• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Nvidia GeForce 378.78 WHQL driver

nvgpu

Senior member
http://www.geforce.com/whats-new/articles/tom-clancys-ghost-recon-wildlands-game-ready-driver

nvidia-geforce-gtx-game-ready-driver-378-78-directx-12-performance.png
 
Notice how its "Game Release" and not driver? Makes it sound like the gains will be from game updates, not this driver. Anyone actually test the driver yet?
 
zero gain in tomb raider with 1070

I figured everyone knew that the in game benchmark was completely useless in Rise of the Tomb Raider😵. There's a guy on Guru3d forums that says he got 15-25 extra frames in game vs DX11. There's also reports of big gains in Hitman as well.

The only DX12 game I have installed at the moment is Deus Ex MD (though I'm thinking about installing Rise of the Tomb Raider), and I saw a 1 FPS boost in a very GPU bound area. I think most gains with these drivers are going to be in more CPU bound scenarios.
 
Much better minimums though so a lot better than zero.
Minimum frame-rates are useless in canned benchmarks, especially if it goes through different scenes with camera transitions.

In which game do you pan the camera across a wide-angle shot while actually playing it?
 
Much better minimums though so a lot better than zero.

The minimums and max don't matter in the benchmark as it seems to capture the loading between scenes. Notice that the min goes up but max takes a massive hit. Same deal the benchmark isn't good for those.

I figured everyone knew that the in game benchmark was completely useless in Rise of the Tomb Raider😵. There's a guy on Guru3d forums that says he got 15-25 extra frames in game vs DX11. There's also reports of big gains in Hitman as well.

For min/max its way off yes, but avg should be fine.

That guy is comparing DX11 vs DX12 fyi:

dx11- 90fps
dx12 - 110fps

So not actually comparing old driver vs new.
 
From your own link:

Hitman performance gains

Most of the performance improvements are going to come from CPU bound scenarios.

You do realize that was posted after my comment right?

That same person got 0 gain in ROTTR which was supposed to have a whopping 33% increase according to Nvidia.

This person had no gain in hitman:

https://www.reddit.com/r/nvidia/comments/5yfe3o/driver_37878_faqdiscussion_thread/depqox0/

So obviously the gains aren't what was expected, but with absolutely no testing data provided from Nvidia in order to reproduce their results, not unexpected.
 
From Nvidia's slide I read 'Game Release' as in performance for the Game Ready driver whenever that game came out and not necessarily the driver right before 378.78. We could have seen incremental improvements since the game came out through today and that is the combined sum. They could be comparing to drivers from over a year ago. If true, its a little misleading and would explain why no one is seeing the claimed improvements moving from more recent drivers.
 
@Krteq posted this in the other thread in general:

index71bt2.png


Guru3D - Quick comparison: Geforce GeForce 378.66 versus 378.78 DirectX 12 performance

Only hitman saw any performance gain, but that was fixed in a previous driver as well as they had (slightly higher) fps of 105 in their Ryzen test last week. So people with month old drivers might see a boost in hitman, but it wasn't from these drivers. The gains @Carfax83 mentioned were from 376.33 which are from december drivers to now, so it was sometime between them.
 
I don't know, I still think they should test at 1080p to come a to full conclusion as to whether there are improvements or not. The bulk of the performance increase for DX12 is always going to come on the CPU side rather the GPU side.
 
I don't know, I still think they should test at 1080p to come a to full conclusion as to whether there are improvements or not. The bulk of the performance increase for DX12 is always going to come on the CPU side rather the GPU side.

Nvidia's test was @ 4k not 1080p.
 
Why do you keep posting that image? Do we really need it 3 times in a single thread considering it's just lies and multiple sites have proven the perf gains aren't there except for some in Hitman?
 
Why do you keep posting that image? Do we really need it 3 times in a single thread considering it's just lies and multiple sites have proven the perf gains aren't there except for some in Hitman?

Show me your evidence.
Show me multiple sites that tested the games release driver vs the latest driver.
Here is the settings they used. Make sure the reviews use the same settings and the same driver versions..and the same games. ANd with a gtx1080
Good LUCK!


Game GeForce 368.61 vs GeForce 378.74 Performance Gain

Ashes of the Singularity, Crazy Preset
The Division 1.6, Max Settings + 1x SMAA Ultra
Hitman, High Settings + High SSAO
Rise of the Tomb Raider, Very High + 2x SSAA
Gears of War 4, Ultra Preset


Here is my article .
http://wccftech.com/nvidia-geforce-378-78-whql-game-ready-dx12-drivers/
 
Last edited:
Uhh... your article is a press release??? Get out of here lol

They didn't test, they just copied the numbers from the image into a table.

Please show where they got the settings from, because wccftech is anything but reliable.
you said they were lies.
Burdon of proof is on you bud.
 
GeForce GTX Driver Performance Gains: July 2016 vs. March 2017 (4K)

Game GTX 1080/GTX 980 Ti

Rise of the Tomb Raider
Even/Even

DiRT Rally
+8%/+7%

Ashes of the Singularity
+11%/+14%

Battlefield 4
Even/Even

Crysis 3
Even/Even

The Witcher 3
+2%/Even

The Division*
-7%/-9%

Grand Theft Auto V
+2%/Even

Hitman (DX12)
+26%/+24%

I’ve already seen some confusion here over whether the GTX 1080 Ti launch driver (378.78) is a special performance driver or not. For the record, it is not.

http://www.anandtech.com/show/11180/the-nvidia-geforce-gtx-1080-ti-review/4
 
The only fact is that NVidia is definitely improving their DX12 driver, which is to be expected I suppose. And with WDDM 2.2 which will be released in the Creators update, these improvements are likely to continue. They're probably going to release a special Creators driver version for the Windows 10 Creators update like they did last year when the Anniversary update became available.
 
Back
Top