Battlefield 4 Gameplay and Performance Preview
Performance and Frame Latency
We tested the BF3 Open Beta using the Radeon HD 7970 and the NVIDIA GeForce GTX 770. Graphics settings were tweaked using the BF4 Settings Editor and set to "Ultra" in all cases, with 4x MSAA enabled, high-quality FXAA, and 100% native scaling.
We used the BF4 Settings Editor for our configuration -- the Open Beta client refused to save in-game graphics settings changes normally.
We tested using FRAPS on a 64-player server in Buffalo, NY (geographically the closest server to my own home). Game frame rates were measured over the course of seven minutes.
We've also graphed the frame time latencies using FRAFS, a handy tool for displaying FRAPS frametime results. If you aren't familiar with frame timing, this is a metric of how long it takes to draw each frame. It shows variation within a single second -- something that FPS, by its nature, fails to catch. And the differences it captures are exceedingly useful. For example, here's the traditional FPS results (min, max, average) for the GTX 770 and the Radeon HD 7970 over an entire run.
Here, the GTX 770 falls about 10% back from the Radeon HD 7970, but this doesn't capture the full range of the performance difference. Let's look at frame times.
The top graphs here are from two separate Radeon runs (colored red) while the green results are for the two NVIDIA runs. See the two dotted lines running across the plot at 16.7ms and 33.3ms? This denotes how much time the GPU spends at or above 60 FPS (16.7ms per frame) and 30 FPS (33.3ms per frame). The Radeon cards only break the 30 FPS mark a handful of times. The 1% latency, meaning the worst 1% of all results, is 27ms for the Radeon cards, or 37 FPS.
The NVIDIA results, on the other hand, are quite different. The GTX 770 broaches the 30 FPS barrier many times in the same seven minutes, with a 1% frame time of 36.9ms, which works out to 27 FPS. This difference is apparent when you play the game; the NVIDIA cards seem to hitch and stutter randomly. It's not terrible, but it's definitely noticeable when racing through the disintegrating streets of Shanghai. The gap in 1% frame time results (27 FPS for the GeForce vs 37 FPS for the Radeon) is a much better indicator of the relative difference between playing with the two solutions.
Note that Mantle, which has been widely talked up these past few weeks, isn't actually in use yet. Both these cards are running in DirectX 11. Mantle's availability in BF4 is set for December, 6-8 weeks after the launch of the game itself, which hits October 29. Exactly how much performance benefit Mantle will deliver is unknown -- the benefit should be significant, or AMD wouldn't have bothered in the first place. 15-20% seems a reasonable estimate, but that's a guess on our part. Even so, the advantage should be significant.
What's more significant here is that the GTX 770 and the Radeon HD 7970 end up tied, despite the fact that the GTX 770 is a $400 card , while the Radeon HD 7970 or its doppelganger replacement, the R9 280X, are both available for $300. People rarely buy a GPU for just one game, but if you're a BF fanatic, the Radeon family, at least currently, is offering the better deal here.
We used the BF4 Settings Editor for our configuration -- the Open Beta client refused to save in-game graphics settings changes normally.
We tested using FRAPS on a 64-player server in Buffalo, NY (geographically the closest server to my own home). Game frame rates were measured over the course of seven minutes.
We've also graphed the frame time latencies using FRAFS, a handy tool for displaying FRAPS frametime results. If you aren't familiar with frame timing, this is a metric of how long it takes to draw each frame. It shows variation within a single second -- something that FPS, by its nature, fails to catch. And the differences it captures are exceedingly useful. For example, here's the traditional FPS results (min, max, average) for the GTX 770 and the Radeon HD 7970 over an entire run.
;)
;)
;)
;)
The top graphs here are from two separate Radeon runs (colored red) while the green results are for the two NVIDIA runs. See the two dotted lines running across the plot at 16.7ms and 33.3ms? This denotes how much time the GPU spends at or above 60 FPS (16.7ms per frame) and 30 FPS (33.3ms per frame). The Radeon cards only break the 30 FPS mark a handful of times. The 1% latency, meaning the worst 1% of all results, is 27ms for the Radeon cards, or 37 FPS.
The NVIDIA results, on the other hand, are quite different. The GTX 770 broaches the 30 FPS barrier many times in the same seven minutes, with a 1% frame time of 36.9ms, which works out to 27 FPS. This difference is apparent when you play the game; the NVIDIA cards seem to hitch and stutter randomly. It's not terrible, but it's definitely noticeable when racing through the disintegrating streets of Shanghai. The gap in 1% frame time results (27 FPS for the GeForce vs 37 FPS for the Radeon) is a much better indicator of the relative difference between playing with the two solutions.
;)
Note that Mantle, which has been widely talked up these past few weeks, isn't actually in use yet. Both these cards are running in DirectX 11. Mantle's availability in BF4 is set for December, 6-8 weeks after the launch of the game itself, which hits October 29. Exactly how much performance benefit Mantle will deliver is unknown -- the benefit should be significant, or AMD wouldn't have bothered in the first place. 15-20% seems a reasonable estimate, but that's a guess on our part. Even so, the advantage should be significant.
What's more significant here is that the GTX 770 and the Radeon HD 7970 end up tied, despite the fact that the GTX 770 is a $400 card , while the Radeon HD 7970 or its doppelganger replacement, the R9 280X, are both available for $300. People rarely buy a GPU for just one game, but if you're a BF fanatic, the Radeon family, at least currently, is offering the better deal here.
