The GTX 280M Overclocking and Benchmarking Results Thread

Well, I don't think 660 core was stable, but right now 640/1600/1000 is running fine. CCC level @1440x900 and 2x AA...mimicing real life

On the Zalman peak is 73C

 

I'm happy with the results

 

yeah, that's one of them ichime. it's the same one falsalhero & i got like 30 fps with, remember...but when we ran the other tool, both scores we're at 19 & 20 fps.

 

11 frames to paradise?

 

now run it using native resolution/gamer
your direct competition 4850m

edit: nevermind..i see it. you ran it already.



a sad day for the gtx280m...loosing by 5 frames.....

 

Hot Dang ...

am i seeing it right ??? !!!

... that the 4850 actually is slightly a better performer over the 280M ???

is this correct ???

according to all the factual data that we are seeing here ??? !!! ???




A Falling Cat named Spangky

9 Lives only

 

You notice you are also using different OS's...

And why not use a more up-to-date and accurate benchmark like 3DMark Vantage.

umm.... why not do a benchmark on a game that actually utilizes a systems full potential....
- Unreal Tournament 3
- Call of Duty 4
- F.E.A.R. 2
- FarCry 2
- GTA IV (really show what the system is made of)
- etc....

 

does he even own any of them games?

 

Hot Dang ...

If I didnt see it with my own eyes ... I would not have believe it ...

Them 4850 are actually a better performer over the 280M ...

half the V-memory

slower cpu

even on a 32 bit os ...

still comes out better ...

I am coming mighty close to pulling my trigger on one of them msi setup ...




A Falling Cat named Spangky

9 Lives only

 

I recommend you to stop making conclusions and assumptions.... especially with so few reviews, tests and details from reviews.

Like I said, just testing one game (especially one that is infamously poorly coded) and taking that to make assumptions is jumping the gun.

Its nice that you are enthusiastic about gaming notebooks, but do not misconstrue the information until there are more details are out.

From my sources that have done tests... as to my high standards...
- the GTX 280M is slightly better performing that the ATI 4870 in overall benchmarks and gaming tests (most important factor).... stock.

we will see more detailed benchmarks and tests as more systems are being reviewed.

 

yeah, something definitely aint right. because im running that same test right now and im getting 20 fps with a stock 9800m gtx@500/799/1250. will post screener when it's done running.

 

The card can't even use decent drivers yet. Slow down peeps.

 

im going to show you how it is suppose to run on old drivers with no over clock. (using the 9800m gtx)

 

Hardly a sad day. GTX 280M can OC its memory to 1000mhz and getting some pretty nice overclocks over the 9800M GTX in the same chassis.

bah, make your decision already. But I'll tell you this; basing your decision on ONE benchmark wouldn't do you much good honestly.

Actually, Crysis Warhead is much better optimized than regular Crysis. As for the GTX 280M being slightly faster than the 4870, it's possible, hopefully with more tests, though so far a single 4870 (from the Asus W90) seems to be winning the battle so far against the single GTX 280M (even with 950mhz memory clocks)...

 

not that much better from my tests.

I would rather use other games that are fully multi-threaded and fully utilizing single and multi videocards.

 

Do share

like?

 

Hardly a sad day. GTX 280M can OC its memory to 1000mhz and getting some pretty nice overclocks over the 9800M GTX in the same chassis.nice over clocks. <-- gonna have to disagree

1: the stock clocks are suppose to be 583/950/1463
2: the stock clocks on a 9800m gtx are 500/800/1250
3: we see so far an over clock of 660/1600/980 stable... going for 675/1600/1000 <-- per his words.
4: 9800m gtx over clocked cards 650/970/1600 max over clock at 700/1000/1700
and here is your proof.
http://service.futuremark.com/resultComparison.action?compareResultId=9427549&compareResultType=14
5: it's starting range is 950 memory and not 800. that's for the 2.1. 3.0 is suppose to start at 950 and up. that's like me starting from 400 core and then going to 650. (no nvidia!)

sorry my friend but there is no comparison. a 9800m is out over clocking the 280m by allot. as of today only..who knows what tomorrow will bring.

side note... 4870 ain't beating me just yet..

 

well I dont have spare change to spare ... so I think for the price ...

I will settle with the gt725 ...



and more importantly not to forget ... I would like to thank everyone in the forum for all their help and information and expertise ...

Its been a swell time ...



cheers everyone ... enjoy your stay ...



Warmest regards to everyone on the forum as always

A Falling Cat named Spangky

9 Lives only ...

 

I listed them over and over. LOL.

Many Unreal Engine games are great for full utilization tests.

As well as....
- Fallout 3 (for a RPG)
- GTAIV (for a new system taxer)
- FEAR 2 and FarCry2 (for good measure)
- Left 4 Dead (another multi-threaded game)
- Stormrise (new and heavy system taxer)

 

dammn...i haven't ran any of these games yet...i see im falling down on my benching...
edit:except for gta4

 

Didn't yallnotice I was running orignal crysis and not warhead? anyway, way too emm drunk and need sleep, ill post warhead @ current OC later, I guarantee I'll beat any single GPU out there

oh and max overlcock atm is 650 1600 1025 again way out of touch here ill talk about it later.....

 

think you better look again...it says warhead right at the top of the screen



technically speaking....you should have already did that
oh and max overlcock atm is 650 1600 1025 again way out of touch here ill talk about it later..... <-- i see you just added this.. your gonna need to be about 100 mhz higher than that on core and memory
not saying it's your fault, just saying nvidia pulled the oki doke move it seems....time will tell though. i sure hope that's not the case. i really wanted to upgrade

 

I thought you disliked the warhead benches since the tool is buggy?

I had 2 benches done at 650/1650/1025 for regular crysis and got 32.5 DX10 high, 19.3 Very high.

 

your miss quoting me.
i said the benchmark tool versions 0.29-0.32. i never said i didn't like benchmarking warhead.

 

-edit sorry link didnt work-
2nd try:
http://service.futuremark.com/compare?3dm06=10698118

11522 3DMarks

615/926/1530

 

[uote=Athinu;4756766] http://service.futuremark.com/home.action;jsessionid=48B4A114C2339C5874CDBD374EB836AC <--this isn't showing.

11522 3DMarks

615/926/1530[/quote]

 

lol sorry should work now

 

yep, working now.

 

can someone throw in DMC4 since it fully stresses out you gpu and doesnt matter the cpu what so ever

also throw in L4D. there is no built in benchmark but use fraps for 5-10minutes to benchmark and see what you get.

 

Haha, look at you, all quick to respond and proud to show off your OC'd 9800M GTX But, you missed a key distinction I made:

What's the highest overclock of a 9800M GTX in the M570 chassis? Pretty impressive considering the lack of $130 worth of additional cooling you seem to have.

Well, sure. a 25%+ overclock over stock clocks is quite impressive. But, like you said earlier, you don't even play games when overclocked, just benchmarks. Why is that?

Some I agree with, even though they're unpopular and no one would even care about those games anymore (i.e. Games that use the Unreal Engine), but GTA IV, though it multithreads still has issues with multi-GPU support. Same with Left4dead and Fallout 3, where mutli-GPU support kicks in at ridiculously high settings (i.e. 2560x1600 with everything maxxed).

Stormrise isn't even out yet and if anything, the ATi cards might end up winning this by a lot because it's going to be a DX10.1 game.

 

Wouldn't happen to know what version of the 4870 those tests are being run with? The prototype GDDR5 4870 or the ASUS W90 GDDR3 4850+?

 

Here is my 9800m GS

 

I'm getting higher scores than that with my 8800M GTX @ 620/1000/1550 (Look in sig)

 

well... that's interesting

Well you've got 500 mhz x2 on him, CPU has a big part in 06 scores these days. If you're going to compare horsepower, buy vantage...

 

yeah tavara is hard to beat in benchmarking with only a 9800m gs. he killed my 3870 single card. i don't want to see what happens if he has the w90 or my computer in his hands.

but try different drivers. 180.80 is the number i think everyone uses. the 280m gtx is suppose to kill the 4870 crossfire claims nvidia.

 

[uote=ichime;4756983]Haha, look at you, all quick to respond and proud to show off your OC'd 9800M GTX But, you missed a key distinction I made:

ummm, no. i think you better look again. that's not my over clock.
pretty quick to judge aint cha? i told you many times before. it's not how high you over clock, but what you get out of your over clock. we're we not listeing?

no! your missing the main thing to all this. his starting clocks and what nvidia said they start at. (and the max they said as well) im trying to understand why he can't over clock at least the same amount as the 9800 series. unless of course the 280m is nearing it's max point already. this card should be the new over clocking flag ship, not the play catch up one.

What's the highest overclock of a 9800M GTX in the M570 chassis? Pretty impressive considering the lack of $130 worth of additional cooling you seem to have.

that over clock doesn't beat out emike with a lessor machine so i still fail to see the logic. and he pointed that out already


Well, sure. a 25%+ overclock over stock clocks is quite impressive. But, like you said earlier, you don't even play games when overclocked, just benchmarks. Why is that?

im here to prove to all the haters why the 9800m gtx is the better all around card. i thought that was pretty clear since day one? till one comes better.(should be/suppose to be the gtx280m) 9800m gtx plays fine with out over clocking..can you say the same thing about yours? i lose 5 fps in crysis warhead, how many do you lose? i over clock in 3dmark..the only thing that moves is test 5. i over clock in co4 which shows about 20 to 30 fps, but if im at 120 fps already..does it really matter? if the game needs to be over clocked to play..sure im all for over clocking to play, but im not the gamer type..can't be sitting around all day playin games. just here to make sure the user gets the full benifits out of his or her machine, instead of waiting for the manufacture to get around to it.

and actually, these are usually your rebutles...why the change of heart for nvidia cards?
edit: to make it even simpler..justify why i spent the money to get these cards and whether it was money well spent or wasted. looks like it was well spent. even if it cost more at the time.

 

Im a benchmark/overclock noob only been playing around with the gfx card in the control panel my last score is 11826 3dmarks.
Someone who know what theyre actually doing should easily get 12k+
My system runs stable how it is now and the temps are quite low too.
I'd love to see some overclock tips/tricks as I am atm a noob at this lol.

 

Summary of this thread:

GTX 280m Sucks. (Its not an improvement much at all)

 

now isn't this interesting...a 9800m gs with 30 plus frames...
and we still say this is the benchmark tool to use?

 

don't jump to a summary yet. it is still young. the asus was garbage until we all put in efforts to get it were it is now(stricker, johnksss, quadzilla, viscous, and many more). the whitebook was crap until we all pitched it(e-wrecked, johnksss, ichime, zfactor, mark, and a whole lot more).

so we have to wait alittle to see what is going on and what we can do to get it going.

 

we trying to get to the bottom of it now...just need an enthusiast willing to embark on that path...with out them in every notebook class..you can't get any where fast....

 

I dont't see what is wrong. 9800m GS is pretty good GPU when OC it!
Look my athers benches! between 9800m GS and ather's 9800m card in real gaming they perform similar

 

tavara

only speaking on one benchmarking tool and not your other accomplishments... great job by the way!

try this benchmark tool for crysis warhead then post back here...thanks.
version 1.5 for crysis
http://www.hocbench.com/
version 1.2 for warhead
http://www.hocbench.com/crysiswh.php

 

Funny thing is, out of all that typing, you still missed my point. For instance, on the issue regarding the 9800M GTX vs GTX 280M overclocks, You say that the linked benchmark run is not that of your system. Fine. But it's not a M570 chassis either (it's a D90xc). You have yet to show me a 9800M GTX in a M570xU chassis to overclock that high, especially in memory clocks and lived long enough to run a game. And where does the whole diminishing return concept come from where so far, we haven't seen any issue of decreased performance with higher overclocks with the GTX 280M yet? Nor has anyone brought it up, but you. So you can type on how I'm quick to judge and don't listen, blah blah blah when what you brought up has NOTHING to do with the point I was making.

Regarding the emike comparison, if you are referring to his 3dmark06 score (which I think you are), you're gonna have to look at that score again (not that it matters much since the whole Whitebook/3870 debacle)

Had you not missed my point, you'd see why. (Hint: one chassis has a better cooling system and more power available than the other chassis).

1) There are no "haters" of the 9800M GTX card. The only thing people "hate" about these cards is (was?) the price.

2) So you're not the gamer type, yet you buy a gaming notebook...great...sure this isn't an e-peen thing? (no offense)

Thing is at stock form, the GTX 280M is the better all around card than the 9800M GTX period. Price, performance, regardless of MXM 2.x 3.0, etc. This card hasn't been out for even a full week, yet some are already making judgments as to how far it can overclock, and not even in a D90xC. How long did it take the 9800M GTX to reach its OC potential when it first came out? If you're gonna make quick judgments or if you wanna see what a better cooling system on these cards would do in terms of overclocking, at least wait till the M980NU becomes available to the public as it has a cooling system that rivals that of the D90xC.

 

wow, all that extra typing..thanks!
we are working on fixing the problem, not debating it..but thanks though

emike does not have a d901c! *facepalm* he has a
http://forum.notebookreview.com/showthread.php?t=339000
looks like a M570 to me and the rest of the world.
his clocks are just as high if not higher. go ask him. (talking about over clocking ratio)(nevermind i found it for you)
http://forum.notebookreview.com/showpost.php?p=4014229&postcount=110
Re: 9800M GTX Thread <hr style='color: rgb(238, 238, 238);' size='1'> <!-- / icon and title --> <!-- message --> Quote:

Originally Posted by itsfun Has you testet the overvolting in Nibitor. I overvolted my 7950 gtx in my Dell XPS M1710 with Nibitor and this works...

I have. Maximum voltage allowable is 1.03v.

Syster, it sounds like sombody was a little rough on their car...

P6504 ! ! !

CPU Score
24274
Graphics Score
5287

Clocks:
712 / 1678 / 977 @ 1.03v 42F ambient

Beat that! (non-sli / quad core people...)
<!-- / message --> <!-- sig --> __________________
-Mike

NP5796 | X9100 3.5GHz | 9800M GTX | 4GB DDR3 1066 CAS5 | 500GB 7200 RPM HD 16MB | X-Fi 7.1| Zalman NC-2000 | 4GB Intel Turbo Memory | Windows 7 Ultimate x64 7068 | Vantage: P6901 | 3DMark 06: 12537

My firearm is my friend, it never leaves my side.

Mojave Underground - Abandoned Mine Explorers of the American West


1) There are no "haters" of the 9800M GTX card. The only thing people "hate" about these cards is (was?) the price.

ummm, again no. you only need to go reread your many post about the performance of the 9800m gtx back then.

2) So you're not the gamer type, yet you buy a gaming notebook...great...sure this isn't an e-peen thing? (no offense)
none taken
definition of a gamer: someone who plays games all day long. that i don't have time to. do i play games, yes. do i want to play high end games when i do want to play them, yes. do i have a system capable of that, why...yes i do.so your point is?
like i said before, my computer is a work station, it's use for more than what you would be accustom to using yours for.

people worry about epeenin when they are out of the loop, remember that.. the ones in the loop dont care. guess every game/sport/house you own/car you drive/school you goto/books you read/life you live. is based on epeenin as well. get cha epeenin on then

and your qx9300 you bought for gaming...quads help very few games....mr epeenin...

 

Thanks for finally showing me what I asked for. Though, the GTX 280M is getting higher memory clocks still. Had you shown me that before, then your point regarding diminishing performance with higher clocks would have been relevant and applied in this situation perfectly.

Hey, remember the last time you said I made certain statements and I ask you to find where I made such statements and you either make an excuse or ended up not finding them? It's no surprise as this is kinda happening again, but I'll help you out this time though:

This is me bashing the 9800M GTX because it performs similarly to the 9800M GT in 3dmark06

This is me bashing the 9800M GTX because it's 1gb advantage shows no performance gain over the 8800M GTX

And this is Johnksss appreciating the price/performance ratio of the 3870s against bunch of non-hardcore ATi fans at...oh wait, we're supposed to be talking about the GTX 280M aren't we?

Well that says a lot, but out of all that, you still don't get it. Gamers don't play videogames all day. Professional gamers, maybe, but "gamers" have lives. I'm a gamer, but I don't play games all day, mostly because I work. When I come back home, I play a games like CoD 4, a bit of Team Fortress 2, throw in some GTA IV, Savage 2 and Counter Strike once in a while, etc. If Starcraft II is as good as it's hyped up to be, then you probably won't see me on these forums or any forums very often

So to you, buying a ES QX9300 for $550 when it should be worth almost twice as much is e-peenin. To me, it's taking advantage of a good bargain to increase my gameplay experience via better performance (and also to increase resale value )

----------------------------------------------------

Aside from all this, I still stand by my assertion that the GTX 280M remains the better card over the 9800M GTX in terms of the overall package. It's just like how on the desktop side of things, 9800 GTX+ is a better card than the 9800 GTX. There is a benefit of having a 55nm die over a 65nm die and with enough patience we would see this benefit soon.

 

ummmmmmm, ok...if you say so ichime.. so im going to exert my epeening and look at my little green rep box's to the left of the screen.. guess im pretty helpful to quite a few people..

Aside from all this, I still stand by my assertion that the GTX 280M remains the better card over the 9800M GTX in terms of the overall package. It's just like how on the desktop side of things, 9800 GTX+ is a better card than the 9800 GTX. There is a benefit of having a 55nm die over a 65nm die and with enough patience we would see this benefit soon. <---- this we agree...i just like to see it with in the first 2 weeks of a purchase, just in case it needs to be returned for full purchase price for not meeting ones expectations.

back to the 280m and getting it where it should be.
Athinu

how goes it, you still working on that?

 

Be careful what you ask for

Vantage Score

Hah! It even shows my old rig since I used the same email. I got my money's worth, heh.



I'm actually impressed enough with the score that'll buy vantage and improve the score even more. Apparently John throws playability overclocks out the window and just goes for the highest number... I can play that game

 

anothergeek
ummm....no. but you'll soon understand over time... and had you been listening you should have been higher the other day.
all we're doing is pumping you up to finally get on the ball is all..

edited:


now back to the 280m
that's a right nice dammn score. this is what we have all been looking for. that 5870 gpu score. now to push that score in the 6k range.

 

Touchy? I'm messing with ya dude. All in good fun, we've all got the same interest it seems.