What are Kyle and Brent's opinions on Tech Report's GTX 660 Ti vs HD 7950 findings?

burningrave101

[H]F Junkie
Joined
Sep 9, 2003
Messages
11,825
Maybe I missed it, but I didn't see a discussion here regarding the recent tests and conclusions from The Tech Report's findings for AMD and nVidia's latest drivers relating to smoothness of gameplay and frame latency.

I know HardOCP has a very hands on approach to in-game testing for the GPU reviews which makes them one of the go to sources for accurate real world testing and I wanted to know what Kyle and Brent thought of these results from The Tech Report and whether or not they line up at all with their own real world testing. The results have really shook things up over at places like the OCN forums since so many have been declaring the 7950 and 7970 as the clear unattested choice on both price and top performance since the release of AMD's recent fps boosting drivers.

The Tech Report said:
In the end, we're left to confront the fact that the biggest change from our prior graphics reviews was the influx of new games and new test scenarios that stress the GPUs differently than before. (The transition to Windows 8 could play some role here, but we doubt it.) For whatever reason, AMD's combination of GPU hardware and driver software doesn't perform as well as Nvidia's does in this latest round of games, at least as we tested them. That's particularly true when you focus on gameplay smoothness, as our latency-focused metrics tend to do.

http://techreport.com/review/23981/radeon-hd-7950-vs-geforce-gtx-660-ti-revisited

The Tech Report said:
The question you have to ask is what matters to you. Do you want the graphics card that scores best in the FPS beauty pageant, or do you want that one that gives you the smoothest gaming experience when you fire up a freshly downloaded game this Christmas? If you just want bragging rights, by all means, choose the Radeon HD 7950. If you're looking for the friction-free fluidity that only comes from consistently quick frame delivery, though, our recommendation remains the GeForce GTX 660 Ti.

http://techreport.com/review/24022/does-the-radeon-hd-7950-stumble-in-windows-8

Link to the discussions at OCN:

http://www.overclock.net/t/1334760/the-tech-report-radeon-hd-7950-vs-geforce-gtx-660-ti-revisited
http://www.overclock.net/t/1337206/the-tech-report-does-the-radeon-hd-7950-stumble-in-windows-8

Thoughts?
 
Last edited:
I poo poo'd the "smoothness" argument when Brent made it his last review but once I examined the frame rate graphs I couldn't really deny that the Nvidia graph is considerably more linear than the AMD line. Some of the swings on the AMD chart were huge and would definitely be noticeable.

Guess its one more time HardOCP was ahead of the curve and its something that's going to have to be addressed in GPU reviews from now on.
 
My thoughts are if you want fluidity in gaming, you should aim for an SSD if you are not using one. SSD's enhanced loading times made the biggest difference in smoothness in gameplay I've seen in a long time. There's a great 'Intel-lecture series' targetted towards developments online somewhere that discussions basically the second-derivative of fps and the added smoothness that occurs from the smaller second derivative. (Changes in changes between frame rates being smaller/more consistent).

It's been a long-time standing issue with ATI-drivers including games like R.A.G.E's launch where the ball was totally dropped. I switch back and forth between generations (Ie X900XT but an 9800GTX followed by a 5870) and do tend to lean towards a preference with nVidia drivers for being nicer in terms of UI, bugginess/strange occurences and performance/smoothness. I think recently with crossfire profiles and what-not, ATI has been making a move in the right-direction towards improving their drivers but the rate that they are improving them is too slow.
 
[H] Conducted a review of the latest drivers from last month and I never read that they had problems with stutter in the HD 7950.

http://www.hardocp.com/article/2012/11/12/fall_2012_gpu_driver_comparison_roundup/

[H] conclusion included the following
The most interesting stand out from this evaluation has been the NVIDIA GTX 660 Ti, which is priced significantly higher than the Radeon HD 7870 while turning in similar performance numbers. The GTX 660 Ti has lost its value at this point in time, lending way to the Radeon HD 7950 which has benefited from falling prices and new drivers.

unless you think much can change in one month.
 
I poo poo'd the "smoothness" argument when Brent made it his last review but once I examined the frame rate graphs I couldn't really deny that the Nvidia graph is considerably more linear than the AMD line. Some of the swings on the AMD chart were huge and would definitely be noticeable.

You mean the Far Cry 3 "review"? Yeah, it runs like shit and has a lot of problems on AMD cards right now. I see the same stutter at 4x msaa, even without it isn't super smooth. It crashes in tri and quad-fire. A lot of people are seeing artifacts in some spots.

You can't make a comparison based on one or two games.

Also you're pretty much asking the same questions going on in this "discussion".

What happened between August and December?

http://cdn.overclock.net/4/4f/4f774678_skyrim.gif

http://cdn.overclock.net/6/6c/6c00c781_skyrim-7950.gif
 
Last edited by a moderator:
You mean the Far Cry 3 "review"? Yeah, it runs like shit and has a lot of problems on AMD cards right now. I see the same stutter at 4x msaa, even without it isn't super smooth. It crashes in tri and quad-fire. A lot of people are seeing artifacts in some spots.

You can't make a comparison based on one or two games.

No, I meant the HardOCP "review" in total. A lot of those frame rate graphs show some pretty big swings with the AMD cards. So Im not basing it on 1 or 2 games. Im basing it on Brent and Kyle coming out and saying it was bad enough that theyd rather have the SLOWER Nvidia card because it played so much smoother. Now we've got a 2nd review site saying the same thing.
 
Oh, I couldn't care less about eyefinity results. That doesn't surprise me.

That doesn't appear to be a real issue with a single display.

Even then its tough to make too much of a decision based off of 3 games.
 
Single-card differences in frame latency consistency? Show me three games with proof of this, from two or more review sources please.

The 660 is absolutely no match for the 7950. To suggest otherwise is ludicrous. However, the response is not at all unexpected. Seizing on new oh-so-important-issues-that-Nvidia-may-be-marginally-better-at is the obvious refuge for the Nvidia fanboy when confronted with irrefutable chasms in both absolute performance and performance/dollar.
 
I gave up an HD 7970 for exactly this reason. NV is just smoother. Even Tri-SLI is pretty smooth.
 
Single-card differences in frame latency consistency? Show me three games with proof of this, from two or more review sources please.

The 660 is absolutely no match for the 7950. To suggest otherwise is ludicrous. However, the response is not at all unexpected. Seizing on new oh-so-important-issues-that-Nvidia-may-be-marginally-better-at is the obvious refuge for the Nvidia fanboy when confronted with irrefutable chasms in both absolute performance and performance/dollar.


Do you have this in a word document somewhere? I swear I've seen you post this same statement in any thread involving a comparison between AMD and Nvidia.

The 7950 is probably the best $300 value card in the past few years though, I liked mine except for the faulty mini display ports.
 
Single-card differences in frame latency consistency? Show me three games with proof of this, from two or more review sources please.

The 660 is absolutely no match for the 7950. To suggest otherwise is ludicrous. However, the response is not at all unexpected. Seizing on new oh-so-important-issues-that-Nvidia-may-be-marginally-better-at is the obvious refuge for the Nvidia fanboy when confronted with irrefutable chasms in both absolute performance and performance/dollar.

The price and FPS performance isn't really the point they're bringing up here. Its showing that a weaker card is still giving a smoother gaming experience in spite of the FPS being lower. This "might" be the same result regardless what AMD or Nvidia cards are used across their line up. If that's the case I don't think its some desperate fanboy attempt to say its important when considering which card to purchase. Its just another factor to take into account (including absolute performance and performance/dollar).
 
No, I meant the HardOCP "review" in total. A lot of those frame rate graphs show some pretty big swings with the AMD cards. So Im not basing it on 1 or 2 games. Im basing it on Brent and Kyle coming out and saying it was bad enough that theyd rather have the SLOWER Nvidia card because it played so much smoother. Now we've got a 2nd review site saying the same thing.

Its not the framerate (frames per second) swings that Techreport are trying to measure, but frametimes (time between frames). A lot of those framerate graphs shows pretty big swings with Nvidia cards as well. Techreport is trying to validate their testing methology with graphs and videos. Message is obviously coming poorly across the internet, since fanboys are trying to make this a blanket statement about Nvidia cards being smoother then AMD cards by default, while other fanboys are trying to discredit Techreport themselves.

For those that take those tests as gospel, go and get an AMD card, since Techreport has shown that AMD 7970 GHZ edition is smoothest:

http://techreport.com/r.x/radeon-hd-7970-ghz/value-99th-2.gif

For those that buhu Techreport, keep in mind that they have tested Skyrim before and they reported those findings as they found it, just as they did now. As you can see, on a previous driver set, Skyrim was pretty much even on AMD and Nvidia cards:

http://techreport.com/r.x/radeon-hd-7970-ghz/skyrim-1.gif

http://techreport.com/r.x/radeon-hd-7970-ghz/skyrim-3.gif

http://techreport.com/review/23150/amd-radeon-hd-7970-ghz-edition/7

Techreport found a difference with new drivers in a new review and they reported their findings as they were. When their findings were different, they reported it as such. Of course, they could have been more clear about that and that this is not a blanket statement, so not to confuse those that think this is a general thing.

If Techreport would have said that AMD cards were less smooth then Nvidia cards in general, they would have invalidated their own previews and methods, where AMD takes a marginal win in smoothness and performance as shown above.
 
Last edited by a moderator:
Since Tech Report noted that they felt stutter with HD7950 on 12.8 or 12.11 drivers, I went to look at their GPU reviews to even later reviews that didn't used older drivers for high-end GPUs. I found their HD7970Ghz launch review in June with Catalyst 12.7 beta.

http://techreport.com/review/23150/amd-radeon-hd-7970-ghz-edition/3

Quote from their review:

"We're relying on our 99th percentile frame time metric for our performance summation, but we've converted the result to FPS to keep our scatter plot readable."
http://techreport.com/r.x/radeon-hd-7970-ghz/value-99th-2.gif

^ What you see is HD7970 Ghz clearly delivering lower frame times than GTX680 or even an after-market pre-overclocked 680. In that review, I even looked closer to the individual tests, games that performed well on NV like Max Payne 3, BF3 or Crysis 2. No unusual stuttering on their HD7970 Ghz card at all. If AMD's drivers had these permanent issues in place, these games should have been stuttering on 7970Ghz compared to 680, but yet the results in that review are nothing worth talking about.

All these games were tested at 2560x1600 with AA and you can see the frame times are imperceptible in smoothness between 680 and HD7970Ghz cards, except Batman AC and Dirt Showdown where each camp nets a win.

http://techreport.com/r.x/radeon-hd-7970-ghz/bf3-99th.gif
http://techreport.com/r.x/radeon-hd-7970-ghz/max-99th.gif
http://techreport.com/r.x/radeon-hd-7970-ghz/crysis-99th.gif

I am not debating Tech Report's testing methodology but this idea that AMD cards stuttered all this time is being spread like the plague and it contradicts even TR's own testing not long ago. If you ask AMD/ATI users who switched back and forth between those brands and NV, I would bet you if they noticed some stutter fest on ATI/AMD cards, they would never go back after trying NV.

If we look at HD7950 vs. GTX660Ti review, the only game that's the same in that review and HD7970 Ghz vs. GTX680 review is Skyrim.

This is TR's HD7970Ghz vs. GTX680 testing of Skyrim with Catalyst 12.7 betas. Not only do HD7900 cards have no problems, but HD6970 is easily outperforming GTX570 under the same methodology. In fact, GTX570 significantly lost to HD6970 in smoothness in BF3, a game that was a staple for NV's good performance. So this idea that AMD cards stutter more as a generalization is unsubstantiated by real world evidence even from the same website!

http://techreport.com/r.x/radeon-hd-7970-ghz/skyrim-99th.gif

And now Skyrim retested with Catalyst 12.11 for 7950 vs. 660Ti:

http://techreport.com/r.x/7950-vs-660ti/skyrim-99th.gif

^ The Skyrim testing area is not the same, and I already noted how the frames per second seem way off to begin with as HD7950 boost demolished GTX660Ti in Skyrim in most other professional reviews that tested 2560x1440/1600 with AA, but still, if AMD cards had these stuttering driver issues for single GPUs all these years, it sure was nowhere to be found in TR's testing this summer. They even concluded that HD7970Ghz was faster and their frame times testing showed it was smoother too (See 1st graph).

This conspiracy theory that AMD was trading smoothness for frames per second all this time and that TR was the first website to expose them is not a valid one since even as of June 2012, these issues are nowhere to be found for single AMD GPUs. My guess is AMD's drivers for recent games have not been polished enough, which is not a surprising given AMD's financial struggles and layoffs, including you guessed it their graphics card department, and the fact that they are still betas. I personally have never felt any stutter difference for single GPUs between AMD and NV. I am currently running 12.8s because 12.11s cause instability for me in bitcoin mining on desktop. I wouldn't have continued buying AMD cards after my Fermi experience if my HD6950 @ 6970 was a stutter mess, that's for sure.

And as I said, frame times is one thing but the Skyrim frames per second are so off, it's almost mind-boggling how they got GTX660Ti to beat HD7950 in this game at high resolution with MSAA. Even a GTX670 max OC cannot even manage to come close to an HD7950 OC and GTX660Ti is way behind by nearly 40%!

This was as of August 2012.
1345736700tJwmf64Bk6_2_4.gif

Source

Once again, where is this frames per second differential in Tech Report's review? If the card is producing low frame rates, the frame times will be higher. There is still a relationship between them. How in the world did GTX660Ti not only make up nearly 40% deficit but is now beating an HD7950 in Skyrim at 2560x1400 + 4xMSAA? I can't explain that, I really can't.

TR's review is raising more questions:

1) HD7970/Ghz cards exhibited no such issues in June
2) HD7950 mysteriously gets creamed in Skyrim by GTX660Ti at high resolution with AA, contrary to 95% of reviews out there.
3) Older AMD cards like HD6970 often beat GTX570 in smoothness. Therefore, we know these developments are recent for single GPUs and in fact, most likely related to the specific games tested in that HD7950 vs. 660Ti review.

In other words, starting to generalize much from this review of HD7950 vs. GTX660Ti outside of those 2 cards and those specific games tested is not exactly confidence inspiring without other professional reviewers also vetting this data.

http://forums.anandtech.com/showthread.php?p=34366294#post34366294
 
Last edited by a moderator:
What happened is that AMD released a new driver that increased performance seemingly out of nowhere. I guess now we know what had to be compromised on in order to achieve that increase.

I don't "know" if this is linked to the performance increase, if its a bug, or even gotten some verification to Techreport's findings (though I find them interesting). If you have some evidence showing otherwise, it would be interesting to see.

What Techreport has shown me, is that there is more uneven frametimes on the AMD cards on the newest beta drivers vs. the older drivers according to their tests. Since they occur on one driver set, while doesn't occur on another, its likely a driver bug (since I consider Scott Wasson knowing enough about computer for this to be a human error).

What Anandtech has shown me, is

*some being butthurt about AMD's performance increase and are trying to make a marketing spin on this.
*some being butthurt because Scott Wasson had a review that was not favorable to AMD and therefore try to discredit him.
*some morons trying to make something more out of this, as a blanket statement regarding the hardware itself, completely disregarding previous tests showing that these issues weren't present in another driver set, but is in a beta (11) driver.

and me being a moron trying to read through those threads because there are a few posters that actually have some useful info on the new method Scott Wasson is trying to review cards with that I find interesting.


Thats one of the few posters there who tries to see and analyze the results and the tests themselves. :)
 
Last edited:
I still can't understand how is it possible to get this numbers:
http://www.overclock.net/content/type/61/id/1173104/
http://www.overclock.net/content/type/61/id/1173105/
I've been tinkering with excel, to check some random numbers, percentiles and what not. If you want to create a real roller coaster and you make something incredibly stupid such as a random chain of numers that start at 0 and finish at 33 you will get: (IE, frame times go from 0 to 33 ms... and now imagine how stupid that is :rolleyes: but its to go for the extreme).
a) Average framerate = 60 fps.
b) Average time frame = 16.6ms.
c) 99th percentile = 33ms.
My point being: its impossible to get a 99th percentile of 40ms AND and average of 60ms unless you go with negative numbers, that is. Its simply not possible to say that the 99th of the distribution sits at 40ms or below yet the average distribution sits at 16.6ms. Impossible. IM-PO-SSI-BLE. Mathematically it makes sense because you can work with negative data... but you can't have negative frametimes and, thus, none of this makes any sense.
So, I have no idea what TR did it... but they clearly have no QC whatsoever, because its entirely possible and very easy to check for inconsistencies, and frametimes, but:
a) You require to work with large sums of frame times (6.000 frames are a meaningless figure, because we are talking about 90 seconds of gameplay, which is close to nothing).
b) You require to dismiss outliers that might be created by whatever it is created... IE might be created by external factors.
c) You require to rinse and repeat until you find data that is consistent with your objective and can be obtained easily.
http://www.overclock.net/t/1337206/...d-7950-stumble-in-windows-8/120#post_18817131
 
Last edited by a moderator:
This thread already has more 'investigation' and speculation then I've seen thus far.

Nice compilation of various ideas around this.
:)
 
*some morons trying to make something more out of this, as a blanket statement regarding the hardware itself, completely disregarding previous tests showing that these issues weren't present in another driver set, but is in a beta (11) driver.

But you have to view this in the context of what those beta drivers have done for AMD. They provided enough performance to essentially reverse the common perceptions of which top-end card is faster - just in time for the Holiday shopping season.

If it's proven that this frame latency issue is completely unrelated to the performance increase in the beta drivers, then good for AMD. But given that the performance increase was quite large compared to past increases that have come from drivers, and these new findings; it would be a pretty big coincidence.

If it is proven that these issues ARE directly related then IMO there should be nothing short of a caveat in every article that has tested AMD cards using the new drivers.

You can't point to numbers from the new drivers as an example of how fast the cards are now, but at the same time, claim that the latency problem doesn't matter because you can simply use an older driver instead...

What's stopping them from coming out with a driver that includes the performance increases but doesn't have the latency problems? It will be very interesting to see which direction AMD takes with their drivers now, and either way, I expect it to be quite revealing.
 
99th percentile frame time is not an average.

I think you might be wrong in your interpretation of 99th percentile in this context. Here it is the time that 99% of frames are less than. So it should be completely possible to have an average of 16ms and a 99th time of 40ms. It would look like this: 12, 15, 12, 40, 15, 40, 13, 12, 65, 12, 13, 12, 9, 11, 13, 16, 12, 9, 10, 9

Your average is 16ms, and you throw out the 65ms outlier, so your 99th time is 40ms because 99% of the frames are less than 40ms. Obviously you'd need a bigger data set, but you get the point.

And I'm assuming the bolded part is a typo and you meant to say 60 FPS and not 60ms. Or did you do your math based on 60ms instead of converting the 60 FPS to a 16ms average? You wrote it both ways there.
 
But you have to view this in the context of what those beta drivers have done for AMD. They provided enough performance to essentially reverse the common perceptions of which top-end card is faster - just in time for the Holiday shopping season.

If it's proven that this frame latency issue is completely unrelated to the performance increase in the beta drivers, then good for AMD. But given that the performance increase was quite large compared to past increases that have come from drivers, and these new findings; it would be a pretty big coincidence.

If it is proven that these issues ARE directly related then IMO there should be nothing short of a caveat in every article that has tested AMD cards using the new drivers.

You can't point to numbers from the new drivers as an example of how fast the cards are now, but at the same time, claim that the latency problem doesn't matter because you can simply use an older driver instead...

What's stopping them from coming out with a driver that includes the performance increases but doesn't have the latency problems? It will be very interesting to see which direction AMD takes with their drivers now, and either way, I expect it to be quite revealing.

Its not a big coincidence. I've had Nvidia driver sets that had provided microstuttering without having any huge FPS improvements. Microstuttering in games can as example be caused by bugs like DPC latency without being related to higher FPS. There are also those that have had microstuttering from time to time in games that others haven't on same driver sets. You have long experience with GFX cards, so you know this already.

Saying that this is a result of performance increase, even though we have seen large performance increases by both Nvidia and AMD in the past without it being about microstuttering sounds more like parroting some Nvidia focus group members marketing spin on this.

Lets be better on [H] regarding this and see what the actual findings were and put the focus group tinfoil hats outside the door.:)

Techreport has an interesting method of reviewing that nobody else uses (except one time on Toms hardware). If this method has real world application, I personally would like to see it in more (or all) reviews. It would be interesting if another site would verify his finding as well.
 
Its not a big coincidence. I've had Nvidia driver sets that had provided microstuttering without having any huge FPS improvements. Microstuttering in games can as example be caused by bugs like DPC latency without being related to higher FPS. There are also those that have had microstuttering from time to time in games that others haven't on same driver sets.

People's subjective impressions of micro-stutter can be almost meaningless. Pick two random people from the forum and their definition of what micro-stutter even is is likely to be completely different. I'd rather discuss measurable data such as that provided by TR.

Saying that this is a result of performance increase, even though we have seen large performance increases by both Nvidia and AMD in the past without it being about microstuttering sounds more like parroting some Nvidia focus group members marketing spin on this.

I'm speculating based on what the actual data is showing me. Your speculation is based on....? Do you have any examples of any version of the recent beta drivers that are able to provide the performance increase without also causing this latency problem?

Lets be better on [H] regarding this and see what the actual findings were and put the focus group tinfoil hats outside the door.:)

Interesting that the TR results don't count as "actual findings" in your book. Easier to move the bar than to re-think an entrenched position in an argument I suppose.
 
I don't "know" if this is linked to the performance increase, if its a bug, or even gotten some verification to Techreport's findings (though I find them interesting). If you have some evidence showing otherwise, it would be interesting to see.

What Techreport has shown me, is that there is more uneven frametimes on the AMD cards on the newest beta drivers vs. the older drivers according to their tests. Since they occur on one driver set, while doesn't occur on another, its likely a driver bug (since I consider Scott Wasson knowing enough about computer for this to be a human error).

What Anandtech has shown me, is

*some being butthurt about AMD's performance increase and are trying to make a marketing spin on this.
*some being butthurt because Scott Wasson had a review that was not favorable to AMD and therefore try to discredit him.
*some morons trying to make something more out of this, as a blanket statement regarding the hardware itself, completely disregarding previous tests showing that these issues weren't present in another driver set, but is in a beta (11) driver.

and me being a moron trying to read through those threads because there are a few posters that actually have some useful info on the new method Scott Wasson is trying to review cards with that I find interesting.



Thats one of the few posters there who tries to see and analyze the results and the tests themselves. :)

There is no hope at anandtech for any reasonable discussion in the video forum. You've got an nvidia sponsored poster who promotes nvidia hardware, trolls amd products and just fills the forum with garbage on a perpetual basis. Then you have the extreme fanboys on either side and it all is allowed.

They need to ban that nvidia focus group marketer, they banned the other one that used to post there and ruin the forum in the same manner, no idea how that one still gets to continue with their trolling/pr spin. As well as deal with the extreme amd/nvidia fanboys who troll each other day.

This TR review is pretty plainly broken on a few fronts. Have seen so much conflicting data from every other review site that goes against it, even TR's own reviews from the past do not agree with this one. Yet it is being posted everywhere by paid nviida promoters ? We've all seen real data relating to inadequacies with dual gpu setups and microstutter, these single gpu claims of techreport and their arbitrary methods are all but worthless though.

Perhaps some other sites will try to duplicate them and not partake in broken methodologies as Techreport has, so we can get a real sense of where TR is going wrong/skewing the conclusions.
 
even TR's own reviews from the past do not agree with this one.

Because they used different drivers. The older drivers that didn't show any latency problems, also didn't come with any of the recent performance increases either...
 
Because they used different drivers. The older drivers that didn't show any latency problems, also didn't come with any of the recent performance increases either...

Take the time to look at their results with the same drivers as other sites have reviewed. Including here at [h], yet they are the only site getting 50% less FPS on the radeon 7950 compared to here at [h] and other sites.

The entire review is suspect for being performed poorly and further suspect because they are inventing their own methodologies and claiming what they can and cannot do....

I'm more than interested in reviews covering cards, but when you one review from one site that contradicts all the reviews at every other site, it's pretty obvious where the error is to be found. Hopefully more sites do the same and Techreport figures out whatever system issues/lack of experience is causing them to turn up these faulty results.
 
People's subjective impressions of micro-stutter can be almost meaningless. Pick two random people from the forum and their definition of what micro-stutter even is is likely to be completely different. I'd rather discuss measurable data such as that provided by TR.



I'm speculating based on what the actual data is showing me. Your speculation is based on....? Do you have any examples of any version of the recent beta drivers that are able to provide the performance increase without also causing this latency problem?



Interesting that the TR results don't count as "actual findings" in your book. Easier to move the bar than to re-think an entrenched position in an argument I suppose.

First of all, its TR's findings that I called "actual findings". Speculations and marketing spin you present are not actual findings.

You are talking about subjective impressions of microstuttering to be meaningless and then you go on talking about rather wanting to discuss measurable data as provided by TR. I can buy that, but then I must ask the question why you bring speculations about stuttering being caused by performance increase and not a beta bug, because its holiday season?

I'm giving you an example that DPC latency can cause microstuttering without being related to performance increase, and you ask me to provide example of a beta driver that provided performance without causing latency issues. Did you not understand the example?

Let me specify it for you then: I gave you an example trying to show you that microstuttering doesn't have to be related to performance increases, so its not given that any microstuttering on drivers that have performance increase is tied to the performance. It can simply be a bug. Its not given either, to specify it further, that microstuttering is caused by DPC latency. Thats just one example of what can cause microstuttering.

Unless you have something else to contribute regarding your speculations that AMD deliberately made the card microstutter because its holiday season and they wanted more
performance, lets close your speculations there, shall we? And discuss TR's findings instead and perhaps also the method itself? :)

There is no hope at anandtech for any reasonable discussion in the video forum. You've got an nvidia sponsored poster who promotes nvidia hardware, trolls amd products and just fills the forum with garbage on a perpetual basis. Then you have the extreme fanboys on either side and it all is allowed.

They need to ban that nvidia focus group marketer, they banned the other one that used to post there and ruin the forum in the same manner, no idea how that one still gets to continue with their trolling/pr spin. As well as deal with the extreme amd/nvidia fanboys who troll each other day.

This TR review is pretty plainly broken on a few fronts. Have seen so much conflicting data from every other review site that goes against it, even TR's own reviews from the past do not agree with this one. Yet it is being posted everywhere by paid nviida promoters ? We've all seen real data relating to inadequacies with dual gpu setups and microstutter, these single gpu claims of techreport and their arbitrary methods are all but worthless though.

Perhaps some other sites will try to duplicate them and not partake in broken methodologies as Techreport has, so we can get a real sense of where TR is going wrong/skewing the conclusions.

Thats why I hope to avoid it here. The GPU wars over there can be tiresome to read and there are some good posters which I want to read. I haven't even bothered getting an account on AT myself.

I don't wish to discredit TR's own reviews. That they have a bit conflicting data can merely show that he is reporting it as he see it then. He showed no microstuttering in previous driver and microstuttering in this. Both were displayed as it was. Thats a good thing in my book. :)

I find his methology interesting and though it might be done better (Anandtech's GPU editor talked a bit about the flaws in methology and hinted he was going to get some better equiptment for this), his is trying to verify his results with the tools he have at hand, so I am keeping myself open to that his method can have real world application as well. :)
 
Last edited:
I can buy that, but then I must ask the question why you bring speculations about stuttering being caused by performance increase and not a beta bug, because its holiday season?

Because the end result is the same either way. All of my speculation essentially boils down to one simple question: Can AMD produce a driver that has the recent performance increases but not the latency issue as identified by TR?

If it's a bug, then great. I never precluded that possibility. I look forward to reading a followup TR article where they test with new drivers, maybe even WHQL? Then we can see if the latency looks more like the betas or the older drivers, and if the recent performance increases remain.

Until then, it appears that the only way to avoid the latency issue is to revert to older drivers. If the current situation, bug or not, is that you can't get the recent performance increases without also having the latency issue, then that to a degree invalidates all of the benchmark numbers using those drivers - or as I said before at least deserves a caveat.
 
Because the end result is the same either way. All of my speculation essentially boils down to one simple question: Can AMD produce a driver that has the recent performance increases but not the latency issue as identified by TR?

If it's a bug, then great. I never precluded that possibility. I look forward to reading a followup TR article where they test with new drivers, maybe even WHQL? Then we can see if the latency looks more like the betas or the older drivers, and if the recent performance increases remain.

Until then, it appears that the only way to avoid the latency issue is to revert to older drivers. If the current situation, bug or not, is that you can't get the recent performance increases without also having the latency issue, then that to a degree invalidates all of the benchmark numbers using those drivers - or as I said before at least deserves a caveat.

Listen, I am not trying to pick a fight with you and I might come out a bit harsher then I would have wished because of all the marketing spin crap I read at AT at the same time about this subject. :)

My point is that there is no need for a spin on this. Here's a new post from Brightcandle showing less microstuttering from 12.10 to 12.11 because of a power bug being fixed:
http://forums.anandtech.com/showthread.php?t=2289404

Its most likely a bug and I haven't seen it reproduced elsewhere or seen it talked about in other reviews with the same driver set. I find it to be a large stretch if AMD should deliberately produce microstutter so they would get some framerate increase.

As you say, we'll see more in later reviews regarding your speculations. Pointless to discuss this at this point. :)
 
Nvidia has a fine engineering team. Their executive and spin branches, however, are nothing short of sickening. Anyone with any sense who has followed the events of the past decade knows that for a fact.

They're slimy liars, and their fanboys are even worse.

Even the fanboys may cry if AMD stops being competitive. And that's a frighteningly-possible situation, given the huge layoffs in engineering recently.


I don't give a shit about spin and about intangible accusations that are easy to level and difficult to disprove. I care about absolute performance, performance/dollar, and a reasonable level of noise/energy efficiency. Based on those logical metrics it just makes zero sense to purchase an Nvidia card right now. If and when that situation changes, so will my position. It hasn't yet.
 
What happened is that AMD released a new driver that increased performance seemingly out of nowhere. I guess now we know what had to be compromised on in order to achieve that increase.


You've established a nonsensical correlation between knockout drivers that completely shifted the balance of power, and highly dubious claims of single-card frame-time latency differences. Your conclusion is baseless and conspiratory. Nice try I suppose.
 
People gave Hardocp shit for saying that games feel smoother on NVIDIA hardware. Now it looks like TR has confirmed it.
 
They still haven't explained why their results are so far off base from the previous reviews that they published.

You have to question that.

People gave Hardocp shit for saying that games feel smoother on NVIDIA hardware.

In crossfire in eyefinity, that wouldn't shock me.

We're talking about single gpu on a single display here. That seems off to me.
 
They still haven't explained why their results are so far off base from the previous reviews that they published.

You have to question that.



In crossfire in eyefinity, that wouldn't shock me.

We're talking about single gpu on a single display here. That seems off to me.

It's looking like AMD sacrificed smooth gameplay to boost fps with their latest driver.
 
It's looking like AMD sacrificed smooth gameplay to boost fps with their latest driver.

What information are you basing that off of? Techreport hasn't claimed that and claims that this has been an issue since 12.8 which is odd since 12.8 is just a whql version of 12.7 beta that they had great results with.

It seems to me like you're jumping to baseless conclusions.

Until techreport narrows down why these results are so different from what they previously published you I'm going to have to question their results. I'm leaning towards a compatibility issue with their new test bench. It happens, I had to switch from a pair of 670s due to some driver issues.
 
single card was fine, now it's not.. not realy hard to see something changed in the meantimes

newer beta driver ? newer bios ? boost ? power tune ?
Pick your culprit, test , compare. Let's be more practical bout this and less fud
 
Last edited:
Back
Top