Fury X vs. Titan X benchmark

Status
Not open for further replies.
razor1 said:
Its not cheating its default driver settings. It has been like that since the 6800 series, They changed the names of things in their control panel though. Performance is now called quality, and quality is now called high quality.

The drivers are set at let the application decide by default.

Also changing the quality slider also adds AA.
Click to expand...


What? No it doesn't. Unless you are talking about AMD, which still I don't think it does.

razor1 said:
I think thats why reviewers tend to use the in game settings vs the control panel settings.

So keep the drivers at default (which is application controlled on both IHV's) and what ever the application does is what is done.
Click to expand...


I think what you fail to realize is that the texture filtering setting in both NVidia's control panel and AMD's CCC has nothing to do with any setting in games. Even if a game uses 16xAF, if the driver is set to "Quality" instead of "High Quality" you are going to see shimmering and blurry textures in game.


Anytime I install a new NVidia driver, the first thing I do is set the global profile to "High Quality" for Texture filtering, from "Quality". This turns off all NVidia "optimizations".
 
pandora's box said:
What? No it doesn't. Unless you are talking about AMD, which still I don't think it does.


I think what you fail to realize is that the texture filtering setting in both NVidia's control panel and AMD's CCC has nothing to do with any setting in games. Even if a game uses 16xAF, if the driver is set to "Quality" instead of "High Quality" you are going to see shimmering and blurry textures in game.


Anytime I install a new NVidia driver, the first thing I do is set the global profile to "High Quality" for Texture filtering, from "Quality". This turns off all NVidia "optimizations".
Click to expand...

The global setting by default is Application controlled.

http://i.imgur.com/TgbMpai.jpg


Now the other part is here

http://i.imgur.com/elHmcYc.jpg

it is set to quality, but the Application overrides this setting.

If you want to change this global setting, and change to high quality, which is done here

http://i.imgur.com/xqrjF0d.jpg

you will turn on MSAA

If you just want to force only AF by the driver go to manage 3d settings global settings, and then change Anisotropic filtering from Application-controlled to what ever level of AF you want.

Using the global slider not only uses AF but also forces AA. This is why reviewers I think don't use the slider because the changes are no longer under their control.
 
Last edited:
razor1 said:
The global setting by default is Application controlled.

http://i.imgur.com/TgbMpai.jpg


Now the other part is here

http://i.imgur.com/elHmcYc.jpg

it is set to quality, but the Application overrides this setting.

If you want to change this global setting, and change to high quality, which is done here

http://i.imgur.com/xqrjF0d.jpg

you will turn on MSAA

If you just want to force only AF by the driver go to manage 3d settings global settings, and then change Anisotropic filtering from Application-controlled to what ever level of AF you want.

Using the global slider not only uses AF but also forces AA. This is why reviewers I think don't use the slider because the changes are no longer under their control.
Click to expand...


Interesting. I've never used the "adjust image settings with preview option". I just tried it myself (dragged the slider to balanced, then back to quality. Sure enough 4xAA was enabled when I went to "Manage 3D Settings".
 
Sharps97 said:
I think the key take away would be to (at default settings) compare the image quality between both cards and then make a judgement call (or at least provide comparative images). I know that [H] has done this in the past - for a long time it didn't matter as both companies default settings were equivalent. Sounds like another round of comparisons might be warranted as NVidia seems to be getting the "itch" again...
Click to expand...

+1 the images from both cards needs to be shown. Otherwise FPS numbers are meaningless. Either team could cheat and no one would be the wiser. Image quality used to be a big deal but no one is talking about it anymore.

Going back and forth from nvidia and ATI/AMD through the years I've always noticed a big improvement in 2D/3D quality going from nv to ATI/AMD.
 
So at minimum its not strictly apples to apples and at worst its default settings cheating. Since when has HardOCP ever trusted anyone so much? Brent, nobody is blaming HardOCP over this but don't sweep it under the rug by saying "oh well, we trust (insert company name here) to do the right thing with default settings"? At least look into it.
 
Fans are still trying to find a reason to buy Fury X.
The argument is image quality. The next will be crossfire scaling.

But the majority still don't want it.
 
WorldExclusive said:
Fans are still trying to find a reason to buy Fury X.
The argument is image quality. The next will be crossfire scaling.

But the majority still don't want it.
Click to expand...

Yeah, but mostly because of anti-AMD bias and double standards. Image quality and crossfire scaling are both extremely important, especially image quality. But since nvidia has worse IQ generally, no one makes a big deal out of it. Although review sites used to take it into consideration years ago.

Not many people truly give both sides a chance, usually they will pick one side and stick with it, and that side is usually nvidia, for the reasons pointed out above.
 
WorldExclusive said:
Fans are still trying to find a reason to buy Fury X.
The argument is image quality. The next will be crossfire scaling.

But the majority still don't want it.
Click to expand...

And yet it continues to sell out no matter how much some wish it wasn't true. ;)
 
WorldExclusive said:
Fans are still trying to find a reason to buy Fury X.
The argument is image quality. The next will be crossfire scaling.

But the majority still don't want it.
Click to expand...

Oh but it is users OWNING both cards that are finding this. Not some witch hunt as you allude. Honestly where is the outcry for neutrality and fair and equal footing? What good are benchmarks if either is playing with settings to gain an edge?

At this point we should in the least being requesting more indepth coverage and interest in what the facts are. But blowing this aside does none of us any justice. I want to know if either is doing this and to what degree. Looking at my CCC for AMD I wonder what really is going on. Sucks its a holiday weekend, Wife gonna drag me all over the place so testing this will have to wait a little.
 
Brent_Justice said:
So, double our workload, and artificially manipulate the experience to our own.

Yeah, no.

I'll stick to letting AMD and NVIDIA provide the gameplay experience they want for their own cards, and we will evaluate that and see which one is better and how they compare.

The reason for using default driver settings has been stated, please reference above posts.
Click to expand...

You need to make sure you explain why the AMD cards are slower by 10% then. Nvidia drivers are lowering quality by default to gain the upper hand in benchmarks which is downright underhand. The end user perception is that Nvidia cards are faster by a large margin. This obviously can make or a break a card such as the Fury X which should and is beating a 980Ti in many reviews but losing in others. HardOCP needs to look at more than just fps in future reviews such as the Fury Pro coming up in the next couple of weeks.

A new article looking at this driver discrepency is in order. HardOCP may be favouring Nvidia over AMD so probably won't do one but someone eventually will.
 
Last edited:
The main difference is AF, the screenshots that were provided for BF4 on OC forum, was when the global 3d settings were forced to high quality, that also forces x4 AA.

AF at x16 only has a 2% performance hit maximum on most of nV's cards.
 
So reading the thread I think a lot of people are jumping to conclusions. Using horribly compressed low ass bitrate Youtube video as proof kind of throws out Sleeping Dogs but those BF4 screens are interesting. I don't mean this to sound tinfoil hat-ish but both of those games are AMD titles. So that brings to mind a big list of questions that need clear answers before anything is solid:

1. Is it true with Gameworks games?
2. Is it true with neutral games?
3. Is it true with all the AMD games?
4. Is it an issue specific to some cards?
5. Is it an issue related to specific drivers?
6. Is it different with different issues?
7. If this is a common problem why hasn't it been noticed by the hundreds of hardware reviewers around the world?

Two games (one of which is hard to take as solid proof) does not a conclusion make.
 
Its not cheating its default driver settings.

It is cheating if Nvidia knows that the reviewers will not change the settings and AMD does not know what Nvidia is up too.. So 10 to 15% preformance must be taking away from any Nvidia default settings review out there..

Oh and I have a 770GTX under the hood for the flame boys..
 
razor1 said:
The main difference is AF, the screenshots that were provided for BF4 on OC forum, was when the global 3d settings were forced to high quality, that also forces x4 AA.

AF at x16 only has a 2% performance hit maximum on most of nV's cards.
Click to expand...

I found that draw distant also effects FPS so you may want to see if Nvidia is playing with that also..
 
fanboy said:
I found that draw distant also effects FPS so you may want to see if Nvidia is playing with that also..
Click to expand...


I didn't see that in any screenshots, I know someone mentioned it seemed like that, but most of the issues seemed like AF not being turned on. As the mip map levels changes with in an engine, without AF, objects that are using lower res mip maps, get washed out and blurred to much more of a degree then the original texture size, and this is actually do to the adaptive filtering algorithms the IHV's use, which is what I think the person was talking about.
 
fs123 said:
A new article looking at this driver discrepency is in order. HardOCP may be favouring Nvidia over AMD so probably won't do one but someone eventually will.
Click to expand...

[H] favored nVidia because that's where the data led them. Not because they are necessarily biased. Many other reviews have come to the same conclusions as [H]. Including myself personally. I skim data and trained myself to ignore opinions due to the technical nature of my job. I went back later to read conclusions and it matched up with my own personal take.

fanboy said:
Its not cheating its default driver settings.

It is cheating if Nvidia knows that the reviewers will not change the settings and AMD does not know what Nvidia is up too.. So 10 to 15% preformance must be taking away from any Nvidia default settings review out there..

Oh and I have a 770GTX under the hood for the flame boys..
Click to expand...

If they are intentionally decreasing draw distance or something of that sort that would make me pause. Some textures being a little off... That wouldn't convince me. If the problems are consistent across a lot of games I'd be disappointed.

I'll wait for an official review from one of the good sites.
 
WuTangClam said:
Yeah, but mostly because of anti-AMD bias and double standards. Image quality and crossfire scaling are both extremely important, especially image quality. But since nvidia has worse IQ generally, no one makes a big deal out of it. Although review sites used to take it into consideration years ago.

Not many people truly give both sides a chance, usually they will pick one side and stick with it, and that side is usually nvidia, for the reasons pointed out above.
Click to expand...

It depends on default settings. If Nvidia's default settings are lower but if raised it on par with AMD, Nvidia is not producing a inferior image, it's just set lower by default. It's just like running a game with no AA vs AA. The image may be worse, but both are capable of the same quality when set to the same AA settings. Now if speed is being debated when set to the same settings, you can have a discussion, but there's a lot more than speed what matters in a buying decision. The Fury X has shutout many potential owners simply on hardware features alone.

Bowman15 said:
And yet it continues to sell out no matter how much some wish it wasn't true. ;)
Click to expand...

Limited stock doesn't mean anything. I can't buy one now, but I can buy a 980Ti, which means they are losing sales. I can't buy one because it doesn't support HDMI 2.0, DVI or I may not want an AIO cooler.
Losing sales is never good, no matter what the perception of "sold out" means.

durquavian said:
Oh but it is users OWNING both cards that are finding this. Not some witch hunt as you allude. Honestly where is the outcry for neutrality and fair and equal footing? What good are benchmarks if either is playing with settings to gain an edge?
Click to expand...

It is a witch hunt. People are demanding Bret and other sites to revise their reviews based on a variable IQ setting.

If I turn on my PC, download the driver and that's what I have, then that is the setting that should be reviewed.
Are we comparing OC vs Stock cards and declaring winners? Or are we comparing stock vs stock?

The moment one setting is tweaked from default, the review is no longer valid.
Bret knows this, and will not invalidate his review to appease the witch hunt.

Even if the Fury X was on par after the IQ settings were increased on Nvidia, the hardware alone is still a no buy for many.
There's a lot more working against Fury X than for it.
 
WorldExclusive said:
Limited stock doesn't mean anything. I can't buy one now, but I can buy a 980Ti, which means they are losing sales. I can't buy one because it doesn't support HDMI 2.0, DVI or I may not want an AIO cooler. Losing sales is never good, no matter what the perception of "sold out" means.
Click to expand...

Of course it means something. No matter what you think people still want this card. Saying it isn't so doesn't make it true. The 980 Ti was limited in stock at first as well. WTF people?
 
Bowman15 said:
Of course it means something. No matter what you think people still want this card. Saying it isn't so doesn't make it true. The 980 Ti was limited in stock at first as well. WTF people?
Click to expand...


AMD stated the first month will be low stock, lets see what happens in the next two weeks.
 
Yeah, perhaps they are both capable of the same quality, but if reviewers don't show the images of each respective card, we can't be certain that any apples to apples comparison is truly apples to apples if the default setting for one team is designed to give a lower quality image in exchange for higher FPS. We should go back to the old days where IQ was compared in addition to FPS.
 
Brent_Justice said:
So, double our workload, and artificially manipulate the experience to our own.

Yeah, no.

I'll stick to letting AMD and NVIDIA provide the gameplay experience they want for their own cards, and we will evaluate that and see which one is better and how they compare.

The reason for using default driver settings has been stated, please reference above posts.
Click to expand...

Oof. Bitching about doing more work? So you can offer less information to the readers?

Good day, sir.
 
Last edited:
Brent, Think you could do a screencap to add to the graphs for each game just to show IQ? No need in my mind to redo every benchmark or touch the driver settings in any way. Just show people the difference or no difference between the cards. I miss that from the old reviews in years back and wouldn't add to much work if you did it at the beginning of the gameplay session from a save.
chris
 
I feel Brent is right on keeping the settings at default and just changing in game settings. But it would not be asking much to take screen shots and have them posted right along with the graphs of fps performance. That then ends the IQ battle because everyone can look at the screen shots and decide for themselves which is more important to them.
This round AMD is getting my money again for graphics cards because no one has a card with that much performance in that size of a card. The Fury X is perfect for sff systems. The only real complaint I have is I wish they would of done the same connector scheme as the HD5800 series and the R9 290 series.
 
When we do specific game performance and IQ reviews, that is relevant, and we will continue to show IQ shots in those reviews and look at card differences.

In the main reviews we do so much gameplay evaluation now it would be a lot to add on.

I will make sure we look for card differences in our game reviews, the next game on my list is The Witcher 3, it needs a full review.
 
WorldExclusive said:
It is a witch hunt. People are demanding Bret and other sites to revise their reviews based on a variable IQ setting.

If I turn on my PC, download the driver and that's what I have, then that is the setting that should be reviewed.
Are we comparing OC vs Stock cards and declaring winners? Or are we comparing stock vs stock?

The moment one setting is tweaked from default, the review is no longer valid.
Bret knows this, and will not invalidate his review to appease the witch hunt.

Even if the Fury X was on par after the IQ settings were increased on Nvidia, the hardware alone is still a no buy for many.
There's a lot more working against Fury X than for it.
Click to expand...

The review is not valid as it stands now. They are comparing a Nvidia card that is doing less work due the default driver settings with an AMD card that is displaying everything as it's meant to. Is that fair in your opinion?
We all know the cards can do the same thing but intentionally setting things lower is only done to acheive one thing, the better fps in benchmarks and reviews which is what most buyers will look at, not image quality.
 
I agree with Brett on this one. You can't be getting the reviewers to go in and tweak the driver default settings. If you feel AMD is getting a raw deal because of default driver settings, then you guys write to AMD and tell them to match the default driver settings of Nvidia. AMD is choosing to make IQ a priority over speed, and it looks as if Nvidia is prioritizing speed over quality. Those are their own choices and they'll have to live with the results. Most gamers aren't going to tweak at the driver level anyway they are only going to touch the in game graphics options usually only going by the preset Low/Med/High/Ultra and maybe customizing the AA levels, so that is what reviewers should be doing also because that is what is what is going to be presented to the buyer of the card in question. Not some driver specific optimized settings, but the actual setting defaulted to by the driver as it was installed.

I do agree though that screen shots of what was shown side by side would be appropriate for the reviews.
 
wantapple said:
Im going to buy fury x anyway... Way better image quality
vG2RNKo.jpg
Click to expand...

Yea I dont think graphics setting such as "Ultra" Would look different on different hardware, Could just be the frame you used
 
Witcher 3 is one of the games nvidia actually recommends using the in driver settings for af to improve IQ so that would be a great one for some screen shots of.
Razor1- I did not know he was using the slider to set up those tests so that is really null and void now. I wish one of us at [H] had both cards to do some comparisons like we all were hoping for in the other thread where he had to rma them.
 
yeah just saw his latest video with same settings on both cards, Fury X takes a larger hit then the Titan X.

I do remember in the past, some games do give performance issues with using drive forced settings this might be something similar.
 
Bowman15 said:
Of course it means something. No matter what you think people still want this card. Saying it isn't so doesn't make it true. The 980 Ti was limited in stock at first as well. WTF people?
Click to expand...

Every high end GPU is limited at release. Out of stock doesn't mean you're moving a lot of units.
I can be out of stock of 30K units sold, or I can be out of stock of 100K units sold. One made more money than the other.

Demand means nothing if you can't supply it.

fs123 said:
The review is not valid as it stands now. They are comparing a Nvidia card that is doing less work due the default driver settings with an AMD card that is displaying everything as it's meant to. Is that fair in your opinion?
We all know the cards can do the same thing but intentionally setting things lower is only done to acheive one thing, the better fps in benchmarks and reviews which is what most buyers will look at, not image quality.
Click to expand...

You're still asking to go from stock settings review, to a variable settings review.
There's no way to set both control panels to render the same quality level to accurately come to a conclusion which card is faster.
Both sides will continue to cry that the AA was too high, or Triple Buffering should be turned off etc.

Nvidia and AND handle IQ differently. Just saying re-run the test with a new variable doesn't mean you will get a single definitive result.
 
WorldExclusive said:
Every high end GPU is limited at release. Out of stock doesn't mean you're moving a lot of units.
I can be out of stock of 30K units sold, or I can be out of stock of 100K units sold. One made more money than the other. Demand means nothing if you can't supply it.
Click to expand...

You words mean absolutely nothing if you don't know how many cards there actually were on release. My point is you are just pulling numbers out of your ass to justify your own argument as you do not know. Did AMD kick your dog, get your sister pregnant or are you just a fan boy?
 
Last edited:
WorldExclusive said:
You're still asking to go from stock settings review, to a variable settings review.
There's no way to set both control panels to render the same quality level to accurately come to a conclusion which card is faster.
Both sides will continue to cry that the AA was too high, or Triple Buffering should be turned off etc.

Nvidia and AND handle IQ differently. Just saying re-run the test with a new variable doesn't mean you will get a single definitive result.
Click to expand...

So if IQ doesn't matter, whats the point in matching in game settings, why not just show whats fastest? There is no reason to try and match "Apple" -> "Apple" comparison for FPS if the IQ is different, because you are really doing Apple -> Oranges at that point.

There are three options here:

1) Nvidia and AMD are equal IQ at default settings - can be matched using in game settings

2) Nvidia and AMD are not equal default settings - can be matched using driver settings and in game settings

3) Nvidia and AMD are not equal default - use same in game settings but note differences and don't just show FPS #s since they aren't valid.
 
fs123 said:
Maybe trying to build his post count up. Do HOCP give any benefits if you have a high post count?
Click to expand...
Well its the type of posts you see when someone is spamming to get to FS forums but he has only made like 7 posts in over a month so its odd. :confused:
 
Status
Not open for further replies.
Back
Top