Some NVIDIA R337.50 Driver Controversy

Subject: General Tech, Graphics Cards | April 8, 2014 - 06:44 PM |
Tagged: nvidia, geforce, drivers

NVIDIA's GeForce 337.50 Driver was said to address performance when running DirectX 11-based software. Now that it is out, multiple sources are claiming the vendor-supplied benchmarks are exaggerated or simply untrue.

View Full Size

ExtremeTech compiled benchmarks from Anandtech and BlackHoleTec.

Going alphabetically, Anandtech tested the R337.50 and R331.xx drivers with a GeForce GTX 780 Ti, finding a double-digit increase with BioShock: Infinite and Metro: Last Light and basically zero improvement for GRID 2, Rome II, Crysis: Warhead, Crysis 3, and Company of Heroes 2. Adding a second GTX 780 Ti into the mix helped matters, seeing a 76% increase in Rome II and about 9% in most of the other titles.

BlackHoleTec is next. Testing the mid-range, but overclocked GeForce 760 between R337.50 and R335.23 drivers, they found slight improvements (1-3 FPS), except for Battlefield 4 and Skyrim (the latter is not DX11 to be fair) which noticed a slight reduction in performance (about 1 FPS).

ExtremeTech, finally, published one benchmark but it did not compare between drivers. All it really shows is CPU scaling in AMD GPUs.

Unfortunately, I do not have any benchmarks to present of my own because I am not a GPU reviewer nor do I have a GPU testbed. Ironically, the launch of the Radeon R9 295 X2 video card might have lessened that number of benchmarks available for NVIDIA's driver, who knows?

If it is true, and R337.50 does basically nothing in a setup with one GPU, I am not exactly sure what NVIDIA was hoping to accomplish. Of course someone was going to test it and publish their results. The point of the driver update was apparently to show how having a close relationship with Microsoft can lead you to better PC gaming products now and in the future. That can really only be the story if you have something to show. Now, at least I expect, we will probably see more positive commentary about Mantle - at least when people are not talking about DirectX 12.

If you own a GeForce card, I would still install the new driver though, especially if you have an SLi configuration. Scaling to a second GPU does see measurable improvements with Release 337.50. Even for a single-card configuration, it certainly should not hurt anything.

Source: ExtremeTech

April 8, 2014 | 07:55 PM - Posted by Anonymous (not verified)

can't say i am surprised, the hint that nvidia wonder drivers are gonna match low level API in cpu overhead and performance boost, didnt make sens to me not even for a sec, the only explanation i found is sli scalling, old games reoptimsed for new hardware, or bugs, and in the end it turned out a little bit of all of them in addition of the disapointing marketing graphs.
but you cant make fan boys listen to reason.
now i am starting to worry about nvidia, they are taking way too many hits at a very close rate, and the bruises will start showing all at once, with a loss of huge portion of market share.
nvidia over the top with propritary eco-system, 3d,physx,g-sync..
AMD reply by open and free 3D,tressfx,bullet physx,freesync
Nvidia focuses all their resources on drivers to beat AMD on performance boost
AMD pushes in Mantle, and the industry forward ( which was an extremly hard thing to do, and they succeeded briliantly), now with DX12 low level, AMD doesnt have to worry about drivers anymore, and strips nvidia from their major winning tool along the way.
and now the R9 295X2, N°1 perfomrance single graphique card, designed for 4K, in a time where 4k is about to Boom.(at half the price of Nvidia's Bi-Gpu)
like in mortal kombat, i dont know if it's a Flawless Victory or a Fatality
what do you think ?

April 8, 2014 | 09:02 PM - Posted by Anonymous (not verified)

Nice bias opinion

April 9, 2014 | 05:19 PM - Posted by Anonymous (not verified)

well i am from the red team yes, but i dont consider myself bias, because i try to stay objective and build my argument on facts in front of me not speculation, and i dont dismiss logique and reason for emotions.
if i am wrong on any of my arguments counter with facts and logicly to convince me otherwise, dont call me bias and hope that you made any kind of point.
here is another argument to why i worry about nvidia, for years they have been free roaming around Pro Gpu market with their insane margins, in 2009 AMD decided to in the market with FirePro and took 10% of the market, in 2013 they have 30%, and projection for 2015 at 50% market share.
add to that all the deals that brings money to nvidia from last consoles are about to come to an end, while AMD secured deals with nintendo/microsoft/sony for the nex generation thats alot of money each year, they also got a good deal with applebook Pro.
AMD is doing well on Gpu front with radeon on desktop, Mantle will certainly hit nvidia very hard in the next couple years by losing market share to AMD, that why Nvidia will keep these silly marketing BS, just to keep attention away from Mantle untill DX12 is here, they wil tell you nobody needs low level API, and when DX12 will come out, they will tell you there is nothing better than Low level API.
as for the reason why i support AMD instead of Nvidia is just because i see more benefit for the industry as a whole and for costumers, if AMD catch up on the market share of gpu desktops to nvidia, concurence will give everyone better prices and more innovation, and less inflated prices on 0 added value, or stupid eco systems for several hundred dollars.
i hate microsoft mostly because how they try to force you to change OS to keep up with directx evolution.
here nvidia maketing this driver as a magic potion for you to get if you upgrade to 700 series.

April 8, 2014 | 08:16 PM - Posted by Anonymous (not verified)

HARDOCP reported higher threshold then previous 335 and other drivers when they tested there 780 TIs

Went from 84c to 87c and 1006mhz to 1019mhz.

337.50 is a overclock driver.

Have to wonder if it was to combat the 295x2 release and why?

April 8, 2014 | 08:48 PM - Posted by Anonymous (not verified)

not an overclock driver i have speedboost disabled and see improvements.

April 8, 2014 | 11:57 PM - Posted by Anonymous (not verified)

you're simply experiencing the placebo effect of installing newer drivers.

April 9, 2014 | 09:12 AM - Posted by Anonymous (not verified)

you're an idiot

April 10, 2014 | 02:02 PM - Posted by Anonymous (not verified)

I have speedboost disabled also and i got nothing like they claimed

April 10, 2014 | 02:22 PM - Posted by Anonymous (not verified)

thats why they only recommend to use with GK104 and 110 older cards dont have speedboost.

Im a fan of nvidia but this is Such a shady move

April 8, 2014 | 09:10 PM - Posted by krvi (not verified)

My system: i5-2500k / H67 chipset / 8GB ram @ 1333MHz / single Nvidia GTX 780 reference clocks but custom fancurve.

I ran performance tests in 3DMark Fire Strike(not extreme), Unigine Heaven and Unigine Valley before (335.23) and after (337.50) updating.
Fire Strike score with 335.23: 7659
Fire Strike scores with 337.50: 7569 / 7599 / 7613
I didn't save the results from the unigine tests, but the difference was negligible; 50.0fps before and 49.9fps after driver update, in Unigine Heaven. The difference in Unigine Valley was also less than 1fps.
These 3 tests do not justify the hype in any way...

Although I have heard that the new driver increased CS:GO performance as much as 40-50fps, I haven't tried that myself. Maybe I should.

April 9, 2014 | 01:17 AM - Posted by wujj123456

Benchmarks shouldn't have bottlenecks and are supposed to stress GPUs and scale really well. So it's not surprising you are not seeing any improvement. Afterall, 3DMark is the best scaling application for SLI even before all the optimization.

I was going to completely ignore this driver since the fact Nvidia used 3930K in their slides speaks everything. Also they are comparing against an old baseline. Well, until they cite massive improvements for SLI. I guess that's still good for people like me running SLI. Just Nvidia really shouldn't have promoted this driver as a huge improvement.

April 9, 2014 | 02:36 AM - Posted by JohnGR

Nvidia customers hear about Mantle everywhere but they have to wait until the end of 2015 to get something equivalent(in reality to get Mantle with a name that Nvidia accepts, DIRECTX12).

But before DX12 we are going to see more than a dozen top titles with Mantle support. So Nvidia started this campaign about a "Wonder" driver promising to give almost Mantle performance in setups with Nvidia cards under DX11. The whole idea was to make it's customers feel better about their hardware.

This marketing comedy is going to continue in my opinion at least until DX12 comes out. Typical driver updates are going to be marketed as the second coming and typical game optimizations that give performance gains in specific games and/or under specific circumstances, things that we are all used to them, as something never seen before, worst as a proof that Mantle is not necessary for top performance.

PS And of course all these slides about the new drivers will use a 12 threads Intel cpu monster to prove that cpu bottlenecks are easily bypassed (LOL) with clever programming and without the need of another API(that's until DX12 comes out. At that time marketing will do a U turn and will start telling us how important is the new API).

April 9, 2014 | 06:17 PM - Posted by arbiter

Well results of mantle have only been marginal 10% boost, and if graphics comparison in BF4 mantle vs dx11, could be just a image trickery as mantle seems to less draw distance then dx, could be where the 10% comes from.

April 9, 2014 | 06:11 AM - Posted by Fane Tractor (not verified)

I got a GTX 760 and after updating the driver to this "new" improved version my fps actually has gone down instead of increasing in at least one game: Thief.

My system shouldn't have any CPU bottlenecks as I have a Core i5 3470 OCed to 3.7 GHz.

Also after updating fonts started to look a little foggy.

June 5, 2014 | 07:30 PM - Posted by ROdNEY (not verified)

Every PC has CPU bottleneck due optimized highly abstract DX API, no matter what CPU you have, the only effective way how to improve CPU performance is add cores since every new generation get only about 10% IPC boost and that is just too low to really see a difference. On the other hand from DX12 presentation there is 550% performance increase on main thread, something you wouldn't gain by HW development in decades. Plus say that my 4core CPU has no bottleneck within API that has virtually no multithreaded optimization and cannot use more than one core efficiently it is just ridiculous. It is the main reason why PC struggle with stable framerate so much and problem is not you CPU but API. With mantle/DX12 your CPU would get same results as 5times cheaper CPU, easily.

April 9, 2014 | 09:13 AM - Posted by Anonymous (not verified)

April 9, 2014 | 11:22 AM - Posted by Anonymous (not verified)

April 9, 2014 | 04:44 PM - Posted by godrilla (not verified)

The improvement claims was by enabling sli in Rome 2 total war which previously did not support sli. Lol

April 10, 2014 | 01:52 PM - Posted by Anonymous (not verified)

Im running 2 780 classified in SLI with 980x @4.4

My tomb raider benchmark comparison WOW i got a whole 2 FPS

Unique heaven i got the same score with both driver.
increase these claims are total bullshit and people who say otherwise prob has driver issues

April 11, 2014 | 08:16 AM - Posted by Klimax (not verified)

When somebody tests driver which makes CPU optimizations in scenarios which are not CPU bound, then of course they won't see much of improvement. And therefore readers should question judgment and knowledge and competency of said reviewers.

There is no controversy over drivers, just bad tests. And author here failed to note that.

(Looks like number of good servers on tech just got another trimming...)

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

This question is for testing whether you are a human visitor and to prevent automated spam submissions.