Bencher
I was stable at 745/1055 ultimately - still nothing to scoff at. I did have excellent temperatures (low that is) due to the use of a Cryo LX with holes cut in bottom and use of OCZ Freeze. As you will see if you read Project Freeze, I had much more planned but have decided that I am ready to move on to an even more powerful notebook than I need
(the beautiful m17x)That being said, if I had carried out the modifications I had planned, I am pretty sure that even better temperatures would be reported leading to better overclock - mainly for the VRAM.
Donda loves my scores hehe -
Well hopefully we can make some real headway with overclocking the GPUs on the 280s, cant wait for monday or tuesday for my machine to arrive lol.
-
dondadah88 Notebook Nobel Laureate
i won't have my sli by then.
-
Well Dahdah you said you had other things you had to take care of first and thats understandable and its not like anyone is going any where . So when you get your second card your going to be one very happy camper for sure. Then the CPU after that. You have alot to look forward to and when its all said and done it will be like having a whole new machine really.
M17x Benchmarks thread !!!
Discussion in 'Alienware' started by The_Moo?, Jun 22, 2009.