gerardfraser:

Just tried the driver on Wolfenstein:Youngblood I must say DLSS on Balanced works fine so far ,looks great and areas where I dropped below 4K 60FPS before this driver are no more.Of course I need to test some more and it is not perfect but this did not bother me at all,white grass https://live.staticflickr.com/65535/49690838288_60c5781c2f_k.jpgYoungblood

I must say that looks really good! And this is coming from a 1080p monitor!

gerardfraser:

Interesting ,where did you come up with 1080P.

Er, my monitor?

pato9:

New driver benchmarked for u my friends, an short but precision benchmark: [youtube=-5r_kzJXkgU]

Hi @pato9 what's up mate? Happy to read you again and know you're still publishing your performance video analysis of Nvidia drivers. Great job mate! If you remember, I statarted performing my regular series of NV driver performance benchmarks (written format) thanks to yours 🙂. Currently, I'm still publishing them in a subreddit we created few months ago, which is related to all types of software and hardware benchmarkings. You can keep an eye on it and on the analysis other users and me are posting there regularly or occasioanally. That said, I've saw your performance comparison are mainly based in FPS Avg and GPU temp (ºC) metrics and comparisons, but would you also consider to include and capture frametime consistency/stability metrics (like FPS Lows or FPS percentiles) at some point? The FPS Avg or raw performance is importante but, usually, and according to my experience, the main and significant changes between driver version are more related with the its correnponding frame time stability levels. I usually find many cases where the FPS Avg can increases relatively, while its corresponding frame-time stability metrics doesn't improve accordingly, or even worsen significantly, being the gap size between its correcponding stability metrics and FPS avg noteworthy higher or lower than on prior compared or recommended versions. Basically, I mean that it's difficult to consider an increase in the FPS avg numbers by itself as the only key performance indicator or component of improvement or regression that could be attributable to driver version changes, without putting, relatively, those raw metrics or FPS avg results in relation with its corresponding metrics of frame time consistency too. Just to share with you some extra insights or suggestions on this matter and issues. Keep up the good job, sir! Regards!

KIBERMAX:

I don't understand why a lot of people write that their driver doesn't work and don't write full information. I always install new versions of drivers and I don't have any problems, this driver version is great all my installed games work fine. Sorry for my english.

Probably they don't even bother to make a clean driver install by using DDU, they just install drivers on top of other drivers.

Smough:

Probably they don't even bother to make a clean driver install by using DDU, they just install drivers on top of other drivers.

installing on top is typically fine to do, we're not in Windows XP era where the files install directly on top of ones in use. DDU is overused, even wagnard says not to use it for every driver update, its a failure recovery tool.

Astyanax:

installing on top is typically fine to do, we're not in Windows XP era where the files install directly on top of ones in use. DDU is overused, even wagnard says not to use it for every driver update, its a failure recovery tool.

Seems you know it all. Still, don't care much what you say, you are never wrong, so at least you can rest assured of that. I just like how several of the claims you've made are nothing more but your personal views on different matters.

Astyanax:

installing on top is typically fine to do, we're not in Windows XP era where the files install directly on top of ones in use. DDU is overused, even wagnard says not to use it for every driver update, its a failure recovery tool.

@Wagnard exactly says:

DDU should not be used every time you install a new driver unless you know what you are doing.

Therfore, if you know what you are doing when using DDU, that is, how to use and set properly such tool, there is no problem to use it. In fact, if the above condition is met, and mainly from an issue prevention perspective, its use is recommended. It is also recommended if someone want to roll back to a prior version or install a specific one. Anyway, I agree with you that an on-top type installation should be usually fine for most users. However, after this type of installation, and in case of image or performance issues or failures, that would not allow us to easily isolate or identify the source or factors that are causing them. Imo, an in deph uninstall via DDU and a clean install through Nvidia's own installer are good practices if you know what you are doing and it just takes a few minutes to be performed.

and if people knew what they were doing in the first place they wouldn't be using DDU every update.

Astyanax:

and if people knew what they were doing in the first place they wouldn't be using DDU every update.

so you telling the use of DDU not necessary on new update ?

endbase:

so you telling the use of DDU not necessary on new update ?

of course it isn't, i've never used DDU on any machine, nor drivercleaner when it was the thing to use.

endbase:

so you telling the use of DDU not necessary on new update ?

Yup, not necessary, I have never used it in like 15yrs of Nvidia driver use, Only time IMO would be if you changed AMD>Nvidia or had a real driver issue that reverting didn't fix. I normally use the clean option though.

I don't want left overs from older drivers so I use it between every driver don't want to manual clean tbh

You don't have left overs between drivers if you tick "clean install"

1) Disable Network Adapter. 2) Uninstall Old Driver. 3) Reboot. 4) Install New Driver. 5) Reboot. 6) Enable Network Adapter. That's pretty much how i do things. DDU is used for swapping Videocard Brands or as a last resort when having unexplainable issues.

pato9:

New driver benchmarked for u my friends, an short but precision benchmark:

@pato9 I just checked 15 of your posts, in more than ten of them you only seem to plug your youtube channel. Please consider this to be a friendly but one time warning, these forums are not your personal YT promotion grounds. And to the other debate yes, DDU should only be used as a last resort to solve an issue, that or if you swap a graphics card and need a clean registry. There is no need to apply it with every driver install, as you are increasing the risk of an issue, not solving them as there isn't anything to solve

EdKiefer:

Yup, not necessary, I have never used it in like 15yrs of Nvidia driver use, Only time IMO would be if you changed AMD>Nvidia or had a real driver issue that reverting didn't fix. I normally use the clean option though.

This! Though it seems the amd removal tool does a more thorough job than DDU on 2020 drivers.

Ever since 442.59 I've not been able to turn off Gsync on the desktop, the usual method of adding dwm.exe (desktop windows manager) to the list of programs with monitor technology set to fixed refresh isn't working. Tried this latest driver, same issue so had to drop back to 442.59

cav800:

These 445.75 drivers are crap for me. After I installed them, I got "DXGI error device removed" crashes for all of my games that use the CryEngine. I reverted back to 441.66 drivers and those games play just fine again. I'm sticking with 441.66.

already fixed.

Does anyone have issues playing HDR content on youtube 8K60fps? playing it normal no HDR works fine 8K60 I have 1080GTX and AMD 1950x Those I can't even play 8k60 HDR, I could play them before, don't know what happened or changed. could be windows update, chrome update, Nvidia updates. [youtube=N1-Jmq7BLFE] [youtube=1La4QzGeaaQ] This works HDR but slow with many fps dropped both CPU and GPU are at 30% [youtube=hVvEISFw9w0]

the 1080ti doesn't support 8k vp9 decoding under 10bit iirc