• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Final Fantasy XV bench

Caporegime
Joined
24 Sep 2008
Posts
38,322
Location
Essex innit!
I will be getting this game, so thought I would have a play. No 3440x1440 option, which is a shame but the full game will have apparently.

388.31 - 6850K @ 4.0 - 1080Ti @2023/5508

2quorh5.jpg
 
Soldato
Joined
1 May 2013
Posts
9,692
Location
M28
I will be getting this game, so thought I would have a play. No 3440x1440 option, which is a shame but the full game will have apparently.

388.31 - 6850K @ 4.0 - 1080Ti @2023/5508

2quorh5.jpg
try this :)

ffxvbench
This patch allows you to customise graphics settings in the recently released FINAL FANTASY XV WINDOWS EDITION benchmark.

The benchmark normally sends performance data to Square Enix. This patch disables the data collection when running with customised settings so as to not taint their statistics.

Details
ffxv.exe commonly uses the following commandline parameters:

  • --graphicsIni <path>: Path to the graphics configuration file. By default, the program will only accept hardcoded paths to standard configurations (with hardcoded settings in the executable). This patch lets you specify any file on your system.
  • --displayResolution <width> <height>: Resolution of the window / display mode
  • --renderingResolution <width> <height>: Resolution the benchmark is rendered at
  • -f: Fullscreen mode
  • --locale=<en/jp>: Language
  • --loop_mode: Run in an infinite loop
The executable lists more options (--ui_lang, --numThreads, --numAsyncThreads, --720p, --2160p, --noNvidiaAfterMath) - these are not used by the original launcher (ffxvbench.exe) though. Feel free to experiment.

The default configurations (Lite / Standard / High) are provided in the release archive.
 
Soldato
Joined
9 Nov 2009
Posts
24,764
Location
Planet Earth
Well the benchmark did have some weird hitching and even GN with its own scripted tests saw the same:

https://www.gamersnexus.net/game-bench/3223-ffxv-gpu-benchmark-technical-graphics-analysis

It appears later runs are somewhat better.

Conclusion: FFXV GPU Results

These results are more cause for investigation. We’re digging into which graphics options specifically are causing the hitching behavior that’s present on both nVidia and AMD hardware. As it stands now, “High” settings, as a blanket, are exacting a detrimental impact on frametime performance and consistency. This is most severe on AMD cards, which are struggling to cope with – we’d assume – the GameWorks settings, although other graphics options do change alongside the GameWorks options. This is something that we are actively investigating, now that we’ve figured out how to toggle settings manually. We should have more soon.

There’s no doubt that the game is visually impressive. The achievement of the visuals, by and large, is done with heavily tessellating things like terrain and ground elements, which provide the apparent depth to the ground. Hair is also tessellated, and the shadow libraries (although the pop-in isn’t great) are pulled from GameWorks. That’s not to discredit Square Enix’s own contributions, though: Texture quality is impressive, and the huge amount of texture detail is indicative of precisely why the 4GB cards are struggling the most. That’s a lot of data to stream, and is also a major contributor in frametime spikiness on the first test pass.

GN,has done extensive testing of the game.

GameWorks & Hidden Graphics Settings in Final Fantasy XV Benchmark:

https://www.gamersnexus.net/news/32...aphics-settings-in-final-fantasy-xv-benchmark

We’ve been working on our Final Fantasy XV benchmarking and already have multiple machines going, including both CPU and GPU testing. This process included discovery of run-to-run variance, pursuant to slow initialization of game resources during the first test pass. We can solve for this with additional test passes and by eliminating the first test pass from the data pool.

One of the downsides to Final Fantasy XV’s benchmark is that there is no customization for graphics settings: You’ve got High, “Middle,” and “Lite.” Critically, the medium settings seem to disable most of the nVidia GameWorks graphics options, which will impact performance between nVidia and AMD cards. We spoke with AMD about a driver update for the game, and have been informed that updated drivers will ship closer to the game’s launch. In the meantime, we’ll be testing High and Medium settings alike, building a database for relative performance scaling between AMD and nVidia. That content is due out soon.

AMD cards will be apparently getting a driver update closer to when the game launches.

Final Fantasy XV (FFXV) Benchmark Variance Run-to-Run:

https://www.gamersnexus.net/game-bench/3220-final-fantasy-xv-benchmark-variance-run-to-run

In the meantime, here are some frametime graphs. This is ad-hoc testing on an 8700K at 5GHz, 1.42V, with 3200MHz CL14 GSkill Trident Z Memory. We are using a GTX 1070 Gaming X card for this test, which is running on the latest drivers. Windows version 1703 is being used. The blue bar demonstrates performance variability during the first pass, something which we must reconcile as we continue to run benchmarks. The frametime spikes are felt in the form of long, noticeable stutters in gameplay.

Be sure to follow our testing as we iterate. We are working all day on FFXV PC benchmarks.

FFXV Pretest: CPU numThread, SMT, NV/AMD GameWorks Scaling

https://www.gamersnexus.net/game-bench/3222-ffxv-pretest-cpu-numthread-smt-nv-amd-gameworks-scaling

The High preset is presently the only time that GameWorks graphics options are enabled, and two of those options supposedly remain disabled for the benchmark utility. The ShadowWorks library is disabled at present, as is voxel-accelerated ambient occlusion. That said, the same user who detailed these settings as disabled later posted a screenshot of the on-screen display, having hacked it to work in FFXV, and now believes that VXAO is enabled for 1080p/High settings. Either way, we previously detailed most of these graphics settings when they were unveiled, back at GDC 2016. VXAO converts the screen space into voxels based upon geometric data, which reduces the complexity present from raw triangles and primitives. VXAO then runs a cone-tracing pass for shadowing computation. The result is that ambient occlusion can theoretically be calculated more accurately, demonstrated with nVidia’s tank asset.

QYXqlcU.png

Let’s pull some quick data out of our upcoming GPU benchmark. This will look at relative performance scaling between the RX 580 and GTX 1060 6GB cards, switching between Medium and High settings. The idea is to see if relative scaling worsens with the higher settings, as that is where nVidia theoretically has more optimization. Keep in mind that more than just the GameWorks settings change between medium and high, but those are most likely to be drivers in performance deltas.

The GTX 1060 6GB card is baseline here, marked at 100% performance. The GTX 1070, under both medium and high settings, maintains 137% of the GTX 1060’s performance. It is almost equal for both presets.


As for SMT, we'll spoil that we've found performance uplift on the R7 1700 (stock and overclocked) by disabling SMT altogether. This seems to coincide with the numThread=8 performance uplift.
UPDATED Section: The RX 580 maintains 81.4% (original test indicated 60%, we discovered some more issues with the benchmark that caused us to rerun tests) of the GTX 1060 performance when using High settings, or 95.5% (original text said 66.6%) of the GTX 1060’s performance when using low settings. AMD is regaining ground at medium settings, which means that at least one of the settings enabled under “High” is more taxing for the RX 580 than it is for the GTX 1060. This comes down to shader-level optimization and/or architectural level differences, where shader-level optimization would also account for driver and library differences involving GameWorks. We cannot conclude that GameWorks is the cause of the 6-percentage-point disparity, but it is a likely contributor, as it is reasonable to assume that GeForce would process those nVidia effects with greater performance.

As for SMT, we'll spoil that we've found performance uplift on the R7 1700 (stock and overclocked) by disabling SMT altogether. This seems to coincide with the numThread=8 performance uplift.

Its probably worth reading the articles since they delve into performance in great detail.
 
Soldato
Joined
18 Oct 2002
Posts
4,571
Location
Mars
Found this info from a video on youtube, not sure if it works

Steps to run Final Fantasy XV in Custom Resolution :

1) Create Short cut of ffxv.exe (You will find it in the folder where you installed the benchmark)

2) In Target, enter the following line after the already present text : --graphicsIni config\GraphicsConfig_BenchmarkStandard.ini -f --displayResolution 2560 1080 --renderingResolution 2560 1080 --locale=en
 
Soldato
Joined
1 May 2013
Posts
9,692
Location
M28
Found this info from a video on youtube, not sure if it works

Steps to run Final Fantasy XV in Custom Resolution :

1) Create Short cut of ffxv.exe (You will find it in the folder where you installed the benchmark)

2) In Target, enter the following line after the already present text : --graphicsIni config\GraphicsConfig_BenchmarkStandard.ini -f --displayResolution 2560 1080 --renderingResolution 2560 1080 --locale=en
#50 ;)
 
Back
Top Bottom