Official Detonator 40.41 out!

[jiffylube1024]

Originally posted by: MrPeepers
Why get excited when a benchmark score goes up? Real games matter, unless you play benchmarks.

It happens all the time with the Radeon 8500. So many people thing it's way faster than the GF3 because of 3dmark2k1. It's criminal, I tell ya!

 

[Rand]

It seems like almost everyone in the GH thread is now reporting decreased image quality and/or stability issues.

Very disappointing. Similar performance, with worse quality and questionable reliability from the looks of it.

I do like the new driver options inteface though, much more convenient to use and it fits in well with WinXP's GUI.
They've also added an anisotropic filtering option for DirectX so one no longer needs a third party tool to enable anisotrophy in DirectX.

Multi-Monitor capabilities seem to have been improved slightly, and their revealing a lot of the options that were formely "hidden" in the registry and now enable for view by default.


If nVidia can fix the image quality issues and improve stability then I'll be happy with these drivers, even if they don't really impact performance at all.

 

[Uclagamer_99]

ran 3dmark2k1 and it is FASTER but the image quality is SUBSTANIALLY worse...every test, the textures looked washed out and the color was dull almost as if it were 16bit textures as opposed to compressed...just terrible...i'm going back to the 30.82s for now

 

[jcarson]

I think video card makers/driver makers are getting smart to the whole benchmark crowd. I would bet money they have 3dmark and are making adjustments to see if it improves a score regardless of actual performace increase.

The label on the new drivers should be

"3DMark Enhanced Drivers: To get your 3dmark scores WAY up" (regardless of actual performance)

 

[Insidious]

If I were Nvidia, and I browsed these numerous threads........ What would I think?

Would I believe the majority of the users of my product were unimpressed by a funciton strip that raised 3DMark scores?

Would I believe the majority of the users were savvy enough to recognize the change in performance without a benchmark tool?

I doubt it........... browsing these threads made me believe that there are a few with astute abilities to evaluate the performance of
a video card as part of a system overall, while the majority of Nvidias' product users are lemmings with no point of reference other than 3DMark scores.....

It's not Nvidia who needs to wise up............. Their hands are tied by the ignorance of their users......... They are being forced to design to the
standards set by the majority......... So sad :disgust:

 

[SunnyD]

Anybody getting blurred lightmaps on textures with ansio (color bleeding)?

Anybody getting random screen flickers, and other graphics corruptions?

Needless to say I went back to the official 30.82 drivers and all is well again.

 

[ViRGE]

Guys, I'd reccomend you take a look at this(page may not be work appropiate). A fellow has been messing with a "texture sharpening" setting new to the drivers, and has done some speed/quality corrolations. It looks like Nvidia might have turned something off in these new drivers to get that performance boost.

Pics:

40 w/sharpening
40 w/o sharpening
sharp zoomed
no-sharp zoom
previous drivers

 

[onelin]

AFAIK SP1 is going to fix the refresh thing, not drivers. In addition, the refresh fix I tried messed up my TV out (TVs don't like being told to go 85Hz) until I reverted. It seems like a tricky matter for dual display (two monitors or monitor and TV...and I do both unfortunately :-/) So for now I live with it or use r_displayrefresh since most games I play are q3-engine anyway. Notice how this is a problem for other videocard vendors, not just Nvidia. It's about bloody time it gets fixed, though :/ ...but it needs to be fixed at the source, not forced across all displays

Any ETAs on SP1? As a side note, I went back to the 30.82's after a brief test of these.

*edit* oh, and I saw something about this in alt.games.quake3 ... this texture blur is due to compressed textures, if they made a quality change it may have been there...which, I don't give a damn about...because the answer to this is: TURN OFF TEXTURE COMPRESSION, IT ALWAYS SUCKS there...got it out of my system. This is why I noticed no quality change while playing quake, I can even lineup screenshots. *end edit*

 

[Jeff7]

*edit* oh, and I saw something about this in alt.games.quake3 ... this texture blur is due to compressed textures, if they made a quality change it may have been there...which, I don't give a damn about...because the answer to this is: TURN OFF TEXTURE COMPRESSION, IT ALWAYS SUCKS there...got it out of my system. This is why I noticed no quality change while playing quake, I can even lineup screenshots. *end edit*

You sure? Most stuff I see indicates that texture compression can actually improve quality, as the videocard can store higher quality textures in less space. There are new lossless compression methods available too. Sorry, but I can't provide any Q3 screenshots right now - I can't find my Q3 install CD.

 

[Taz4158]

Any ETAs on SP1? As a side note, I went back to the 30.82's after a brief test of these.

Within 10 days.

 

[Xtasy]

What util w/clocking can i use in conjunction with the 40.41's? I saw nice nature frames (almost as good as 23.12) on the beta ones when my gf3ti200 was clocked at default! Hopefully when i set my card to ti500 levels with these drivers, i can get nice increase in fps.

 

[happybob]

anyone with problems in games with this new driver? i noticed that there seems to be a bug in the direct 3d antialiasing options with some games. when i select "let application choose", everything runs fine, but when i set it to manual at 2x, 4x, etc... i get problems:

GTA3: it's basically unplayable when i try to set the antialias settings manually.
NHL2002: playable, but with lots of artifacts...

anyone else noticing these problems? i know that when i was using the last Detonator, none of these problems were there.

 

[Mem]

anyone with problems in games with this new driver? i noticed that there seems to be a bug in the direct 3d antialiasing options with some games. when i select "let application choose", everything runs fine, but when i set it to manual at 2x, 4x, etc... i get problems

Only had one problem with anisotropic filtering, at 2x setting(D3D) in Might & Magic 9 I get weird artifacts,at 1x it`s fine, just one game so far.

 

[ElusivePeon]

When I try coolbits the sliders top out at 315/560 on a Asus Ti4200. Anyone know how to increase the max speed settings on the sliders? I know the card can go higher...

 

[kyu614]

Hi, im kinda new to this stuff, but does it matter that it says WHQL Certification Pending? Does that mean its some kinda beta driver? How long does it take to get the certification?

Also, do i just run the file after i download it to install the new driver? or do i have to go to the hardware profile and select update driver and select the downloaded file?

thanks.

 

[Mem]

kyu614,remove the old drivers first,see this here .

 

[Crapgame]

NVIDIA Acknowledges Bug And Will Fix
Posted by Matt Burris on August 30th, 2002 - Friday 6:31 pm

Just got word from Brian Burke at NVIDIA this bit of information that should help clarify things for a lot of you out there in regards to the new Detonator 40.41 drivers:


The default setting on the new driver and the old driver are both the same. The problem that is being identified is a bug that occurs when the AF settings are changed. It seems that once the AF slider is moved away from 0 to another setting, then back to 0 again, it triggers a bug that causes the control panel to misread the register setting. That setting is not point sampling, but it is something less than 0. The only way to "reset" back to true 0 is by reinstalling the driver or deleting the registry setting.
We are going to fix this bug and others and submit the new BETA driver to WHQL, before the driver moves from BETA to "official".

Also he had this to say:


A lot of the improvements in performance that are seen are due to efficiencies in the driver code that controls vertex and pixel shader. That is why you see the jump in peformance for Nature and Aquanox. Those apps also happen to be the ones that are used for benchmarking.

 

[Jeff7]

Does anyone happen to have any idea or information as to when support for these Detonators will be added to Rivatuner? The current version supports up to 30.82; the current tweaks don't have any effect on the 40's, and I'm not about to leave Rivatuner - it has too many good things that I don't want to give up: Good interface, lots of tweaks, and the Launcher that lets you load custom tweak-sets for different games.

 

[jdogg707]

Ok...Nvidia is doing something really messed up with these drivers...with the 30.82's I got a 3D Mark 2001SE of 12,045. With the 40.41's I get a score of 13,819 with Texture Sharpening disabled and a score of 10,666 with TS enabled!!!! What's up with that? I think I'll go back to the 30.82's for the best of both performance and visual quality...

 

[BFG10K]

The Good

• GLQuake now works at every resolution my system supports. The driver is correctly reporting the available modes to the program.
• Slightly faster performance in some programs like Quake2 and Quake3 (with anisotropic filtering).
• Great new control panel design including anisotropic filtering control in Direct3D.

The Bad

• RivaTuner's refresh fix no longer works (fixed by using Refresh Force instead).
• All games are dark. It's like the gamma default is lower when switching to 3D games.
• Serious Sam still has serious Z buffering issues in certain maps in OpenGL mode. I have verified that this is an nVidia specific bug because when I force a 24 bit Z on my Radeon the problem is nowhere to be seen.

No Change

• From my quick IQ checks in 3D and 2D I couldn't see anything significantly different. That's not to say that nothing's changed, just that nothing hit me as immediately different.