PDA

View Full Version : New 162.22 Vista and 162.18 XP WHQL


Pages : 1 [2]

Redeemed
07-27-07, 03:07 AM
These seem to have solved the problem I was having in Lost Planet. They also seem to have given me about a 5fps increase over the previous drivers (158.22 I believe they were). Atleast, in Lost Planet anyways. Have yet to try any other games.

WeReWoLf
07-27-07, 04:48 AM
Still no fix for Vista and Scarface *sigh*

Revs
07-27-07, 05:11 AM
Any feedback on DX9 games in Vista with these drivers??

I aint had chance to install em yet. With the Beta's, some DX9 games were perfect and others were poor. HL2 E2 just would not run in SLI, but was fine in non-SLI.

Any changes in this respect??

PTK1982
07-27-07, 05:50 AM
It is amazing how this driver gets WHQL certification. I know that WHQL certification is as useful as umbrella in sun shine but i would atleast expect them to test the driver in basic windows usage. Then M$ would notice that driver crashes every time that favorites is opened in IE7. But i quess M$ was too busy count dollars that actually test driver.
Just amazing... back to 158.45 last good driver without that ******* drivers stopped responding messages.:(

Electronic Punk
07-27-07, 06:05 AM
That isn't what WHQL is.

Revs
07-27-07, 06:05 AM
I get the feeling this is a Vista error. If something doesn't timeout as fast as Vista would like or something.

On my slow ass work machine, when I open Outlook I get a 'Not Responding' message in the bottom right corner. But it's not that it's not responding, it's just not responding fast enough for XP's timeout settings.

I read on a thread somewhere on this site there is reg. entry u can alter that can stop the driver error in many cases, but I don't know if it effects anything else in a bad way.

Knacki99
07-27-07, 07:02 AM
Actually I am more interested in a new driver that fixes the RAM bug than an old beta that has turned WHQL 2 weeks after it was released.

Buckeye
07-28-07, 10:25 AM
Installed 162.22 and nTune 5.05.47.00

nTune does not seem to want to adjust my second GPU fan's speed at all, first one adjusts ok. Drop down from GPU 0 to GPU 1, adjust fan speed, hit apply, no effect.

SLI vusual indicator seems to be working ok, I get the green bar on the left side and it expands nicely. Can't remember but do we get two bars with SLI turned on ?

WoW is working fine with SLI turned on, pegged at 60 fps everywhere. No texture flashing like I had before with SLI turned on.

LoTRo is running fine max'd out settings with SLI turned on. I do get some strange bright flashes though. I was in a area that had some rain falling and sometimes as the rain passed over me the screen would brighten for a sec, could be lightening or something not sure.

As soon as RiveTuner gets updated I am going back to that instead of using nTune. I tried RiveTuner 2.0.2 and only see 1 GPU and can adjust that fan ok.


update:
Driver not responding hit me hard again, I was just moving some icons around on the desktop then Bam ! Once it starts and even doing a hard reboot to get it to stop it comes right back :(

No more messing around with drivers for me until they get this nailed down, to much wasted time.

jabloomf1230
07-30-07, 09:53 AM
That isn't what WHQL is.
Yeah, most people think that there are a bunch of guys in white lab coats at MS, who sit all day with thousands of computers and test stuff. Although MS does independently test some hardware, most software and device drivers are tested by the manufacturer itself (or specialized contractors). To get WHQL certification, a device driver has to pass a series of tests, which they run with software that the manufacturer obtains from MS. The company then passes along the test results to MS and if the driver passes all the tests, then it receives WHQL certification (and signed drivers). Thus, WHQL can't possibly assure that a device driver will work for all possible combinations of hardware and software and in fact, it never does. If anyone is doing something more boring than watching paint dry, they can read the MS website that talks about WHQL, Windows Logo and driver signing:

http://www.microsoft.com/whdc/GetStart/default.mspx

vmsein
07-30-07, 03:49 PM
Installed these to see if they resolved the TDR errors I have been getting with any driver after 158.45, and regrettably, they have not. Back to 158.45 for me:thumbdwn:

ragingWS6
07-31-07, 09:23 PM
Looks like I won't even bother to try these out. I'll stick with the 158.45s then.

$n][pErMan
08-03-07, 10:16 PM
Hmmm .. after reading this thread it looks like im staying with my 158.45 set :p

jolle
08-11-07, 10:40 AM
Waiting for a fix for the horrible performance of WOW in Windowed mode.
This set wasnt it..

fasedww
08-12-07, 01:01 PM
[pErMan']Hmmm .. after reading this thread it looks like im staying with my 158.45 set :p
Stay with 158.45 I tried these and right away the not responding crap, so I'm back with 158.45 too:(

Phyxion
08-12-07, 01:43 PM
Stay with 158.45 I tried these and right away the not responding crap, so I'm back with 158.45 too:(
Maybe you should try to revert back to stock speeds, 3.5 MHz cannot be a proper speed for that CPU:captnkill:

fasedww
08-12-07, 02:02 PM
Maybe you should try to revert back to stock speeds, 3.5 MHz cannot be a proper speed for that CPU:captnkill:Read this; At least with Vista x64 these WHQL drivers act no differently than the beta 162.22 drivers regarding the frequent TDR / nvlddmkm errors. Not surprising since I think the drivers are the same other than getting signed. I can very reliably trigger a TDR / nvlddmkm event by opening IE7, activating the explorer bar, then navigating through the favorites via mouse. Aero causes the problem; Dreamscene may exacerbate it. If I turn off desktop composition (disabling Aero in the process) the problem goes away.

I am still convinced these errors are a driver issue, although it may only manifest in systems like mine configured with SLI and four DIMMs. As I have after trying all the subsequent driver releases, I wound up reverting back to 158.45 which is the last set that does not cause these errors in my system. Since the major difference between 158.45 and later drivers is support for DX10 in SLI, I wonder if that is the underlying problem...

My system is completely stable using the 158.45 drivers with the E6420 @ 3.33GHz using linked/synched FSB with memory running @ 833mHz (4-3-3-10-14-2T). RAM is four 1GB sticks of Crucial Ballistix Tracer PC2-8500. I did try dropping the FSB to 1333, cutting the CPU to 2.67GHz and the RAM to 667mHz, keeping the same voltage settings. I don't overclock my 8800ultra GPUs. I also tried disabling Nvidia GPU Ex in the BIOS and UAC, even though I log on to the Administrator account. None of this had any effect on the errors on the drivers sets after 158.45, so it definitely is not just a problem related to overclocking.


:rolleyes: Hell No' I have been playing F.E.A.R. for two hours straight at these speeds and with 158.45, I know what I'm doing on my overclock and setting voltages etc. to have it stable. It's this driver set with my system I get driver not responding Etc.

Vik1dk
08-12-07, 02:18 PM
Maybe you should try to revert back to stock speeds, 3.5 MHz cannot be a proper speed for that CPU:captnkill:
went over his head.....;)

Talon
08-12-07, 06:38 PM
went over his head.....;)

:lol:

emgarf
08-15-07, 06:43 PM
Using the 162.22s on Vista 32... I set nTune to load a custom profile forcing the fan control mode to automatic. This fixes the 100% fan speed issue, but only temporarily - when the system resumes from sleep, the "Direct fan control" mode has been reenabled, even though the profile is set to use "Automatic fan control".

The 100% fan speed is driving me nuts :argh: since this system was built as a "quiet" system... anyone know the most recent driver that doesn't have the fan speed bug?

cheers,
emgarf

emgarf
08-18-07, 06:27 AM
Update: Went back to 158.18 for now; fan speed is now correctly in automatic mode at all times. If anyone can suggest a better driver than 158.18 that doesn't have the fan issue, please advise... cheers!