Nvidia 195.62 WHQL for mobile released:

http://www.nvidia.com/Download/index.aspx?lang=en-us

Enjoy!

 

Thanks for the heads up
I'm downloading it now!

 

post some benchmarks pls!!!thanx!!!

 

Nvidia 8800M GTS - overclock to 610/975/1525 - Win7 64bit.

3DMark06

186.61 = 9411
195.62 = 9506

 

Is there really any difference between these and the desktop version?

 

some real game benchmarks!!!

 

Worthless.. 3DMark scores easily jump 100 points..

Have anyone got that brightness issue with these?

 

Depends on your system. I've never had a 100 point jump with any single driver running the test multiple times. I've only witnessed 100+ pt jumps with different drivers.

What's more, in game results between 186.81 & 195.55/195.62 have been dramatic.

Are you referring to wake from sleep issue? If so, then yes, it exists. I've remedied the issue by setting the manual (using Fn + brightness control keys) to the same level as windows power profile brightness.

 

Can't seem to get this to work on my 8800M GTX, reverting back to old 186.81 drivers

 

hmmm interesting. maybe i should post this in the sager forum, see if it improves on the gtx260/280 performance

 

on the details page for the driver, says something like: "If your notebook is a [...] XPS M1530 [...] we recommend installing this Dell update before updating the drivers..."

But there is no link on "this Dell update"... what is this?

--------------
EDIT: Nevermind, the link appears on the english version of that same page. It's a link to A12 BIOS update.

 

my 3d mark 06 score!!!

 

Excellent Driver.
Fix the It fixes the "unable to throttle back to default frequency in Windows Desktop 2D Mode" issue that had plague previous driver releases.

 

That's a really good score for the 9600GT!!

 

It was at 1280x 800 though.

 

It's still a very good score. Besides, most laptops that are 16:10 are tested at this resolution.

These didn't do much for my 3DMark06 score: I get 3926 whereas the highest I ever had was 3969 (CPU: T9300, GPU: DDR2 8600M GT). It could just be the general deterioration of Windows -- I've noticed quite a few things behaving slower than they did before. In a couple of weeks, I'm going to get rid of Vista x86, install Win 7 x64 and then test again with the same drivers.

 

HDX 9300 - Dragon
8800M GTS - OC 610/975/1525
T9300 2,5ghz
Win7 64bit
1280x1024

 

I just had bad experiences with this driver for my 88 m SLI. In first try, the screen blacked out and I had to force shutdown. Later, when I installed it succesfully, it doesnot allow be to benchmark. If I run any, then the system will freeze. Same stuff is happening with games so I wanna stick with 186.82.

 

this release seems to reduce fps and image quality for GTA4 on my 260m GTX, anyone?

 

Well, haven't loaded a game yet to try, but right of the bat, it's got the downclock problem for me.... ie, GPU stays at full 3D clocks all the time, even idle at desktop. 186.82 worked flawless in this area, will probably be going back, as it freakin kills my battery when on the go, and keeps the GPU in the 60's constantly, even idle.... oh well

 

Hmnn 186.82 gives the BEST fps at l4d 2.

 

What did you use to test the clocks?

 

No they aren't. If a screen supports the program's default resolution then that should be used for comparison. Aspect ratio has nothing to do with it.

 

On this site, they are (or at least they were, prior to the coming of the hideous 1366x768 screens). Take a look at the reviews of the 16:10 laptops and you'll see that the main resolution used is 1280x800. This argument has come up quite a few times and it's more useful to have results at the same resolution that can be easily compared to all other machines to get them at the native resolution of a given machine.

 

so are these drivers good or not?

 

Checked them with both Everest Ultimate and GPU-Z, both show them staying stuck on full 3D clocks (ie, in my case 550,900,1375)

Also, just ran about a half hour in NFS Shift, and these run hotter than the 186.82's do.... hottest I ever got before was 93C after a couple hours of playing, these hit 96C in half an hour...

Think I'm gonna drop back to the 186.82s, those worked great for me atleast.

EDIT: Just walked away for a few minutes, came back.... GPU had cooled to 67C, but then went up to 86 sitting idle at the desktop! Now it's back down to 72C. All this is at 3Dclocks. Odd.

As a comparison, with 186.82's, idle desktop would will range from low 50's to mid 60's then back down to low 50's, but this of course at 2D clocks.

 

Windows 7 64 bit here brings no improvement.
Here, to compare my results with Windows Vista Home Premium 32 bit.
(OS and this may very well maintained with you the reason for his poor results)

At the same settings as 2.0. The same notebook, same CPU and GPU. However, I have no SSD but see signature.
610/975/1525 = 9552

my max overclock
650/1040/1625 = 9891

At the time the brightness problem on the track. See also my posts
to 195.55 Beta: http://forum.notebookreview.com/showpost.php?p=5544831&postcount=8490
and: http://forum.notebookreview.com/showpost.php?p=5549493&postcount=8537
The problem is the 195.62 WHQL also available.
Furthermore, the maximum brightness on my laptop is no longer accessible via Fn + F7. Only about Mobility Center (Windows key + X).

The min brightness in turn can only be reached via Fn + F7.

For example:
1.Start Mobility Center
2.brightness about Mobility Center push to max. not close (!)
3.brightness via Fn + * reduction (*= in my F7) and then bring about Fn + ** (**= at my F8) to max.
This will not reach the maximum brightness. Also moves the slider in the Mobility Center does not.
4.A click on the slider in the Mobility Center will show again the maximum brightness.

Max brightness can only be achieved through Mobility Center! (at HDX9xxx WinVista32)

At minimum brightness of it, this effect exactly the opposite.

When you uninstalled the driver and before all power schemes will change to 100% brightness, then after reinstalling everything first ok. Also, the controller in the mobility center is also moved.
Once in power scheme, however, the brightness changes and restarts the fault is again present.

 

finally! Been waiting for it long time now.

 

http://a.imagehost.org/0585/failscore.png 195.62 WHQL

http://a.imagehost.org/0283/winscore.png 186.82 WHQL


186.82 definitely gets my vote. same CPU and GPU clocks. Nvidia GTX 260M @ 600/1000/1550 and Intel P8700 @3.102GHz

 

On the 8800M GTS, GPU-z .0.3.4 & .0.3.8 both show clocks downclocking to min clocks. Temps a degree or two cooler on load with overclock playing MW2 & L4D2 than 186.82.

Less A/V stuttering also.

Had good experience with the 195.55 beta, figured I'd give the WHQL a go.

Chalk it up to the HDX Dragon's custom made for HP MXM card.

 

Good Lord man those are some great temps on that machine for those clocks... I'm jealous! Sheesh!

And yep, going back to 186.82's myself, they have been fantastic drivers for my machine.

 

Goodness, these benchmark feedbacks and replies are good to me for not to test this Driver myself. Because i don't think any Driver can beat the almightly 186.82 easily. I wonder when will Nvidia released a Driver based on 186.82 modification.

 

Yes but does this driver perform better IN GAMES?
Seriously...who cares about -500 points in 3DMar06 if I get +5FPS in games ? 3DMark06 was released almost 4 years ago ...

 

I lost ~2.5 FPS in mass effect, 4max and 6 min FPS in l4d, and 2FPS in the crysis benchmark average FPS going from 186.82 to 195.62

 

Agreed ^^^

 

No real change in FPS for me with latest update and temps seem a little hotter. But will keep playing about with it to see the flash 10.1 beta improvements it promises

 

Since using the 195.xx series of drivers, I've noticed that most flash-based video players are acting in an odd way. Whether it be the navigation controls not rendering in the correct place, or videos crawling to about 1FPS. This is in Mozilla Firefox 3.5.3, Windows 7 X64 btw.

 

Yeah, these 19x.xx series seems horrible

 

I can't seem to get these drivers to work on my 8800M GTX. Their supported products page does state that my GPU should work with said driver though. Anyone with an 8800M GTX has gotten it to work?

 

in the AW forums Cha, it seems people are having issues with it, having to use a modified .inf. On another note: I also had random stutterings in mass effect dropping me randomly from 60FPS to 5-10FPS (while running around the presidium) so these drivers are definitely a no go for me.

 

Im using these drivers right now as they're the only ones that let me easily completely disable powermizer with my SLI setup, which seems to fix the audio stuttering I get. Unfortunately I get some brief freezes so they're not perfect in the slightest, but the older drivers dont completely disable powermizer and I still get the audio stutter. Until a new driver comes out which improves this whilst retaining the ability to switch powermizer off completely then I don't have much choice in what to use (195.55 also had the option, but didnt completely disable it, only locked the clocks so still got audio stutter).

 

Yeah, I did start a topic there regarding these drivers. But seeing as to how many people are having problems with'em, I will stay with 186.81 for the meantime

 

where do you get 186.82?
i only see 186.81 on nvidias site

 

How are you guys able to overclock so much, my 8800m gtx only goes to 575/1400/854 beyond that , something crashes and driver reboots with a rigid lower clock speeds and it doesnt got back to normal until a reboot
tried different drivers , monitored temps as well nothing beyond 70 either

 

NVIDIA GeForce v186.82 WHQL Drivers

 

did you do a driver sweep before installing the new driver?

 

Every individual card is different. Both of my HDX's can overclock to different amounts before issues. Some HDX owners can overclock even higher than what I'm able.

 

A few 3DMarks isn't enough to dissuade me from moving to the latest Verde driver.

186.81 - 3856 Marks, 1523 in SM2.0, 1417 in SM3.0
195.62 - 3840 Marks, 1517 in SM2.0, 1416 in SM3.0

OC'd results below were run at 675/475/1425 speeds:
186.81 OC - 4811 Marks, 1959 in SM2.0, 1827 in SM3.0
195.62 OC - 4795 Marks, 1950 in SM2.0, 1827 in SM3.0

And yes, CPU scores were down between 186.81 and 195.62, despite not changing CPUs.

 

my GPU scores went up but the CPU one's went down... haven't tried gaming yet...

 

Is that possible that the PhysX affects the CPU Scores?
Because Forceware 195.62 WHQL have newer PhysX compared to Forceware 186.81 and 186.82.