• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

BF4 IQ Comparison - Nvidia control panel settings

The only ones talking about AMD here are the Nvidia "shills" since they love that word. He noticed IQ differences and also a fix for them in the nvidia control panel.
I posted it here so people could help investigate the issue, my nvidia card is a 500 series so would be invalid to use if it was an issue with 900 only.

There was no fix, the guy that originally did the screenshots and videos, forced AF to x0 in the control panel. That's why AF wasn't turning on when the application wanted AF.
He screwed up either by accident or on purpose.
 
There was no fix, the guy that originally did the screenshots and videos, forced AF to x0 in the control panel. That's why AF wasn't turning on when the application wanted AF.
He screwed up either by accident or on purpose.

I'm saying he fixed it by forcing it on in control panel, I don't know why it was wrong before and thus why I created this thread: to figure it out.
 
I'm saying he fixed it by forcing it on in control panel, I don't know why it was wrong before and thus why I created this thread: to figure it out.


no one else was able to duplicate the it if it was driver bug, So that kinda of tells us, he doesn't know how the driver settings work, if he did it by accident.
 
no one else was able to duplicate the it if it was driver bug, So that kinda of tells us, he doesn't know how the driver settings work, if he did it by accident.

The problem is the original guy claimed he never touched the default driver. Everyone thus far claiming they don't have the same issue has always set their own settings in NVCP. The only way to be sure is to remove drivers and get a fresh install so any changes before are not still saved. Then run the test. So far I have only seen definitively that the 700 series doesn't do this. I have yet to see a single 900 series tested correctly as of yet.
 
The problem is the original guy claimed he never touched the default driver. Everyone thus far claiming they don't have the same issue has always set their own settings in NVCP. The only way to be sure is to remove drivers and get a fresh install so any changes before are not still saved. Then run the test. So far I have only seen definitively that the 700 series doesn't do this. I have yet to see a single 900 series tested correctly as of yet.

Essentially we don't know whether GTX 980. 970. 980 TI and Titan X were tested with correct settings. I figure reviewsites do quick swaps with driver unistalls/installs they dont bother to double check Nvidia IQ settings.
 
I want to be clear, BOTH AMD and NVIDIA should be looked at with scrutiny. This isn't a call to just look at NVIDIA. We should all be a bit concerned and equally so when a lot of how we buy is based on these reviews. I am more concerned with fact than any ego-contest.
 
Everyone thus far claiming they don't have the same issue has always set their own settings in NVCP.
This isn't true though. I had already linked a few users with a 900-series cards who explicitly reset the NVCPL to defaults (which results in ignoring NVCPL ignoring all users settings), and there was no problem:
http://forums.anandtech.com/showpost.php?p=37533420&postcount=83
http://forums.anandtech.com/showpost.php?p=37533433&postcount=91
http://forums.anandtech.com/showpost.php?p=37533331&postcount=62

The only way to be sure is to remove drivers and get a fresh install so any changes before are not still saved. Then run the test. So far I have only seen definitively that the 700 series doesn't do this. I have yet to see a single 900 series tested correctly as of yet.
This seems to be a bit of stretch, since there is no difference in NVCPL profile defaults between GPU models. If behavior of NVCPL was already proven with a 700 series, that should be enough. You don't even need to re-install to wipe out all user profile settings and restore a fresh state. If you click the 'Restore' button on the Global Settings tab, NVCPL will do this for you.

I don't think some of you realize how much of an eyesore and instantly obvious having AF disabled is in any game. If this was in anyway a wide-spread problem, both the BF4 forums and NVIDIA forums would have been blowing up for months now, yet that never happened. Not even trashy rumor mill sites picked this up, since there is no story if this cannot be reproduced. The only people suggesting this imaginary issue needs to be looked into, are users on forums who do not even own an NVIDIA 900-series card.

Even more so, reviewers like Brent here at [H] have stated that this BF4 AF disabled issue doesn't exist and difference between Quality and High Quality texture filtering on NVIDA is being massively overstated. His reviews thus far should likely fulfill your requirement of performing clean driver installs which switching between AMD/NVIDIA GPUs on his test bench, not to mention always doing extended playthroughs. There is really no controversy here, since both AMD's and NVIDIA's default texture filtering quality is actually quite good and it's been this way for quite a few years now. GPUs are so fast now, there only negligible performance impact from high levels of AF, to the point of it being almost free.

As stated previously, nobody thus far has been able to reproduce any kind of AF disabled issue in BF4 when testing properly. This strongly suggests that the original user's problem is either user error, not knowledgeable enough to properly troubleshoot the issue, or some strange bug or corruption specific to his system.

I want to be clear, BOTH AMD and NVIDIA should be looked at with scrutiny. This isn't a call to just look at NVIDIA. We should all be a bit concerned and equally so when a lot of how we buy is based on these reviews. I am more concerned with fact than any ego-contest.

This is already the case though. Anytime there has been an improper driver optimization which increases performance greatly at the expense of quality it quickly gets called out, and in extreme cases even by AMD and NVIDIA themselves. Though there is a fine line here between scrutiny of driver optimizations which are not bugs, and scrutiny of unintentional driver regressions or bugs. The former is generally can result in positive and productive discussions, while the latter does not from lack of intent.

Ever since the initial comparison video was posted, it was already clear this could not possibly be driver optimization. At NVIDIA defaults, the driver honers the Direct3D AF setting enforced by the game itself, nothing is overridden. If this was not happening, it would have been an honest bug and nothing to overly scrutinize anyone over. Mistakes like this with unintentional regressions occurring in various drivers and game patches happen all the time, and they eventually get fixed. In this case though, the issue from the video in OP hasn't even reached the threshold for being a reproducible bug which warrants any discussion at all...

Before this thread was even created, 900-series users around the web had already debunked the idea that this was even some kind of wide-spread BF4 bug. The issue in question ended up being neither a optimization nor a bug which anyone outside of the original user could reproduce. At that point, anyone focusing on the already invalidated video in the OP as reason for needed scrutiny, or ignoring the testing already done to debunk it have been doing nothing but spreading dishonest FUD about NVIDIA, even if unintentionally.

This thread was on its way to die a quick death, yet durquavian, you personally revived it with a post which was seemingly nothing but a hidden agenda to continuing the spread of FUD. Even if you are personally in denial about the honest testing performed by users of NVIDIA cards, you should have realized the negative connotation that reviving this thread would have. It seemed rather disingenuous, even if you do have an honest concern.

I'd be happy to have a mature discussion which scrutinizes how AMD and NVIDIA go about performing optimization in their drivers, and if this does or doesn't have a direct correlation to in-game quality, but this thread which has such a video in the OP which entourages such a negative discussion is not the proper venue for it. If you desire such a discussion which is untainted by FUD, I'd suggest you create a new thread which doesn't focus on the videos or testing done by that single user, who at this point has lost all credibility as a trustworthy source.
 
Last edited:
This whole "controversy" just shows how pathetic AMD shills and fanboys have become. The fury x got soundly beat by nvidia so now they resort to making up lies about image quality differences.

Yeah it sounds so far fetched it has never ever happened before ...
 
Yeah it sounds so far fetched it has never ever happened before ...

But what part of "hasn't happened since G70" don't you understand?

Yes, it did happen TEN YEARS AGO. No, if a game suddenly stopped running HQ AF, I'd definitively notice. I owned an Nvidia 7900GT, so I know the impact of lower-quality AF. I bought an HD 4850 to smack Nvidia around for that major fuckup :D

But I've been running Nvidia without any AF issues for the last 5 years now! Nvidia fixed their AF quality with G80, and have since cranked it up by default. Water under the bridge, eh?
 
Last edited:
Ever since the initial comparison video was posted, it was already clear this could not possibly be driver optimization. At NVIDIA defaults, the driver honers the Direct3D AF setting enforced by the game itself, nothing is overridden. If this was not happening, it would have been an honest bug and nothing to overly scrutinize anyone over. Mistakes like this with unintentional regressions occurring in various drivers and game patches happen all the time, and they eventually get fixed. In this case though, the issue from the video in OP hasn't even reached the threshold for being a reproducible bug which warrants any discussion at all....

The person that posted the video owns and uses a Titan X (OCed at that) and found the IQ differences and thus made the comparison video. He thought his drivers were clean but something must have been bugging them out to cause the issue of AF being removed.

I was asking for other people to confirm or refute it as a normal setting or something wrong. Other people have found that it was just an issue with his setup for some reason, now he's just manually forcing AF and its working fine for him, and works like it should for other people, so no issue.
 
He's already posted that his monitor got flagged as a generic PnP monitor. RTFT already. It was only 7 pages long...

EDIT: Now 8 pages. Gee that took me all of three extra minutes to read...
 
Last edited:
He's already posted that his monitor got flagged as a generic PnP monitor. RTFT already. It was only 7 pages long...

EDIT: Now 8 pages. Gee that took me all of three extra minutes to read...

Having your monitor flagged as generic PnP shouldn't effect AF settings, I'd bet most people have theirs setup that way. I also addressed this a long time ago. Please let the thread die as he already found out it was his issue and not a default issue.
 
Hey Guys, long time reader of [H] but first time poster. I will try and clear a few things up.

The initial video I did was full of errors and mistakes and none of them were intentional. Most were down to it being my first time of using a capture card but anyways, here goes:

Ran DDU, cleaned Nvidia drivers and installed AMD drivers - Recorded a set run on BF4.
Ran DDU, cleaned AMD drivers and installed Nvidia drivers - Recorded the same run.

Comparing IQ I could see something was up and AF wasn't there correctly. Others said that they didn't have the same problems that I was seeing on their system, so I set "prefer max quality" and I then had the same image quality as the AMD run. I have since redone the video with everything set correctly. I also noticed that colours we're different on both the TX and FX and after some head scratching on what I needed to set on the NCP, I couldn't find anything, so went into windows display properties and found that the capture card was set to record the generic plug and play monitor, so I set the profile to the ROG Swift and bingo, matching colours.

I have also clean installed Nvidia drivers since and I got the same AF problems that I had previous but as others are not getting it, it is clearly something up my end and possibly clean installing windows would sort it.

But to confirm, I have all video's set to "prefer max quality" and I have the monitor, capture card and other noob errors all sorted now.

A massive apology for the confusion and it was all down to me being a massive noob but if anyone would like to check my vids now, they will see that the latest batch and a few of the redone ones are spot on.

I hope this clears up things.
 
Back
Top