AMD 7000 Initial IQ Issues *Now fixed* [Tom's Hardware Review]

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
This was the first I had heard of this, but the image quality screenshots do speak for themselves. I don't want to start a flame-war here (I like both AMD and NV offerings) but issues like this should make AMD accountable, if true. If you look at the results, there are some big gains in fps between some the games such as Metro and Skyrim specifically when you look at the performance levels.

I definitely do not like the default details turned down to get a few fps. NV did this in the past, and they paid for similar behavior.

Thoughts?

http://www.tomshardware.com/reviews/image-quality-driver-optimization-graphics,3173.html
 
Last edited by a moderator:

Nintendesert

Diamond Member
Mar 28, 2010
7,761
5
0
Based on our original findings, it wasn’t a stretch to question if AMD had decided to try an old trick with the 7000 series, giving up hard-to-notice texture detail for a couple of frames per second in some games. We’re glad that AMD responded, clarified the situation, and provided proof in a fixed pre-release driver, plus a commitment to include that resolution in the upcoming Catalyst 12.4 WHQL driver. The driver it sent us for testing suggests that the issue can be solved without sacrificing performance.



I think it's a non-issue since the fixed driver doesn't negatively affect the current scores and performance so AMD didn't gain anything by these texture bugs/settings.
 

Ferzerp

Diamond Member
Oct 12, 1999
6,438
107
106
Doesn't appear to be malicious as they were given a patched driver that fixed the problem at barely any performance penalty.
 

djsb

Member
Jun 14, 2011
81
0
61
Once again Toms decides to post image quality shots in animated gif form. Way to go.

Will read the article anyway, but sheesh.
 

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
It is a worry these bugs and others are still getting through the quality process, but it doesn't appear to produce any advantage in performance so it does not look like cheating.
 

Olikan

Platinum Member
Sep 23, 2011
2,023
275
126
it's a drivers bug, very common in new arquitectures

happen the same with 6000 series and nvidia in the past

will be fixed in 12.4 driver without performance lost
 

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
Once again Toms decides to post image quality shots in animated gif form. Way to go.

Will read the article anyway, but sheesh.

Its horrid. It would be slightly less bad if it was just an image of the 7970 verses 6970 etc graphics card to show the difference in one image, but they are cycling through a number of different images.
 

Concillian

Diamond Member
May 26, 2004
3,751
8
81
I really do not like the animated gif way of showing that.

But... this is an instance where you have to give it to Tom's. Good on them that they found this issue, then worked with AMD to get it resolved. Whether you praise or chide AMD for this is up for debate, the journalism, however is unarguably commendable.
 

videogames101

Diamond Member
Aug 24, 2005
6,783
27
91
I really do not like the animated gif way of showing that.

But... this is an instance where you have to give it to Tom's. Good on them that they found this issue, then worked with AMD to get it resolved. Whether you praise or chide AMD for this is up for debate, the journalism, however is unarguably commendable.

Their article title is trolling for site hits =/
 

Ferzerp

Diamond Member
Oct 12, 1999
6,438
107
106
Their article title is trolling for site hits =/


Depends. There is the assertion that any time an article poses a question, the answer is *always* no. In that way, if you follow that mindset, no ;)

Honestly though, they had an AMD response prior to publishing it and have the proof that that isn't the case, so they shouldn't have published with that title really.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
Depends. There is the assertion that any time an article poses a question, the answer is *always* no. In that way, if you follow that mindset, no ;)

Honestly though, they had an AMD response prior to publishing it and have the proof that that isn't the case, so they shouldn't have published with that title really.

That's a great statement. The title was more sensationalist than it should have been, based on the results.

Thankfully performance was mostly untouched.
 

LOL_Wut_Axel

Diamond Member
Mar 26, 2011
4,310
8
81
Nice sensationalist title from Tom's for a problem that's already fixed. They've really lost a lot of credibility in recent months.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Nice sensationalist title from Tom's for a problem that's already fixed. They've really lost a lot of credibility in recent months.

Whats funny is that the author of that article posts on various forums taunting AMD cpu owners and trolling. Not that I disagree about bulldozer (it is garbage), but the author has a pretty extensive history of trolling / being unprofessional towards AMD stuff. At least you can count on some reviewers being, you know, objective regardless of what they think on other websites.
 
Last edited:

LOL_Wut_Axel

Diamond Member
Mar 26, 2011
4,310
8
81
Whats funny is that the author of that article posts on various forums taunting AMD cpu owners and trolling. Not that I disagree about bulldozer (it is garbage), but the author has a pretty extensive history of trolling / being unprofessional towards AMD stuff. At least you can count on some reviewers being, you know, objective regardless of what they think on other websites.

Some people have too many problems overcoming their personal bias. These type of people shouldn't be hired to be editors.

Alas, that seems to be too much to ask for now-a-days.
 

utahraptor

Golden Member
Apr 26, 2004
1,053
199
106
I love the article, hate the GIFs. I want side by side!

Also, another vote for Tom's being equal to Anandtech period :)
 

SolMiester

Diamond Member
Dec 19, 2004
5,330
17
76
This was the first I had heard of this, but the image quality screenshots do speak for themselves. I don't want to start a flame-war here (I like both AMD and NV offerings) but issues like this should make AMD accountable, if true. If you look at the results, there are some big gains in fps between some the games such as Metro and Skyrim specifically when you look at the performance levels.

I definitely do not like the default details turned down to get a few fps. NV did this in the past, and they paid for similar behavior.

Thoughts?

http://www.tomshardware.com/reviews/image-quality-driver-optimization-graphics,3173.html

Eh?...I thought that was AMD in the 5xxx series, that lowered IQ for FPS....
 

notty22

Diamond Member
Jan 1, 2010
3,375
0
0
It was AMD with the 68XX Bart's launch. Where they introduced a 6850/6870 that was slower than the 5850/5870 while using those questionable drivers. Though those releases were priced much more competitively to counter the GTX 460 at the time.

Benchmark Reviews can confirm that issues with filtering still exist, and pointed this out in our Radeon HD 6850 and Radeon HD 6870 launch articles. We also made it public that certain AMD partners were sending 'juiced' video card samples to reviews sites, ours included, with details published in our 1120-Core "Fixed" Radeon HD 6850 Review Samples Shipped to Media article. So could this be AMDs last ditch effort to compete with NVIDIA by manipulating performance?
 
Last edited:

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
You need to go back much further than that to get to the Nvidia time when they did it.

Yeah, the Geforce FX was probably the most notorious (and egregious) example I can think of since the 'modern' GPU was born in ~1999.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
Eh?...I thought that was AMD in the 5xxx series, that lowered IQ for FPS....

Any example of lowering IQ to get extra fps is downright wrong. Unless the company comes right out up-front and explains their actions and provides a tangible reason (can't think of any myself) it's downright sneaky.

This example that THG found is interesting because it is hard to tell if it was truly a QA issue, or something planned. Definitely not at a 'conspiracy' level or anything, but it's difficult to know if it was an honest mistake or not. Either way, AMD should get a little credit for acknowledging the issue and fixing it with the a driver release quickly. I doubt anyone will know the real reason, but its good to see it resolved IMHO.
 

BoFox

Senior member
May 10, 2008
689
0
0
Their article title is trolling for site hits =/

It's just a question - the very question that the article investigates.

Better yet, it's a rather good thing that Toms investigated it, else HD 7xxx owners would probably be suffering these IQ issues for some more time (even indefinitely).