Fable Legends Early Preview: DirectX 12 Benchmark Analysis
by Ryan Smith, Ian Cutress & Daniel Williams on September 24, 2015 9:00 AM ESTDiscussing Percentiles and Minimum Frame Rates
Up until this point we have only discussed average frame rates, which is an easy number to generate from a benchmark run. Discussing minimum frame rates is a little tricky, because it could be argued that the time taken to render the worst frame should be the minimum. All it then takes is a bad GPU request (misaligned texture cache) which happens infrequently to provide skewed data. To this end, thanks to the logging functionality of the benchmark, we are able to report the frame rate profiles of each run and percentile numbers.
For the GTX 980 and AMD Fury X, we pulled out the 90th, 95th and 99th percentile data from the outputs, as well as plotting full graphs. For each of these data points, the 90th percentile should represent the frame rate (we’ll stick to reporting frame rates to simplify the matter) a game will achieve during 90% of the frames. Similar logic applies to the 95th and 99th percentile data, where these are closer to the absolute maximum but should be more consistent between runs.
This page (and the next) is going to be data heavy, but our analysis will discuss the effect of CPU scaling on percentile data on both GPUs in all three resolutions using all three CPUs. Starting with the GTX 980 Ti:
All three arrangements at 3840x2160 perform similarly, though there are slight regressions moving from the i3 to the i7 along most of the range, perhaps suggesting that having an excess of thread data has some issues. The Core i7 arrangement seems to have the upper hand at the low percentile (2%-4%) numbers as well.
At 1080p, the Core i7 gives greater results when the frame rate is above the average and we see some scaling effects when the scenes are simple (giving high frame rates). But for whatever reason, when the going gets tough the i7 seems to bottom out as we go beyond the 80th percentile.
If we ever wanted to see a good representation of CPU scaling, the 720p graph is practically there – all except for the 85th percentile and up which makes the data points pulled out in this region perhaps unrepresentative of the whole. This issue might be the same issue when it comes to the 1080p results as well.
Facebook
Twitter
RSS
141 Comments
View All Comments
medi03 - Thursday, September 24, 2015 - link
"AMD had driver with better results, but we didn't use it", "oh, Bryan tested it, but he's away" adds some sauce to it.Oxford Guy - Thursday, September 24, 2015 - link
"we are waiting for a better time to test the Ashes of the Singularity benchmark"L-O-L
Frenetic Pony - Thursday, September 24, 2015 - link
This is as usual a trolley, click bait response. The truth is far more complex than whether one side "wins". Here we can see Amds older card once again benefiting greatly from Dx12, to the point where it clearly pulls ahead of Nvidias similarly priced options. Yet on the high end it seems Nvidia has scaled it's gpu architecture better than Amd has, with the 980ti having the advantage. So, technology, like life, is complicated and not prone to simple quips that accurately reflect reality.jospoortvliet - Friday, September 25, 2015 - link
True. One thing has not changed and becomes more pronounced with DirectX12: AMD offers better performance at every price point.Th-z - Thursday, September 24, 2015 - link
Are you referring to AotS results? That benchmark stresses different things than this classic flyby benchmark, both are useful in their own right. AotS's more akin to real gameplay and stresses draw call capability of DX12, which is *the* highlights of DX12.My question for Ryan, Anandtech didn't test AotS because you said it's still in the early developmental stage, however this one isn't? I say test them all, make interesting before-and-after study cases regardless. Also have you considered improving your comment section?
DoomGuy64 - Friday, September 25, 2015 - link
Incorrect. The $650 Ti is the only Nvidia card better in dx12, and it has 96 ROPs, compared to the Fury's 64, not that Fury is actually doing that bad. AMD on the other hand, is cleaning up with the mid-range cards, which is what most people are buying.masaville888 - Saturday, October 10, 2015 - link
I left AMD for good after too many years of technical issues such as artifacting, poorly optimized drivers and so forth. I always had a good experience with nVidia and after going back I have no regrets. AMD seems more interested in tech announcements than user experience. If they figure out the customer side of things they have the potential to be great, but not until then.lprates - Thursday, October 15, 2015 - link
I totally Agreelprates - Thursday, October 15, 2015 - link
I totally Agreelprates - Sunday, October 18, 2015 - link
I totally Agree