AMD Frame Pacing Explored: Catalyst 13.8 Brings Consistency to Crossfire
by Ryan Smith on August 1, 2013 2:00 PM ESTTotal War: Shogun 2
Our first detailed benchmark is Shogun 2, which is a continuing favorite to our benchmark suite. Total War: Shogun 2 is the latest installment of the long-running Total War series of turn based strategy games, and alongside Civilization V is notable for just how many units it can put on a screen at once. Even 2 years after its release it’s still a very punishing game at its highest settings due to the amount of shading and memory those units require.
For the sake of completeness we’re posting our frame rate charts for each of our individual games, but in general there’s nothing here we haven’t seen before in the 7990 review, in other reviews, or in Bench. The 7990 and GTX 690 still swap places fairly regularly.
Looking at our expanded delta percentages for Shogun, we can see how the 7990 and other Crossfire solutions stack up to the GTX 690 and other SLI solutions. For all AFR configurations the results match what we saw in our summary, with NVIDIA’s solutions offering lower deltas than AMD’s even with the new drivers.
This is actually AMD’s weakest game, with both the 7970GECF and 7990 exceeding 20% variability on this game. However it’s also the only non-action game in this collection, so it’s the game least affected by higher levels of variation and consequently the game AMD can afford to do the worst at. Nevertheless the improvement over Catalyst 13.6 without frame pacing is nothing short of amazing.
Meanwhile we’ll hit upon this a few times, but as a reminder AMD’s frame pacing improvements apply to older cards too, so the 6990 has its frame pacing problems resolved like the rest of AMD’s multi-GPU cards. It actually does better than the rest, we believe due to the fact that the lower framerate and higher frame times give AMD’s drivers more time to analyze and schedule frames.
Looking at the FCAT graphs, we can see that the higher variability of the 7990’s frame times is represented well. Though NVIDIA’s frame time spikes are more extreme than AMD’s.
Finally we have our 95th percentile frame times. Despite the fact that AMD’s framerates are down slightly versus Catalyst 13.6, their 95% percentile times are way up. Simply by instituting frame pacing they’ve dropped from 36.2ms to 21.5ms per frame.
Facebook
Twitter
RSS
102 Comments
View All Comments
chizow - Saturday, August 3, 2013 - link
Written on the walls at AMD Driver HQ I'm sure, quickly referenced when questioned about microstutter being worst on CF.krutou - Friday, August 2, 2013 - link
BUT, BUT, BUT, RADEON PRO?!?LordOfTheBoired - Saturday, August 3, 2013 - link
Took them by the hand? Looks more to me like "waited until people cared, then released a benchmark to prove they didn't have the problem, while offering no constructive commentary."And the hell of it is... AMD's stance makes sense. This IS a market that hates VSync because ZOMG LAG. "The market" has made it ABUNDANTLY clear that they have no interest whatsoever in technologies that improve the visual experience at the (real or imagined) cost of responsiveness.
But apparently that's only true when there's the visual equivalent of a record skip on your screen and not when it's a subtle frame rate fluctuation. The former is a good thing because it means you "aren't lagged", the latter is a horrible thing because it means you "aren't lagged".
chizow - Wednesday, August 7, 2013 - link
@LordOfTheBoired - this is the type of indignant attitude that got AMD and their fans in this predicament to begin with.mikato - Wednesday, August 7, 2013 - link
Well one result of Nvidia releasing the software was to allow tech journalists to shine a bright spotlight on a problem of their competitor's products. Your "holding AMD by the hand" idea is pretty amusing.novastar78 - Wednesday, August 7, 2013 - link
What's sad is that AMD has ripped ATi to shreds to the point where they are spread pretty thin. Just getting a working product out the door is a task. You are being waaay too hard on them.They were using traditional methods to test and frankly it's not so unimaginable that they were caught by this. Granted, maybe it should have been caught sooner, but to demonize them or put them down for it seems a bit harsh.
They know now that it's a big problem are definitely committed to fixing it. You can clearly see that they are trying to stabilize the company and there is lots of turmoil. The moves they are making and the people being brought on board are a good sign. There is definitely a process change that needs to happen but when the tree is being shaken so many times too many apples can fall.
Give it some time, I think we will see great things form them over the next few years.
Wreckage - Thursday, August 1, 2013 - link
It did not help that certain people were claiming that AMD does not have a stutter problem.DanNeely - Thursday, August 1, 2013 - link
The problem is that AMD didn't think they had a stutter problem; or more precisely, until 3rd parties shoved the reality in their face they assumed (without testing) that they and nVidia had equal amounts of the problem.I suspect at one point in the past they were right; and that the genesis of nVidia's "years in the making" tool to measure the problem dates back to when they discovered it was a problem internally and began working on fixes so that they could announce the same tool at the same time that their drivers had a negligible impact from it. It'd be interesting to see what would happen if someone tested SLI card/driver configurations from a few years ago to see how well nVidia did at the time.
BrightCandle - Friday, August 2, 2013 - link
NVidia has had this fixed for a lot longer than that. The 680's were noticeably smoother on their release day compared to the 7970 crossfire. NVidia has claimed they have been fixing this since the 8800 and there is no reason not to believe them as HardOCP and other review sites have been noting the difference for years.HisDivineOrder - Friday, August 2, 2013 - link
Exactly. This has been a problem for as long as there has been Crossfire. I think a lot of people are so shocked by this fact (because some of them owned Crossfire and just dealt with it and didn't realize they were getting shafted) they can't accept it.Yes, you got reamed. Yes, for years, you were using Crossfire and suffering from microstutter when you needn't have to. Yes, AMD took you for a ride. It's so infuriating it drives people not to want to believe it because to believe it would be so horrible as to suddenly be intolerable.
Do what most people do. Just stick with nVidia since you know they do proper testing. Boycott AMD for a five year period and come back once you're sure AMD's got their act together.