AMD & NV image quality comparison

[wand3r3r]

The point would be to investigate why/how/when some people are getting reduced quality.

If benchmarks are done with a 10% performance boost they are invalidated.

It's just PC components which none of us should care about other than the integrity of the data, without ulterior motives.

 

[TeknoBug]

Between a R9 290 and GTX970 I saw no difference in DX11 (and OpenGL in Linux), but Mantle had a different outcome however has been addressed since then and is identical to DX11 now.

 

[gamervivek]

All you needed to do was to check my video's... This has been there for some time.

https://www.youtube.com/watch?v=_r5pGUKOh1I

That is DX11 against Mantle at 1080P (both on the Fury X). First you tell me I need 200%, then you tell me I need 1080P, it seems no matter what, you have it that I am doing it wrong, so what next? You make out you know me but really you don't. I have A titan X, A fury X and a 7950 but your personal agenda is getting in the way. Not sure why I replied, as this will get picked apart but I do bring evidence to the table whilst you don't actually bring anything other than rude insults.

Thanks for the kind words everyone else and I thought I did quite well seeing as I had to squeeze all this testing in between doing 12 hour shifts and keeping the wife happy 😀

My bad, I did glance at your videos but didn't see that comparison. You're getting worse framerates contrary to the two reviews that I've seen so there is something different. And judging by this whole fiasco, I am more willing to accept their results if you don't mind. :hmm:

But of course it isn't as bad as the 5fps video you posted.

I didn't want you to test it at 200% at 1080p but just repeating that that's what was triggering that 5fps behavior according to you while the reviews I posted were at 1080p or even lower where Fury was getting a boost.

And I don't have a personal agenda against you, just the reality. Rude insults? hahaha, amusing.

No, respectful to everyone in this forum you interact with, including members you don't share an opinion with.

I asked you, for the purpose of clarification, to cite the video you are referring to separately, so that individuals reading this thread don't have to dig through fifteen pages to try and figure out what you are talking about. You haven't done that.



Have fun amusing yourself somewhere else.

-- stahlhart

Of course, you guys aren't the only game in town. But this is just too amusing to pass up.




If you have an issue with a mods direction, edit, infraction, warning, etc., make an MD thread and complain about it there.

You do not get to quote the mods edit and them publicly call him out in the thread. Ever.


esquared
Anandtech Forum Director

 

[railven]

So we've all already forgotten that this was replicated on 290's?

Or are we all agenda pushing here?

Because it's getting pretty old.

The folks driving an agenda are clear as day lately. Say something remotely negative about a brand and they jump on you.

Say something positive about a brand and you're their best friend. Now we're back to the 10% cheating theory. Woof.

 

[wand3r3r]

The folks driving an agenda are clear as day lately. Say something remotely negative about a brand and they jump on you.

Say something positive about a brand and you're their best friend. Now we're back to the 10% cheating theory. Woof.

If you are referring to my post, I am looking for investigation into the issue, not claiming there is a 10% performance boost. :hmm:

 

[Headfoot]

An ignorant is just a person that doesnt know any better.

If there is some special way of using it in english in terms of slang that makes it rude. Then I am sorry about that.

The danish wordbook doesnt have any offensive meaning to the word.

just to FYI, I am a native english speaker. spanish is my second language so I completely understand how many different nuances there can be. Here's how the context works in american english at least:

"Person X is ignorant" usually means a judgment of the person's character and can be taken as an insult. In formal contexts (legal/judicial settings for example) it has no insulting character.

"Person X is ignorant of topic Y" has no insulting connotations in almost all contexts. Sometimes this particular form doesn't make grammatical sense though.

Gotta hand it to people who learn english as a second language. It's ugly and full of warts. Really not an easy language to learn. Id hate to have to learn it

 

[godihatework]

The point would be to investigate why/how/when some people are getting reduced quality.

If benchmarks are done with a 10% performance boost they are invalidated.

It's just PC components which none of us should care about other than the integrity of the data, without ulterior motives.

great. where's the investigation into the 290's then?

no posts on that in your history.

ton's of stuff about the 980 tho, how we need to "trust the source" and how we need to "investigate the issue".

Maybe your OP (EDIT: and thread title) should be edited as well, as it seems pretty clear that the initial problem was PEBKAC.

 

[wand3r3r]

great. where's the investigation into the 290's then?

no posts on that in your history.

ton's of stuff about the 980 tho, how we need to "trust the source" and how we need to "investigate the issue".

Maybe your OP (EDIT: and thread title) should be edited as well, as it seems pretty clear that the initial problem was PEBKAC.

What are you referring to with 290's? I'm not opposed to looking into image quality for any card. Actually to the contrary I'm very interested if there is something wrong with the image quality.

As for the title it falls into the overall topic. What's your problem with it or suggestion? I think it's quite neutral like the issue should be.

 

[godihatework]

What are you referring to with 290's? I'm not opposed to looking into image quality for any card. Actually to the contrary I'm very interested if there is something wrong with the image quality.

As for the title it falls into the overall topic. What's your problem with it or suggestion? I think it's quite neutral like the issue should be.

read your own thread.

 

[DEW73]

...By the way, you don't even have to put the NCP in highest quality to fix gta5, you only have to select 16x AF in the control panel instead of "let the application decide". This may be all gregster needs to do to fix his Bf4 issue, i checked bf4 myself and didn't have his issue so I can't try it myself.

Isn't the issue at hand that AMD's "Use application settings" actually let the game in-game settings take precedence over the vendor controlled settings?

From reading the thread it appears Nvidia's default is "Let application decide", which I read as should do the same, i.e. let the in-game settings decide the IQ.

Apparently that's not the case and it needs to manually be set. How does AF affect performance of games nowadays?

 

[godihatework]

Isn't the issue at hand that AMD's "Use application settings" actually let the game in-game settings take precedence over the vendor controlled settings?

From reading the thread it appears Nvidia's default is "Let application decide", which I read as should do the same, i.e. let the in-game settings decide the IQ.

Apparently that's not the case and it needs to manually be set. How does AF affect performance of games nowadays?

No, it's not the case.

Seriously. This is the only forum still spouting this nonsense.

 

[DEW73]

No, it's not the case.

Seriously. This is the only forum still spouting this nonsense.

Isn't that what Gregster was finding though that he had to set it manually to 16xAF in Nvidia's tool? I may have misread the thread, but that was my impression...

 

[godihatework]

:wub:

Isn't that what Gregster was finding though that he had to set it manually to 16xAF in Nvidia's tool? I may have misread the thread, but that was my impression...

because of user error.

this was, and always has been, an example of PEBKAC.

 

[Keysplayr]

No, it's not the case.

Seriously. This is the only forum still spouting this nonsense.

Well, the thread needs to be kept near the top for the most exposure. True or not. Anandtech is one of the most popular forums on the web, so it makes sense for all the focus to be here.

 

[desprado]

What are you referring to with 290's? I'm not opposed to looking into image quality for any card. Actually to the contrary I'm very interested if there is something wrong with the image quality.

As for the title it falls into the overall topic. What's your problem with it or suggestion? I think it's quite neutral like the issue should be.

there is no issue hence some member are trying create an false excuse of Fury X fail launch.

 

[Keysplayr]

there is no issue hence some member are trying create an false excuse of Fury X fail launch.

This can't happen. What's done is done and could have been avoided. That is the sad part. But this is Off Topic. Sorry.

 

[DaveSimmons]

It might make sense for a mod to edit the first post to note that it has been debunked. Most people won't wade through 16 pages to find that information.

 

[wand3r3r]

We still don't know why it required manually overriding settings to get the same image quality.

The whole point is to figure out what's going on. It seems like you can get similar image quality, but most importantly is what can cause the issue to occur where image quality is lacking.

It seems like a relatively easy issue to track down by switching between AMD & NV cards and comparing images.

 

[myocardia]

Well, the thread needs to be kept near the top for the most exposure. True or not. Anandtech is one of the most popular forums on the web, so it makes sense for all the focus to be here.

Indeed. Case in point:

We still don't know why it required manually overriding settings to get the same image quality.

The whole point is to figure out what's going on. It seems like you can get similar image quality, but most importantly is what can cause the issue to occur where image quality is lacking.

It seems like a relatively easy issue to track down by switching between AMD & NV cards and comparing images.

 

[godihatework]

It might make sense for a mod to edit the first post to note that it has been debunked. Most people won't wade through 16 pages to find that information.

i think at this point a lock is probably more warranted - as again, the OP is just repeating the same stuff over and over.

EDIT

it's classic sh*tstirring - "just asking questions", "we need to investigate", "is there a problem"

all of which creates the impression that it is a problem.

It's like anti-vaxxers just "asking questions"

 

[railven]

We still don't know why it required manually overriding settings to get the same image quality.

The whole point is to figure out what's going on. It seems like you can get similar image quality, but most importantly is what can cause the issue to occur where image quality is lacking.

It seems like a relatively easy issue to track down by switching between AMD & NV cards and comparing images.

I think OP is right, we need to find out why AMD cards are suffering from low IQ when set to Default. Clearly someone has to know why THIS is happening, and most importantly if AMD is cheating in review benchmarks!

EDIT: This is with default settings in AMD CCC.

Spoiler

Just having a laugh at OP, here is the post for that above:
http://forums.overclockers.co.uk/showpost.php?p=28276323&postcount=102
http://forums.overclockers.co.uk/showpost.php?p=28271368&postcount=366
If you didn't already know, its a bug found in BF4, neither AMD nor NV are cheating. False alarm, close this thread already.

 

[ShintaiDK]

Seems BF4 is in highlight both ways. Anyone tried to delete the graphics settings file when changing cards in that game?

 

[Sabrewings]

Can this thread die now? Or does it need to be renamed "Why does BF4 have crappy graphics at times?" since we have it confirmed to be a problem on AMD setups as well as no issues have been demonstrated outside BF4?