HD 7950 May Give Higher Framerates, but GTX 660 Ti Still Smoother: Report

[ThunderStorm]

I just hope AMD optimizes its drivers nicely, especially in CFX configuration. Micro-shuttering for me is just meh, not a big deal at all since I have experienced this phenomenon on both AMD and nVIDIA cards.

 

[NdMk2o1o]

Wow, just WOW!!!!!!!!!!!!!!!!!

I mean the end of the world is upon us and this is all you can think to argue about, priorities people!!!

OT

I couldn't be any happier with the performance of my 7950 coming from a 570 and don't notice any microstuttering in any of my games, not saying it's not there I just think it's probably a moot point to 90% of people, the other 10% of people are probably those who claim they can see and perform better in games running at over 120FPS

 

[esrever]

I would really wish some other site were able to confirm or deny these findings. I stopped reading TTR after reading their stupid article about AMD censoring material and the original 660 ti review due to their blatant bias. They use a different method to test cards/cpus which is good but their numbers are often times inconsistent with everyone else. The review themselves are just written as best of they can to down play any AMD cards and even to exclude any test that would showed AMD winning.

I wouldn't actually be surprised if TTR just randomly made up the graphs at this point because of how they insist on doing the same articles to make nvidia look better at this point and using data that not only is selected based on favoritism but also inconsistent with other sites, they only hide behind their testing method at this point.

First it was the original 660ti review where they included TWIMTBP games where AMD suffered in the final conclusion but removed gaming evolved titles where nvidia suffered. Then the update which they hid behind windows 8 drivers and a selection of games that were different, Removing BF3 and other AMD favored titles and adding a few of TWIMTBP titles. The conclusion was stupid and the testing was stupid so when they were called out, The made the follow up review which at this point I am not even inclined to believe the numbers.

The whole year of review from the tech report has been praising nvidia. Thats not so bad but when they often skew the price of AMD and nvidia cards in the conclusion and not including AMD favored results. It starts to be something.

 

[Hilux SSRG]

well i returned by gtx670 after stutter after stutter in games. This smoothness thing that people keep mentioning is all in the head bcoz i couldn't find it I will try the amd 7 series and see if that helps with higher frame rates, if not i will return that also and wait for next gen

Did you happen to have a launch gtx 670 ASUS or EVGA? Many people did RMA and it fixed the problem quick.

 

[the54thvoid]

I know this is about Skyrim stutter but i previously mentioned graphical stutter in FC3. Just played the co-op version and it was silky smooth. Even the SP is now smoother after a sizable download for the game (mysterious 350mb file).

I'd put it down to drivers. Fixable. And it's not an issue to me so I'm happy.

But it is a bit shameful the ferocity of the arguments in this thread.

 

[LAN_deRf_HA]

I just checked, and those arguing for a particular side matches what is in their 'system specs'.

A few of you have also spoken against tpu for posting something like this. This is a great topic to make sure that you aren't speaking in a biased tongue before clicking 'post reply'.

That being said, this review should have taken into account more than one game if they are going to go public, as it is unintentionally slandering a company, and is pretty serious in a business sense. While I haven't noticed it on CF 5750s or my 7950, the videos do show that there is a real problem in Skyrim, nothing more, so take it as that. I trust that AMD is looking into this, and if they see it as a serious issue, they will take measures to fix this.

I guess you missed the 2 articles before this. It's not just Skyrim. And you're damn right there's a system spec correlation. I've dealt with enough AMD bugs to know I should steer clear. Even if this microstutter is a newer driver problem, which I don't see how it could be as I've seen this issue since the 7950s launch, it would still just be the most recent in a long line of critical bugs. People can cry to their hearts content that anyone so harsh on the brand they blew money on must be a fanboy, but it's not really about that. It's about experience.

 

[crazyeyesreaper]

I could care less eitherway i just want more testing done

7950 vs 7950 boost does the clock speed fluctuations cause the problem to be more pronounced

how bad is the issue on say the 7970 and 7970 GHz edition does clock speed on the GPU lower overall latency giving higher clocked cards any kind of benefit or does it further cause problems

does the 7870 experience the same issues

Its an interesting glimpse but 1 GPU 1 Driver revision is not enought to say theres a huge problem here. more data would be appreciated in this case to get a clearer view.

Doesn't make a helluva lot of difference imo. TR are only claiming the discrepancy regarding the 7950 Boost, they aren't extrapolating the results and saying it affects all models.
Having said that, given the standing that Scott has in the community, I'd say that someone will test the 7970, 7970GE etc. It wouldn't surprise me if AMD were optimizing for outright speed at all- that is the primary metric used in the majority of reviews, and a major bullet point in PR slides.

AMD had no problem reaping the rewards of reviews using the all new 7950 Boost...are you saying that TR should gloss over any potential problem because the bulk of the other models are possibly OK ? HardOCP and Tom's Hardware amongst others are also reporting various issues in single/multi card configs- at what point do you say its worth reporting?

first i didnt say TR did im saying essentially i want more testing done before saying somethings wrong drivers / auto boost clocks going up and down etc can all have an impact on stuttering 7970 stock vs 7970 ghz edition and 7950 stock would be welcome tests on the same driver to see if boost clocks are causing some issues. Seems your taking my post and trying to paint it in a light that makes your post more meaningful, i simply want more testing done nothing more, Altho why people are up in arms over skyrim surprises me AMD has had issues with the Gamebryo engine for hell 9-10 years now Morrowind / Oblivion / Fallout 3 / Fallout New Vegas / Skyrim all exhibit this issue. an easy way to fix it is 1 mod the game a great deal and 2 lock it to 59Hz. That said stuttering can be issue ive seen it many times. not saying TR is wrong but what im saying is their can be easy solutions that fix the problem.

I do have to say this brings back memories of my 4870x2 and the stuttering in Oblivion due to crossfire.

 

[slatanek]

How about shadow flicker?

Micro-stutter being one, but what are your guys experiences with shadow flickering? I have a HD7850 and as there is movement in a scene (i.e. camera panning) the shadows are a mess! I get this in every game I play. Another issue - corrupted frames. Every now and then a frame is fucked up, like there are solid, black triangles, untextured objects etc. I mean it happens from time to time and at first I thought "oooh, its just nothing". Now the longer I'm being exposed to it the more irritating it is.
The problem is these issues are small at first, but after you've seen something once it begins to bother you more and more with time. And I really don't care if its software or hardware - what matters is the user experience. Next time I'll give nVidia a try.

 

[SIGSEGV]

Micro-stutter being one, but what are your guys experiences with shadow flickering? I have a HD7850 and as there is movement in a scene (i.e. camera panning) the shadows are a mess! I get this in every game I play. Another issue - corrupted frames. Every now and then a frame is fucked up, like there are solid, black triangles, untextured objects etc. I mean it happens from time to time and at first I thought "oooh, its just nothing". Now the longer I'm being exposed to it the more irritating it is.

what kind of driver version have you used?
in case you have already used 12.11 beta driver then you should submit your problem here instead posting off topic here

Next time I'll give nVidia a try.

that's your own choice whether you'll gear up your rig with radeon or nvidia, end of story

 

[slatanek]

WHQL driver 12.10

I'm using the latest WHQL driver, which is 12.10, so I won't post here :shadedshu

 

[ensabrenoir]

....i went from single ati to xfired 6870's to single gtx 670..., nvidia is a different beast all together. As much as i love, recommend and buy amd gpus in my experience /opinion. Green teams got em beat. Amd is purpose designed and built as the value alternative....

 

[sergionography]

i wonder if the screen being 60hz has to do with anything, because if anything i dont see how higher fps can be constantly less smooth, in this case its either the graphic card is rendering multiples of the same frame, or due to the 60hz which can only handle 60fps? is skipping anything over 60
or idk i guess i just dont understand this whole latency thing, it just sounds fishy to me that more frames are taking longer times in between to display
another thing would be the mechanism in which the frames are transmitted, its clear on the amd the frame overlaps the other, where for nvidia maybe each frame fades while the other displays?

im very ignorant on this matter. but can anyone clarify?

 

[jihadjoe]

i wonder if the screen being 60hz has to do with anything, because if anything i dont see how higher fps can be constantly less smooth, in this case its either the graphic card is rendering multiples of the same frame, or due to the 60hz which can only handle 60fps? is skipping anything over 60
or idk i guess i just dont understand this whole latency thing, it just sounds fishy to me that more frames are taking longer times in between to display
another thing would be the mechanism in which the frames are transmitted, its clear on the amd the frame overlaps the other, where for nvidia maybe each frame fades while the other displays?

im very ignorant on this matter. but can anyone clarify?

The average frame rate is higher, but when the Radeon encounters a specifically difficult frame it tends to lag, hence the stutters.

 

[Melvis]

Some one should start a Poll? see what us the people that use these cards both nvidia and AMD to see if yes i can see this problem or no i dont see it at all on either cards?

 

[DOM]

Imo is just BS since amd has a faster card

If someone wants to give me a 680 I can do a comparison

 

[sergionography]

The average frame rate is higher, but when the Radeon encounters a specifically difficult frame it tends to lag, hence the stutters.

but in the video at 240hz camera shot the amd consistently appears to have less frames as u can almost tell every frame thats changing, so what i was saying how could more fps be more apperant
yes i get your point but that is when for example an intensive scene comes up and it stutters for a bit or in some cases fps takes a dip thats why some reviews have min fps in their graphs, but i didnt notice that in the videos here, its just consistent frame by frame change is what i notice

 

[Rahmat Sofyan]

Lets hope, The Smoothest FPS appear on next catagories on HWBot, and The Fastest Card on 3D Mark dissappear...LoL

Since Radeon 7000, and now HD7000... I didn't find that trouble, in my opinion we are just too lazy to do some tweak and wait the new fix, and I believe everybody on this forum can do it.

 

[Frick]

It's about experience.

Which is what makes it so weird to me. TR has hard data, you can't really argue with that, but why are there people with experiance and know-how saying they don't have it at all? Is there other things at play here, or is it just a case of people simply not seeing it?

i just want more testing done

Indeed so.

Some one should start a Poll? see what us the people that use these cards both nvidia and AMD to see if yes i can see this problem or no i dont see it at all on either cards?

I was thinking about that as well, but I don't think a poll would cover it.

 

[m1dg3t]

I was thinking about that as well, but I don't think a poll would cover it.

Don't worry, my BS will cover it

 

[Ferrum Master]

Huang paying double salary to Nvidia's Trolling Division

 

[3870x2]

I have had stuttering on both AMD and NVidia cards now that I remember (they are talking about the long pauses you see every once in a while, not the ones happening every few frames that are consistent in the videos).

I thought it was because my CPU was doing something, and since I couldn't pinpoint what it was exactly, I would just restart and not see the problem again for a very long time. This has happened maybe a half dozen times in the last five years.

I just now thought of this after looking at the video again. The first time I remember it happening was on the 9800GX2, then on a 4870.

 

[kristimetal]

That stuttering

After reading the review amd all the comments (both on TPU and techreport), I had the preconception that 660ti is stutter-free.
Well, i watched the videos (both 120hz and 240hz videos) and i ended up with the conclusion (as i said, thinking that 660ti has more "magic" to it and no stuttering) that both GPUs stutter like hell, of course, the effect was amplified by recording the game at 240hz, where the actual frame rate was ~80fps, so the choppiness.
In the areas where 7950 stutter, like where you start going up the small hil after you pass that road (i think), but also the 660ti stutter, just as an example.
Unfortunatelly, video cards have stuttering, and Skyrim is a bad example to exhibit, cause Skyrim is poorly optimize, remember SkyBoost.
We should see more games tested, maybe TPU will conduct a test like this.

 

[tacosRcool]

smooth as butter

 

[Melvis]

I was thinking about that as well, but I don't think a poll would cover it.

Yea im not sure either, but you will get BS from both sides so it should even out in the end?

 

[Frick]

Yea im not sure either, but you will get BS from both sides so it should even out in the end?

I meant that you'd have to gather a lot of specs, so a normal poll wouldn't cut it. Driver versions, OS versions, what systems they were using, what games etc.