[TECH Report] As the second turns: the web digests our game testing methods

[chimaxi83]

What's the reason for immediate flame bait and thread crap starting a mere 6 posts into this thread?

Anyway,

Right. Now all we need to see is if they do what they say and if the fix has any effect on the performance or the fps reported. It is very good that AMD acknowledged this problem.

Agreed :thumbsup:

 

[HurleyBird]

AMD Fanboys were pretty bad in the last thread. The entire "all anyone said was more testing is needed" excuse rings hollow against all but the most liberal interpretation of the phrase, and that's obvious for anyone without a strong AMD bias.

That being said, the behavior of AMD fanboys in the "smoothness" thread doesn't even scratch the surface compared to antics of the half dozen or so Nvidia zealots (we all know who they are) we've accumulated on this forum. Compared to them, the AMD fanboys are pretty mild.

Why everyone needs to make such a big deal and wear these companies on their sleeves, I have no idea. Being in love with tech in general is cool, but being in love with one specific company is frankly pathetic.

 

[3DVagabond]

Flat out "call Shens"

So, are we ready to call Shens. on this, yet? There's just too much evidence, even from their own reviews, that contradicts their findings. Even looking at last gens results, where latencies were well into the range we're being lead to believe is noticeable, nobody's cards were stuttering then. And as you point out, nobody was saying AMD was "slower but smoother".

So do you have any response to the points raised?

Why are the results from the 570 far worse than the 6970 in the same comparison, but it wasn't less smooth than the 6970? Nobody ever claimed the 6970 was slower but smoother. Especially not the people who are so excited about the current results.

Why was the performance so much worse in later reviews than prior reviews for the 7950? And, no it wasn't the current drivers, because they showed it wasn't the 12.11B causing the issue. They claimed it was present in older (slower) drivers.

Again, it's a case of attacking the poster not discussing the content of the post. Try and make a post that adds to the discussion, and addresses the article(s) in question not just troll individuals.

 

[Ben90]

At the risk of incurring the wrath of the intrawebs, haven't we known this since before unified shaders? I've been wary of ATI since I became an enthusiast. I thought they would have fixed things by now.

 

[Imouto]

I loved when this issue was irrelevant with GTX 570.

Seriously guys, WTF.

 

[Keysplayr]

I loved when this issue was irrelevant with GTX 570.

Seriously guys, WTF.

Its irrelevant with the 7 series. So you have your
Homeostasis.

 

[lavaheadache]

I love how people justify things like this by pointing out the fault of another. It just makes no sense

 

[Keysplayr]

So much love.

 

[KingFatty]

So if I recall correctly, the issue that AMD acknowledged for the 7950 GPUs is also an issue in the laptop 8XXXM GPUs - or am I remembering that wrong?

 

[Lepton87]

I don't see how my response was taken as "nah, the problem doesn't exist"
It's confined to GCN architecture and all I did was saying that I'm not seeing any of this on my QUAD VLIW-4 rig. Good that I didn't jump to "upgrade". So 7850 might actually be less fluid than 6950 in games in spite of delivering more fps. Interesting. Glad to see that older VLIW cards are unaffected.

 

[SirPauly]

So do you have any response to the points raised?

Why are the results from the 570 far worse than the 6970 in the same comparison, but it wasn't less smooth than the 6970?

Possibly the memory differences between Sku's?


OT:

IF members can go beyond attacking and pointing fingers would be appreciated.

 

[f1sherman]

[H]onorable Kyle's take on the issue

http://hardforum.com/showpost.php?p=1039478800&postcount=10

 

[ArchAngel777]

The point is, there were people posting who wouldn't admit to something being off or wrong if AMD themselves had come forward right away and said it.

Are you daft? Care to show me the people who denied it after AMD confirmed it?

 

[Ferzerp]

[H]onorable Kyle's take on the issue

http://hardforum.com/showpost.php?p=1039478800&postcount=10

That reply is pure gold.

 

[f1sherman]

@ArchAngel777 1 post above you

And I disagree with Tviceman:
As much as I am convinced that NVIDIA would hardly be this open, am just as much sure majority of NV posters would be all over NVIDIA about the issue.

Yes there would be couple of "everything working fine here" and "nothing to report" posts,
but hardly mass denial, relativizing and questioning of TR honesty/accuracy.

That's why official NV forum is a useful place to report bugs, and AMD's official forum is a useless internet dump.

 

[SirPauly]

I missed this post and quite the read:

http://forum.beyond3d.com/showpost.php?p=1689708&postcount=1

7) This is precisely why it is important to measure frame latencies instead of frame throughput. We need to keep GPU vendors, driver writers and game developers honest and focused on the task of delivering smooth, consistent frames. For example (unrelated to the above case), allowing things like a driver to cause a spike one frame so that it can optimize the shaders and get higher FPS number for the benchmarks is not okay!

This caught my eye!

 

[ArchAngel777]

@ArchAngel777 1 post above you

Fair enough, there is at least one idiot. But, quite honestly, what I react to is the hyperbole being used around here. Most people are level headed around here, with a handful on each side acting like idiots. That is hardly a reason to state one side or the other is full of deniers.

 

[Hitman928]

Same guy after people questioned those comments and explaining why they don't thing that is what's happening :

Completely agreed, see my clarification to caveman-jim above. I didn't mean to imply that or attempt to explain what's going on in this specific case... it was just another example of why measuring and optimizing for frame rates is a bad plan.

 

[f1sherman]

Fair enough, there is at least one idiot. But, quite honestly, what I react to is the hyperbole being used around here. Most people are level headed around here, with a handful on each side acting like idiots. That is hardly a reason to state one side or the other is full of deniers.

Notice the trend:

• AF shimmering with HD6000,
• automatic FP16 demotion
• HD7970 release driver AF blurring
• TR findings on frame (in)consistency

In all above cases it goes like this:

Phase 1: 3rd party website/reviewer notices an issue.
Phase 2: AMD posters go into denial mode, they blame NVIDIA, biased reviewer and whatnot
Phase 3: AMD releases a fix.

 

[ArchAngel777]

Notice the trend:

• AF shimmering with HD6000,
• automatic FP16 demotion
• HD7970 release driver AF blurring
• TR findings on frame (in)consistency

In all above cases it goes like this:

Phase 1: 3rd party website/reviewer notices an issue.
Phase 2: AMD posters go into denial mode, they blame NVIDIA, biased reviewer and whatnot
Phase 3: AMD releases a fix.

You must be new around here, because these tactics were used by nVidia trolls first. Sure, both sides use them now, but not always. I guess you forgot about the nVida 7800 GTX shimmering issue? Trollo making light of it, etc... Like I said, you must be new around here, or perhaps you are just trolling yourself.

 

[bunnyfubbles]

For those hoping there won't be a performance hit, the way I see is is if there is a minor performance hit, it will be worth it as long as the microstutter is gone.

Reported framerates mean little if they're so inconsistent that it feels like a lot less.

 

[Final8ty]

I love how people justify things like this by pointing out the fault of another. It just makes no sense

While that can be the case it is also used to show double standards and hypocrisy.

 

[SirPauly]

Same guy after people questioned those comments and explaining why they don't thing that is what's happening :

He never made any conclusions -- raised points on the importance of latency. Personally glad to see discussions and investigations that go beyond frame-rate.

 

[ArchAngel777]

http://forums.anandtech.com/showthread.php?t=1822635&highlight=7800+gtx+shimmering

Yep, no double standards here. nVidia never takes shortcuts... Wait, delete that link!

 

[96Firebird]

While that can be the case it is also used to show double standards and hypocrisy.

Which lends a helping hand to attacking posters and hardly the issue at hand.