9262 Benchmarks with 9800 GTX

Ok, here is my first 3dMark score. This is done on a stock system.
The only change I made was to change power mode to performance and change NVidia settings to performance for all apps.

System: NP9262 w.4gb RAM, 9800 GTX 1gb, Vista x64 SP1, 1680x1050 display

Moving onto Crysis bench next. Question: I am going to run the built in batch files for the tests. Do I need to set the crysis settings before hand or will the batch file do this for me?

 

thats a good score. Not that much of an improvement but an improvement overall. It gets 1K more than standard 9800GT. Big kudos to you sir, waiting for more test

P.S.
u have the 9262 with the Q9650 so actually the 9800GTX in the coreduo systems will get even smaller gap between the 9800GT. I am actually a bit dissapointed. well we can look at 5% more gains from the 9800GTX for $400 more.

 

First of all, congrats on your new machine!

It will use the settings that you have the game currently set to. So yes, check and change settings before running the tool.

 

Unfortunately, with the Crytek benchmark you have to run the game first, select the settings you want, accept them, and then exit the game to run the benchmark. I know this is a pain, but considering all the confusion about the third party benchmark, it's probably worth doing.

 

Grats for youre machine first of al .
Yes seems like a 5% increase, i hope the crysis benchs will show higher % increase but i doubt since these scores are the 2nd i see from this card and are quite close.

 

3dmark scores are a dissapointment so far. i am anxiously waiting for the crysis benchmarks. crysis will tell us if the gtx is indeed a card worth it.

 

Congrats mate I'm jealous Looking forward to the Crysis benchs. Especially:

All settings HIGH at 1900x1200
All settings VERY HIGH at 1900x1200
All settings ULTRA HIGH at 1280x1024

This with DX10 and DX09 in Vista.

I would say that if we focus on this we will be able to compare. This rather then spread our attention into so many resolutions and possible qualiy settings.

Trance

 

Ultra high... uh?

 

he has a wsxga screen so no chance for 1920x1200

 

Ops just noticed that our LCD does not support 1900x1200. Well just the use the highest possible resolution.
He can get ULTRA HIGH in Vista with DX10.

Trance

 

I only have the 1680x1050 display, so I will do them at that res and 1280.
How do I get it to use dx10 or dx9? I am not even sure which dx version is installed with stock vista.

 

dx10 or 9 does not matter. it's the same thing for high settings. you only need dx10 to enable very high settings.

 

just start benching already! we are dying here waiting for you! gonna have a cig break now

 

LMAO. First test is running right now. 1680 HIGH

 

i bet we will have higher results with new drivers

 

I've had a hard time finding a good comparison to your graphic score (average of sm2 and sm3)... single card, same OS, same CPU speed. However, if I reduce the score I get with a single card by 5% to account for the XP/Vista difference, the number I get suggests the 9800M GTX is 6-7% faster than an 8800MGTX.

Looking forward to your Crysis tests.

 

I still think that’s a good score, especially since he’s using default drivers. It’s too soon to see something just blow the 8800GTX out of the water, but it is an improvement and that’s always a welcome.

9800M GTX #1!

Please post many pictures of the unit itself, and don’t forget the hard drive tune test. I know, it’s a lot of work but somebody has to do it.

 

I am having hard time trying to find where Crysis wrote my benchmark log. Dang vista wont let you write to "Program Files" folders....grrr. Might have to reinstall to another folder....So, hang in there for results

 

Do you mean you just don’t have 1920x1200, or it’s not available for that unit?

 

just locate the Benchmark_GPU file in your cysis installation directory and run it. the results will be posted right the dos command prompt right away.

 

I meant his LCD as his Clevo is not WUXGA.

Trance

 

I didnt go into DOS box first, just click thru explorer, so the batch file got confused since the current directory was not what it was expecting.

Re-running... sorry its taking so long.

 

I think it saves them in the "logbackups" folder but could be wrong.

 

No I only got the 1680, it also comes in 1900 but that res what too small for my eyes

 

not sure what you mean by that but as soon as you run the batch file it is automatic from there. just make sure to make any changes to game's video profiles in-game before you run the test as it uses the game's settings for the run.

 

Yes, but since I installed this in Program Files (x86), under Vista, by default it will not allow an app to modify files under this tree. So it virtualizes the files and writes them somewhere else.

 

No problem. Just think of us a vultures, sitting around waiting for a zebra to die.

 

Yep... sounds like Vista!

 

AAAAAAHHH! Autoupdates kicked in during GPU run!!! grrr. Moments away I assure you.

 

Great question! If you can't select between dx9 and dx10 in the game's option menu (I can't find it, but I'm running XP so perhaps that's why) I have no idea how it works. The third party benchmark offers a selection, but I'm convinced that doesn't work properly except at very high settings.

I assume you have dx10. Perhaps if you run "dxdiag" it will tell you.

 

LMAO!

Get those scores up! LOL!

 

Hmmmm, your score seems a little low from what I can see. I got a 10466 with the same os, ram, and cpu. My hdd was a 320gb 5400rpm drive and I have a wuxga display. My gpu is a 9800m gt. I also had my power mode set to performance, but I never changed anything in the Nvidia settings. Can anyone explain why I would have a score close to his when he has a better gpu and better hdds?

 

Because the GTX is shaping up to be an overpriced joke of a GPU?

 

Thats a bit harsh, I think people set their expectations to high for this gpu, there was a about a 15% difference between the 8800m gts and gtx and they where 32 sp apart. For only 16sp apart it shouldn't be anymore than maybe a 7% difference. For a game like crysis that runs in the 20-40fps thats only 1-3fps difference, =/ you would have to use a game that runs much faster, to see any real difference in a benchmark, but odds are you will never see the difference while gaming.

 

I m with ya. I dont think its too much to ask for than 5% performance from a $400 price tag

 

Yea, his scores are about 4% higher than mine, but he also has two 320gb 7200rpm drives. His better hdds should held accountable for at least 1% more than me.

 

What is the clock speed on those 9800M GTX? Im sure its just a overclocked 8800M GTX

 

No it is not. 8800M GTX and 9800M GT has 96 stream processors, and the 9800M GTX has 112 stream processors.

 

He meant it's performing like one.

 

oh

Btw, Kobe! Did you OC youre VGA to get 10990 in 3DMark06? Or just the CPU? What clocks are the 8800M GTX running on? (It would have been cool if you put a GPU-Z result too on the screenshot )

 

No GPU overclock, just CPU. But as you know, it’s a software overclock which means you’re also raising the FSB speeds of everything else. I’m sure you would like for me to post everything, but I’m sure I’m going to need every advantage I can think of once you get your 5796.

Seeing what you were able to pull off with that Gateway, I have my work cut out for me. Trust me, I left room for improvement.


BTW...it's 10999!

 

nah, I will not OC anything on my 5796...maybe when it gets cheaper, I will get an x9100 in it, but before that, I leave everything on stock settings

 

I wanted to leave everything at stock, but Crysis looks and plays so much better at 1650+ res. everything high and no lagging. At stock speeds, you will still have a ball! Your review is one I'm waiting to see, as you will have a direct comparison between....nevermind.

Later on down the road you can combine that x9100 with a 9800GTX, that sounds good.

 

Just an update, when I set nvidia to performance, I got a 10560 score in 3dmark06. That puts the difference down to about 3.3% between me and sliqsystems. I'm sure something has to be off here because there is no way that the upgrade in gpu and the upgrade in hdds can only be that much better.

 

The performance looks like its about right to me, people shouldn't be expecting more than 5-6% difference between these cards.

 

This is why I stuck with the dual 9800M GT's for $145 more than one 9800M GTX. Many people warned me of the potentially low increase, and I heeded their warning.

 

I really hope nvidia will get their act together. If their next mobile GPU will only be 5% faster then the 9800GTX then the gap between laptops and desktops is going to be huge. With Quad SLI on the desktop side and only 9800GTX SLI in our notebooks the performance gap is enormous.

 

i think alot of people miss the point with gaming laptops in one way. power consumption. quad sli sounds awsome but so does it's power consumption. what does that quad sli consume, 1500w? you would need $70 dollars a month in more electricity usage just to run your computer 18 hours a day.

220w vs. 1000w / 1800w is a huge difference in cost to operate.

 

Assuming the quadro 3700 specs are correct and it has 128 sp, thats probably going to be the next best gaming card for a while, whenever it's actually released.

 

its funny how johnny bravo didnt came back with the bench scores hes a lier LOL