The 7970 is the clear winner in that TechReport review, it's within 2 FPS of the titan in 2 of those games at those settings.
Were it loses it's like 31 fps vs 26 or 28 fps vs 21 and 23 fps vs 17 fps
You should read the review and other charts. The clear winner was the 690 except in Crysis 3, then Titan wins by that review. Read the bottom of each page. The new review methods with FCAT require you to read a bit more. Looking at the frame latency chart is interesting. Nvidia mostly has flat lines, while AMD has slopes. That means a lot of variation in frame deliver still exists, and why in most they found the 690 was smoother in person.
Maybe the settings for the games in this review are lost to you. At those settings the single gpu cards even the mighty titan are struggling, bordering on unplayable so of course the 690 is going to be best. Unlike you I cant simply ignore the fact that it's $400 vs $1000. Did you read the bottom of the Bioshock page?
To prevent the dip to 30 fps (1/2 60hz), just use dynamic vsync. Hence, if the cards are able to sustain >60 fps to cause stutter, it wont, as vsync is on. If during scenes which cause it to drop <60 fps to deliver a viewable 30 fps, vsync is turned off.
Ive tested a 59 fps frame limiter as well with radeon pro, it works too.
There are other games other than Crysis 3 and when you lower the settings for playable FPS, the dual GPU setups become playable with less adjustments.
Bioshock was the only one in which the 690 didn't win, however, as you pointed out, the settings aren't really playable (I dislike it when they bench at unplayable settings). However, in other reviews, used with playable FPS, everything changes: http://www.tomshardware.com/reviews/radeon-hd-7990-review-benchmark,3486-6.html
Reading the conclusion a 2nd time kind of brings to light what the benchmark setup was about. It wasn't really about testing the cards under normal use, but under extreme conditions to see if micro-stutter is better or worse. I guess that is why the tests were performed in such a matter.
I'd like to vote for AMD but it's £900 in UK or 3 x7950 for £720, A year ago maybe at this price but today it's madness
Not really, I think it was more about were a single card fails to be playable would the addition of a second in Sli or XF help, so in the case of a 680 to 690 then yes for the most part and the 7970 to 7990 not so much.
Sure, we're only talking about five games, tested under specific conditions. But we created these conditions in order to answer a pressing question about the impact of multi-GPU microstuttering. We've had the tools to detect its presence for a little while, but does microstuttering really have a negative impact on gameplay? The answer appears to be yes. Also, microstuttering tends to grow worse as frame rates drop, calling into question the true value of multi-GPU schemes like CrossFire and SLI.
Nvidia has mitigated the effects of multi-GPU jitter via its frame metering capability, and that feature appears to work reasonably well most of the time. The GeForce GTX 690 is tangibly superior to the single-GPU GeForce GTX 680 in each of our test scenes, although the difference is pretty minor in Crysis 3. That case is a reminder that frame metering and pacing schemes aren't perfect. The 690 has near-pristine frame delivery in Crysis 3, but the smoothness of the animation is compromised by a see-saw pattern of frame dispatch, as we measured with Fraps. Fortunately, such early-in-the-pipeline jitter is usually confined to small spans of time—just a handful of milliseconds—on the GTX 690 and similar frame-metered SLI solutions.
Maybe it is a bit of both, this was on the last page of the review:
Yes and Vsync fixes ALL of this stuttering.
Im sure all the Tech websites just ignore Vsync because they would have nothing to write about if they took any notice
Not at those FPS levels. It might even add to the stutter.
What FPS levels?
Basically if your GPU cant do 60fps 95% of the time then your GPU isnt powerful enough and your settings are to high.
I often think that the whiners and the moaners are playing at 40fps all in the name of 4xMSAA.
I ran a 460 SLI for 2 years and never had any issues because it was setup correctly and my FPS was always 60. Vsync fixes stuttering.
Did you look at the review we were talking about? Only one game had over 45 FPS. What if someone doesn't want to lower settings to get 60 FPS?
I don't even know if Crysis 3 is capable of always being over 45 FPS, as I know of a couple people who have to use a 45 FPS limit on that game with two 7970's OC with voltage increases. While much of the game is over 60 FPS, it stutters like crazy in some areas.
Settings are up to high. Crysis 3 can be turned down like every game.
I have a Single 7970 on 1900x1200 and i finished Crysis 3 on 60FPS with only one level with the bugged graphics that caused my FPS to drop.
If you cant play at 60 FPS then your system isnt powerful enough. Everyone wants to max a game out but most cant. Soon as your frames hit the 40's then stuttering and spluttering is going to happen. GPU 100% and hot as hell isnt fun either. Screen will be tearing up and it will look crap.
Setup the game correctly and turn on Vsync with a 59 FPS cap and the game will run smooth and responsive. GPU will be between 80-95%.
Just because you like it that way, does not mean everyone does. You are making compromises to not have stutter. Good for you, but not everyone wants to compromise.
I'd rather play a game at 30fps and have it look like I want than crank everything way down to achieve 60fps.
They are not compromises.
Anyone can basically try and run a game that your system cannot handle.
Setting up a game correctly is no different than setting up a race car. They dont set the race car up so it can do 200mph but cant get around the corners without crashing.
If you play games at 30fps on a PC then i feel sorry for you. Really should buy better GPU or turn down the settings. 30 FPS and 4xMSAA or 60 FPS and 2xMSAA i know what id choose.
There is no such thing as "correctly" setting up a game. Everyone has a different opinion on what is the best compromise. You sacrifice visuals for higher FPS, other would prefer to sacrifice FPS for higher visuals.
I happen to have higher requirements than you on FPS. I shoot for 80 FPS, unless I'm in 3D, then I have no choice but to live with 60 FPS, and I tune to never drop below.
You have to look outside your own box some times. Not everyone has the same opinion as you.
Oh, condescension, my favorite.
This isn't about what you or someone else considers a worthless graphical effect, it is about people's preferences. It is arrogant to pity someone and deride their opinions because they are OK with lower FPS but more eyecandy. If someone is OK with 30 FPS that is their call, don't be so arrogant to tell them you pity them and that they should go play on a console.
Don't parade personal opinion as fact and deride anyone for daring to be different.
This isn't about what you or someone else considers a worthless graphical effect, it is about people's preferences. It is arrogant to pity someone and deride their opinions because they are OK with lower FPS but more eyecandy. If someone is OK with 30 FPS that is their call, don't be so arrogant to tell them you pity them and that they should go play on a console.
Don't parade personal opinion as fact and deride anyone for daring to be different.