Asking Ryan Smith of AT if a special examination could be done?

[NIGELG]

Congratulations. This should have been your one and only post in this thread. Welcome!

You are not a mod so don't tell me what to post.

I still maintain that your motives are suspect in this whole issue.Mountains out of mole hills.

 

[Keysplayr]

Not according to TechReport's findings -- one can dismiss it -- attack the web-site -- attack the messengers -- but one can ask for more investigations based on the surprising results from the specific site and other web-sites.

Agreed. No harm in asking for more investigations.

 

[BrightCandle]

Not according to TechReport's findings -- one can dismiss it -- attack the web-site -- attack the messengers -- but one can ask for more investigations based on the surprising results from the specific site and other web-sites.

TechReport is pretty clear on this, NVidia's cards produce a better feeling of motion than AMDs in single cards as well as dual cards. The problem is that some people are happy when micro stuttering is occurring and some are not. But its also true that some people are happy at lower frame rates than others. We can still objectively measure the impact and its still a worthwhile number to get in every review. Further more we can even estimate what FPS a stutter effectively gives you so its comparable.

TechReport is not some fly by bunch of morons, they are currently on the forefront of graphics card performance testing, clearly way ahead of the old guard like Anandtech. Anandtech needs to update and start measuring what actually matters, the perception of motion.

 

[Keysplayr]

You are not a mod so don't tell me what to post.

I still maintain that your motives are suspect in this whole issue.Mountains out of mole hills.

Doesn't matter what you think my motives are. Even if my motive WAS to explicitly show Nvidia in a better light than AMD, it wouldn't matter. The potential issue is STILL there and needs further investigation. My motive doesn't matter one iota, one way or the other. Sorry.

 

[krumme]

Well Keys,
There is a "good relationship" between Anandtech and NV, so if this was a big issue i am sure it would have been investigated, as NV technical marketing would be flooding Anandtech with proofs easy to communicate.

You just shows people where their money goes when they buy NV cards - as not a single card is competitive today - ; into maintaining a lot of bs overloading the forums. Instead of paying for the focus groups they should rather lower the price imho.

 

[Keysplayr]

Well Keys,
There is a "good relationship" between Anandtech and NV, so if this was a big issue i am sure it would have been investigated, as NV technical marketing would be flooding Anandtech with proofs easy to communicate.

You just shows people where their money goes when they buy NV cards - as not a single card is competitive today - ; into maintaining a lot of bs overloading the forums. Instead of paying for the focus groups they should rather lower the price imho.

Whatever you say.

 

[ICDP]

Not according to TechReport's findings -- one can dismiss it -- attack the web-site -- attack the messengers -- but one can ask for more investigations based on the surprising results from the specific site and other web-sites.

I haven't attacked TRs site or methods, nice attempt at a strawman though. Every other TR review since January with AMD 79x0 cards has not shown this stutter problem. In fact they show that frame latency on AMD cards in single GPU config have not been an issue. Or should we throw out the dozenss of other review because this single one paints Nvidia in a better light than most recent reviews?

What "other" web sites are showing this problem in single GPU configurations? Again if you go to other web sites they point to multi GPU configurations and the well known fact that Nvidia is better out of the box. Once again we have people taking one single review that showed peculiar results in Windows 8 and implying that "other web-sites" are getting the same results.

 

[Silverforce11]

What if CF with radeon pro turns out to be faster AND smoother than SLI, while being significantly cheaper?

Btw, its not really a hypothetical question.. because its actually true, dynamic frame rate control in radeon pro is extremely precise at churning out constant frame times. Been testing it in all my games with fraps monitoring, its spectacular how consistent the frames are, even for a single gpu the improvement is there.

IMO, AMD needs this feature in CCC and not just in a 3rd party tool, even if they support its development, something this important should be official, period.

 

[chimaxi83]

Why are you Nvidia people so angry that the claims of two websites can be easily and repeatedly dismissed by using a third party tool? Just because you shouldn't have to use a third party tool? Well guess what, you all use Afterburner, or Precision, or GPU Tweak, so where's all the crying and moaning over having to use those to overclock and overvolt your beloved Nvidia cards? Hypocrites.

 

[SirPauly]

I haven't attacked TRs site or methods, nice attempt at a strawman though. Every other TR review since January with AMD 79x0 cards has not shown this stutter problem. In fact they show that frame latency on AMD cards in single GPU config have not been an issue. Or should we throw out the dozenss of other review because this single one paints Nvidia in a better light than most recent reviews?

What "other" web sites are showing this problem in single GPU configurations? Again if you go to other web sites they point to multi GPU configurations and the well known fact that Nvidia is better out of the box. Once again we have people taking one single review that showed peculiar results in Windows 8 and implying that "other web-sites" are getting the same results.

One may ask for for more investigations from other web-sites and from the site that offered the surprising results.

One may dismiss the results; one may attack Tech Report; one may attack messengers; one may ask for investigations -- my points wasn't about you! TechReport offered their findings -- they're surprising -- worthy of investigation for some -- others it may not be.

 

[SirPauly]

What if CF with radeon pro turns out to be faster AND smoother than SLI, while being significantly cheaper?

Wonderful! The awareness may help improve nVidia products.

 

[omeds]

What if CF with radeon pro turns out to be faster AND smoother than SLI, while being significantly cheaper?

Btw, its not really a hypothetical question.. because its actually true, dynamic frame rate control in radeon pro is extremely precise at churning out constant frame times. Been testing it in all my games with fraps monitoring, its spectacular how consistent the frames are, even for a single gpu the improvement is there.

IMO, AMD needs this feature in CCC and not just in a 3rd party tool, even if they support its development, something this important should be official, period.

The same results can be achieved on each technology using the same methods via 3rd party tools. Great news for us enthusiasts.
But, out of the box experience is important for the majority who just want plug and play. This is the main target audience for both Nv and AMD, and why this issue should be addressed from both camps.

 

[Jaydip]

The same results can be achieved on each technology using the same methods via 3rd party tools. Great news for us enthusiasts.
But, out of the box experience is important for the majority who just want plug and play. This is the main target audience for both Nv and AMD, and why this issue should be addressed from both camps.

Exactly :thumbsup:

 

[VulgarDisplay]

The same results can be achieved on each technology using the same methods via 3rd party tools. Great news for us enthusiasts.
But, out of the box experience is important for the majority who just want plug and play. This is the main target audience for both Nv and AMD, and why this issue should be addressed from both camps.

You do realize that the number of people who are using SLi/Crossfire that just plug and play is probably zero right?

If someone is just going to plug and play they bought a dell or some other mass produced garbage. They also have no idea what microstutter is, and probably don't even know what a framerate is.

 

[AnandThenMan]

One may

One may

one may

one may

one may

LOL.

Anyway, I agree with some others here, this thread was created with the sole intention of spreading FUD, nothing more.

 

[bryanW1995]

Well Keys,
There is a "good relationship" between Anandtech and NV, so if this was a big issue i am sure it would have been investigated, as NV technical marketing would be flooding Anandtech with proofs easy to communicate.

You just shows people where their money goes when they buy NV cards - as not a single card is competitive today - ; into maintaining a lot of bs overloading the forums. Instead of paying for the focus groups they should rather lower the price imho.

I disagree. Ryan Smith made a dumb decision one time to let an overclocked gtx 460 run in a comparison with the 6870 launch, but since then I've not seen a single thing that gave me the impression that he or the site unfairly favor NV over AMD. And let's face it, AT doesn't need to kiss jhh's butt to get cards for evaluation, anyway. I'd be much more worried about smaller sites sucking up to major corporations than big ones like AT.

 

[wand3r3r]

This thread is somehow like putting this on repeat.
http://www.youtube.com/watch?feature=player_embedded&v=EhgYrY8fN-0

 

[blackened23]

TechReport is pretty clear on this, NVidia's cards produce a better feeling of motion than AMDs in single cards as well as dual cards. The problem is that some people are happy when micro stuttering is occurring and some are not. But its also true that some people are happy at lower frame rates than others. We can still objectively measure the impact and its still a worthwhile number to get in every review. Further more we can even estimate what FPS a stutter effectively gives you so its comparable.

TechReport is not some fly by bunch of morons, they are currently on the forefront of graphics card performance testing, clearly way ahead of the old guard like Anandtech. Anandtech needs to update and start measuring what actually matters, the perception of motion.

Like you, have used both setups in multi GPU - and while I have seen MS before in prior generations of GPUs, I have not seen any type of MS with vsync enabled on either setup. I used vsync, which I think is representative for how most people play games; both setups can be choppy with vsync turned off, perhaps nvidia less so with adaptive vsync? I don't know, but with vsync on they are both smooth. I don't witness MS with 680 sli with avsync on, and I never witnessed MS with 7970s with vsync either.

If there is an objective measure that proves beyond a shadow of a doubt that there is a substantiation of this, I would by all means welcome it. When someone says "I feel like such and such is smoother" without any real objective data, and while playing with vsync off -- it just raises red flags. I want data. Not subjectivity. Also, I understand that vsync off is desirable for benchmarking purposes but that likely is not how most people play games.

I really hope Ryan can get an objective measure for this and post it for all to see.

 

[lavaheadache]

I have 7970's in crossfire and I do see micro stutter. I can make it go away, however.

 

[wand3r3r]

nm.

 

[omeds]

You do realize that the number of people who are using SLi/Crossfire that just plug and play is probably zero right?

If someone is just going to plug and play they bought a dell or some other mass produced garbage. They also have no idea what microstutter is, and probably don't even know what a framerate is.

Not true, theres plenty of "average" gamers that want more performance which can only be achieved via MGPU, and can notice issues when they play. Most people dont want to have to mess around with 3rd party tools, finding the best frame caps and setting up profiles for each game. Many just want fast, smooth gameplay with minimal hassles.
I know plenty of pro and semi pro players that couldnt give 2 shiits about their hardware either, but will scream when its not smooth. They dont care about PC's, they care about gaming and want a trouble free experience. You see, most PC gaming enthusiasts are not computer hardware enthusiasts. Those are two different things.

It's also beneficial for both vendors to improve the MGPU plug and play expereience, meaning they can sell more hardware, instead of average consumers being put off by the issues and messing around required.

I dont see the problem anyway, whats it to you if both vendors improve in this area?

 

[SirPauly]

I disagree. Ryan Smith made a dumb decision one time to let an overclocked gtx 460 run in a comparison with the 6870 launch, but since then I've not seen a single thing that gave me the impression that he or the site unfairly favor NV over AMD. And let's face it, AT doesn't need to kiss jhh's butt to get cards for evaluation, anyway. I'd be much more worried about smaller sites sucking up to major corporations than big ones like AT.

Ryan also missed the problem with blending mip-maps with the 58XX series.

 

[Keysplayr]

Ryan also missed the problem with blending mip-maps with the 58XX series.

Hey, nobody is perfect and someone can miss something once in a while.

But I know Anandtech's members would love to see Ryan's take after any investigation. Who wouldn't?

 

[Silverforce11]

Seeing what Radeon Pro can do, I want AMD to make these features official, there's no point in arguing against it. There's simply NO good reason not to do it. Pay the author a small fee and get it done.

 

[blastingcap]

And let's face it, AT doesn't need to kiss jhh's butt to get cards for evaluation, anyway. I'd be much more worried about smaller sites sucking up to major corporations than big ones like AT.

http://www.anandtech.com/print/2731

"Early last week Charlie over at The Inquirer posted a story saying that a number of reviewers were cut out of the GeForce GTS 250 launch. We felt a bit hurt, by the time the story launched we weren't even asked to be briefed about the GTS 250. Cards had already gone out to other reviewers but we weren't on any lists. Oh, pout.

Magically, a couple of days after Charlie's article we got invited to a NVIDIA briefing and we had a GTS 250 to test. Perhaps NVIDIA was simply uncharacteristically late in briefing us about a new GPU launch. Perhaps NVIDIA was afraid we'd point out that it was nothing more than a 9800 GTX+ that ran a little cooler. Or perhaps we haven't been positive enough about CUDA and PhysX and NVIDIA was trying to punish us.

Who knows what went on at NVIDIA prior to the launch, we're here to review the card, but for what it's worth - thank you Charlie "