The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Installed just fine, even better than the latest Hotfix driver as these do not pop out a warning about installing the driver which you have to accept as in the hotfix drivers.
I heard this driver does not include the DPC Latency fixes for Pascal cards. I wonder if mobile Pascal cards will also be affected, hope they figure it out until launch...
Nice results @ stock. The 980 (desktop) is already 2.5x faster than my 680/K5000M, can't wait to get my hands on a new notebook (even though I'm still in love my trusty x7200)!
I think that non-counting/almost 5k score is torturing me more than you...my SSD is so full and cluttered that I don't find the time to set it up and make a proper run...
Beta and hotfix drivers are never approved by 3DMark. It doesn't necessarily mean the score is invalid, just that the guys that work there are not burning any calories to validate them and they won't validate your score. Kind of makes sense because they would waste an inordinate amount of time validating drivers that are not WHQL certified, which is basically meaningless. WHQL certification is about as worthless to me as an old piece of gum stuck to the bottom of one of my shoes. I don't consider WHQL certification to be important for any drivers... video, WiFi or anything else... just doesn't matter to me. It's just a money-making gimmick for the Redmond Mafia.
What would be better is if Futuremark would remove the flag and add a message to say "Valid Score - Driver not WHQL Certified" or something along those lines. Too many noobs in the world think that Futuremark flag means the score is not accurate when all it means is the video drivers were not validated by Futuremark.
Having a "Graphics driver not approved" flag from Futuremark is acceptable to HWBOT. Here is what HWBOT.org has to say about it.
A – DRIVERS Unlike Futuremark, HWBOT chooses to allow people to use whatever driver they feel is best for their videocard, be it an official version of the videocard driver. Pre-modified performance drivers, such as Starstorms, are not allowed for the simple reason that the HWbot crew is unable to be certain that there are no cheats implemented in those drivers. There is one exception: the Omega Ati drivers. These are allowed because they give the user the opportunity to unlock for instance pipelines on their video card.
Official drivers causing a benchmark not to render correctly are not accepted at HWBOT. Incorrect rendering is defined as clear visual indications that parts of the benchmark are either not visible or in someway improperly rendered. The main effect of this incorrect rendering is an out of line performance increase due to this rendering flaw. Do note that this rule will not be actively applied to hardware and drivers released prior the year 2010. Also, the HWBOT staff will not be actively testing out different combinations of drivers, hardware and operating systems to ensure propper rendering. We hope that the overclocking community will be able to help us with this list and find problematic drivers.
Graphics driver is not approved This message is displayed if your benchmark result was obtained using a set of graphics card drivers that are not (yet) approved by Futuremark.
In order to ensure that drivers do not breach our Driver Approval Policy, Futuremark tests video drivers and validates the results. Drivers are sent for testing and approval as soon as they are published. This process takes normally no more than 3-4 working days after driver release. Any results uploaded during this time initially shows as "not approved" but will turn to approved results when the driver approval process is complete for the specific driver version - you do not need to do anything, your result will transform into approved result in time.
Results obtained using beta, "hotfix" or otherwise non-WHQL drivers will also show this message and normally non-WHQL drivers are not approved for use with our benchmarks unless specifically requested by the graphics card vendor.
If there are no WHQL certified drivers yet for your video card/chip - usually if you buy your hardware as soon as it becomes available - you will have to wait until the first WHQL approved drivers with support for your card are made available for valid results. You can benchmark even with beta drivers - the results just won't be shown as valid and the result page will show that the drivers are not approved.
These are the types of flags that indicate potentially inaccurate benchmark results:
Non default settings were used in benchmark. Score could not be calculated. This warning means a required test was not included in the benchmark run. 3DMark and PCMark only return an overall score when using one of the default benchmark presets. You can use sub-scores (Graphics, Physics etc.) for comparisons but you must take into account the modifications done to the settings and how they alter the test workload.
Non default settings were used in benchmark. Result is not comparable to other results. This warning is shown when you run the benchmark using custom settings. 3DMark and PCMark Advanced and Professional Edition benchmarks offer a wide range of options to customize a benchmark run for detailed investigation of system performance. Results from such custom runs are not comparable with results from other systems, however, unless the exact same settings are used in both cases.
Usually custom settings are used for investigating how the performance of a single system changes when varying the benchmark load. For this purpose, comparison is still possible by directly comparing scores from individual tests, such as the Graphics Score and Physics Score in 3DMark 11 for example.
Time measuring inaccurate This message indicates funny business with the system clock during benchmark run. In most cases, this means that, no, you cannot cheat in 3DMark by adjusting Windows timers during the benchmark run or otherwise tampering with the measurements done by the benchmark. If this message persists and you have not done anything out of the ordinary, it may indicate a hardware issue with the real time clock in your system or the presence of a background program that somehow twists the time-space continuum of your operating system in such a way that this anti-cheat detection is tripped.
The last one is what makes Windows 8 and 10 sometimes not honored by HWBOT.org because the score is actually calculated incorrectly with CPUs older than Skylake.
The biggest issue is the benchmarks run with non-WHQL certified drivers are hidden and do not show up in 3DMark search results unless you remove the check mark in the box.
Yeah, that hit me by surprise as well. 3DM11 however works better with W7. DX based performance was way better with W8.1 and W10. I had installed all 4 (W7,W8,W8.1 and W10) to run some VR tests at work, and did some benching on the side. FireStrike and Skydiver were way better on W8.1 and W10. Time Spy however didnt like a stock clean install of Windows, no matter what version and seemed to have an affinity towards my garbage daily use install.
Of course... a "
garbage daily use install" is what most gamers have for an OS. That's probably a good reason for them to call 3DMark software suite the "The Gamer's Benchmark" and Time Spy DLC the "DirectX 12 benchmark for gaming PCs" LOL. It's for gamers, and reviewers focused on highlighting the ability of a particular product to play games, including BGA jokebooks with decent video cards, but not actually designed to be a top notch tool for overclocking performance enthusiasts. While it can be used for that, it does not appear to be what they had in mind. Really no different than Passmark Performance Test, which is geared toward establishing a "normal" baseline for their reference database used to compare belly-button hardware configurations running at stock clock speeds. (I have actually had Passmark "invalidate" some of my extreme overclocked benchmarks because they were too out of scope with the average for the same hardware.)
I guess as long as it is fun and we enjoy running the benchmarks, that is the most important thing. I know I enjoy benching and it doesn't really matter to me how they intended for the software to be used.
So how are these for gaming? Any noted increase in CPU temps between this and 365.19? 362.00 had hotter CPU temps for me for some exceedingly odd reason.
It's not exactly that I want to move off of 365.19, but if I ever want to use Vulkan in something I will have to, and it'd be nice to know if these are something to consider.
Fox Have you looked at your physics score with latest 3DMark version (The Time Spy version). It is crippled by Futuremark!! 3Dm11 isn't crippled, only the 3DM suite.
And the XenForo software problems will never end!!
Has been in contact with Jarno KOKKO (Futuremark). The latest version of 3DMark is a disappointment and is buggy. Several of you should contact Futuremark and complaining!! I dont like the new Suite. It's more like a big mess!! And very very UGLY. More like the new mess from M$
Try the latest 3dmark Version. Run bench and compare Fire Strike score with a older one. Jaro Kokko on Futuremark confirmed to me in mail that physics in custom settings have a bug!! Probably also in full test.
Yes, it should count for HWBOT. Unless there is something specific to this driver I haven't heard about yet, they don't care if Futuremark validates the drivers.
The entire 3DMark suite has been buggered up since they updated it with Time Spy.
I agree... both are awesome and I like both more than the 3DMark suite. I do like Sky Diver though. That actually taxes the CPU more than Fire Strike does.
nVidia's site is really really really really really really really really really really slow...
Should have results in 20 minutes or so... just ran Time Spy with 368.20, will report back with a comparison between 368.20 and 369 for Time Spy and 365.19 for everything else. Only reason Time Spy was 368.20 was because Microsoft forced me into 368.20 with the Anniversary Update.......
Still too hot to use the laptop so it will all be my 780 Ti.
Problems? See this thread at archive.org.
Indeed.
![GeForce GTX 980 (369.00) [W10].png](images/storyImages/GeForce GTX 980 (369.00) [W10]-preview.png)
