Destiny 2 Performance Review

Benchmarks: FPS, Frame Time & Smoothness

Benchmark Sequence

Destiny 2 does not have an integrated benchmark, so we had to create a test that represents what you'll see across various maps. Rather than using the opening scene, we collected data from The Farm. It's a controllable location with indoor and outdoor environments offering a large field of vision, and numerous reflection, shadow, and lighting effects. Though different from most of the other benchmarks out there, this sequence passes through several demanding areas that perfectly represent what you can expect from real-world gameplay.

Performance at 1080p

Given that Destiny 2 is supposed to be playable on modest configurations, our first results were run at 1920x1080 with quality set to Highest.

With these settings, the Radeon RX 460 2GB and GeForce GTX 1050 2GB can't maintain a minimum frame rate of 30 FPS. Moreover, you'll need a GeForce GTX 1060 6GB to reach 60 FPS.

On the Radeon side of things, RX 470, 570, 480, and R9 390 offer more or less similar performance, and they all trail Nvidia's GeForce GTX 1060 3GB. One last surprising finding: Radeon RX Vega 64 flirts with a 99 FPS average, but drops as low as 47 FPS, particularly in scenes with considerable depth of field. We suspect, however, that there is a driver bug causing these frame time spikes.

After we finished our benchmarks on RX Vega 64, AMD released its Radeon 17.10.3 drivers. Therefore, we redid our tests to see if the results would change. Sadly, they did not.

Performance at 1440p

We retained the same Highest quality preset for our 2560x1440 benchmarks.

Not surprisingly, Radeon RX 460 2GB and GeForce GTX 1050 2GB forfeit this round, with both cards running mostly below a 25 FPS average. On the other hand, Radeon RX Vega 64 makes a surprise comeback by surpassing the GeForce GTX 1080 FE!

In general, all of the Radeons, from RX 470 to the RX 480 (and including R9 390/RX 570), pass the GTX 1060 3GB and 6GB, whereas they trailed the Nvidia cards at 1920x1080. Without a doubt, AMD still has work to do on its drivers in order to improve performance at 1080p.

Day vs. Night

Destiny 2 offers a complete day/night cycle, and with different lighting effects at night, we wanted to know how performance would be affected at both test resolutions. Our quality settings remain unchanged.

Radeon RX Vega 64 pulls ahead of GeForce GTX 1080 FE at 1080p. Might this help shed some light on our daytime results at the same resolution? One thing is certain: some setting that was causing AMD issues during the day (shadows or ambient occlusion, perhaps?) is no longer an issue at night.

On the other hand, Radeon RX 470, RX 570, and RX 480 continue to trail both GTX 1060s.

At night and at 1440p, the rankings are nearly the same as during the day.


MORE: DiRT 4 Performance Review


MORE: Prey Performance Review


MORE: Mass Effect Andromeda Performance Review


MORE: Ghost Recon Wildlands Performance Review

Image
Destiny 2 (PC)
.
  • NinjaNerd56
    I’m running a GTX960 on High, and never get below 40-50fps in D2. Most of the time, I’m exceeding 100.

    I have 32GB of RAM, and have my whole house on a realtime SNMP/WMI monitor.

    D2 uses 10-12GB of RAM and drives the 4 cores in my I5 at 40-50% CPU.

    Looks awesome....
    Reply
  • SumTingW0ng
    Do you recommend this game? Should I get one?

    My setup is i7 4790k @ 4.5Ghz, 32GB DDR3 RAM, and ASUS Strix GTX 970.
    Reply
  • firerod1
    My evga 1060 3gb sc doesn't keep stable 60 fps on mixed medium/high settings and I have an i5-2500 that runs around 75-85% utilization with 6.5gb of RAM out of my 8 being used, Its been annoying.... to say the least. Probably going to upgrade to zen+ or i5-8600 h370 when it comes out.
    Reply
  • AnimeMania
    Will clearing out 68 Gig of SSD space to run the game provide any benefit or should it be run off the hard drive.
    Reply
  • clonazepam
    Disabling Depth of Field can give you back 10-20 fps average. I don't miss the effect in a first person shooter. I run it off of a normal hard drive. Once I figured out my normal overclock settings on the GPU just weren't going to work (I'm down to just +100 on both core and mem), it runs like a dream.

    The gunplay is just top notch. For normal play, you can 100% rely on the weapons, and add in the various abilities as you go. They aren't essential until you're in the harder modes and really looking to maximize teamwork and efficiency. It's a great game.
    Reply
  • nick sottiaux
    20411987 said:
    Will clearing out 68 Gig of SSD space to run the game provide any benefit or should it be run off the hard drive.
    Loading times are significantly improved while on my SSD. Running a 6700 paired with 16gb and a 1060. Takes from 20-25 seconds to load a map on SSD and close to a minute on my SSHD.

    Hope that helps

    Reply
  • Blytz
    Why does the iv card list say Vega 56 and the benchmarks are 64's ?
    Reply
  • lordsabathaan
    Why the hell did you use so old drivers for Nvidia since 388.31 give up to 53% performance increase in Destiny 2? And you update with trying new amd drivers , but you even dont use up to date drivers from Nvidia..
    Reply
  • cpt.pants
    These benchmarks are way off. I run a Ryzen 1700x and a RX 480 8gb. On 1080p with highest settings I average well above 100fps while playing solo and see a low of 80 fps during Public events or raids/crucible. Either toms is using bad drivers on amd hardware or they are just bias as usual towards nvidia.
    Reply
  • Yannick_G_THFR
    20414974 said:
    Why the hell did you use so old drivers for Nvidia since 388.31 give up to 53% performance increase in Destiny 2? And you update with trying new amd drivers , but you even dont use up to date drivers from Nvidia..

    388.00 was the last public version when all the benchs were done.

    20415809 said:
    These benchmarks are way off. I run a Ryzen 1700x and a RX 480 8gb. On 1080p with highest settings I average well above 100fps while playing solo and see a low of 80 fps during Public events or raids/crucible. Either toms is using bad drivers on amd hardware or they are just bias as usual towards nvidia.

    We always try to find some demanding areas for the benchmark sequence. Of course (and like you), our RX480 show better results - like all the others GPUs, so the final ranking doesn't change globally - in some places :)
    Reply