For Honor Closed Beta Gameplay (FPS Test) - Fx 8350 & GTX 1060
Test System:
CPU: Amd Fx 8350
GPU: GTX 1060 6Gb Gigabyte Mini Itx
Ram: Corsair Vengeance 8GB 1600 Mhz
SSD: Sansung 750EVO 120GB
OS: Windows 10 (64 Bit)
Graphics Settings Explanation:
For Honor’s impressive visuals are derived in a few ways. The color palette is exceptionally orchestrated and tone-mapped, for one, but use of tessellation and high-detail textures makes for an environment that fits the game’s gritty undertones. Looking carefully at the scenes in our benchmark course (see: video accompaniment), we can see that For Honor also relies upon heavy geometric complexity with its more extreme settings. The graphics options, as they are present in the beta version of the game, are defined in their entirety below:
Quality Preset: Governs all options for quick settings configuration. Options are low, medium, high, extreme, and custom.
Texture Filtering: This is your normal anisotropic filtering. Although For Honor technically changes this setting as the graphics preset is scaled, AF is known to have effectively zero impact on the FPS of the vast majority of games. Anistropic filtering is a means to improve object and texture smoothness proportionate to the screen’s viewing angle, and has been around for years.
Anti-Aliasing: The method of smoothing over jaggies in the game. TAA is the preferred setting and functions temporally, or frame-to-frame, to better improve edge clarity during movement. This theoretically reduces shimmering/marching ants effects from traditional single-frame AA methods.
Render Scale: You’ll generally want this at 100%, unless under specific circumstances involving supersampling. Lowering render scale will reduce the rendered output of the game’s graphics while retaining the display resolution configured in the “Display” options menu. This would allow you to configure a 4K display resolution with a 50% render resolution, resulting in a 1080p render.
Geometric Detail: Like most settings with this game, you’re getting GPU impact with most things – but geometric detail does tend to serve as the first candidate for improving CPU performance in games with older APIs. Geometric draw calls must go through the CPU, and so reducing geometric detail would assist with CPUs that struggle to keep up with the GPU.
Texture Quality: As of the beta, texture quality is grayed out and stuck to ‘high.’ We believe that this is simply because the other texture resolutions haven’t yet been shipped with the title (or enabled), but the menu suggests that low and medium options will eventually be offered. Texture quality most directly impacts VRAM consumption in nearly every game that exists, as this setting almost exclusively refers to the resolution of the textures used. Higher texture resolution means higher VRAM residency by the game. Lowering this setting would help for users of 2GB video cards. We do see some differences between 3GB & 6GB variants of the GTX 1060, too, and this would be a place to look for improvement. More below.
Dynamic Shadows: As far as we can tell, the quality and edge clarity of shadows attached to dynamic host objects is impacted by this option. That’d be things like character shadows, where the shadow corresponding to character movement has its quality dictated by Dynamic Shadows quality.
Environment Detail: We haven’t studied this one too closely just yet, but it seems to be a mixture of decal and/or mesh detail for the static environment objects.
Dynamic Reflections: Reflection dynamism (on/off) for water and shiny surfaces.
Motion Blur: Self-explanatory, really. This toggles on/off.
Supersampling Anti-Aliasing: Defaults to ‘off’ in all graphics presets given the tremendous workload produced by the option.
Other Videos By Dibyajyoti Ghosal
Other Statistics
For Honor Statistics For Dibyajyoti Ghosal
At this time, Dibyajyoti Ghosal has 563 views for For Honor spread across 1 video. His channel published less than an hour of For Honor content, less than 0.51% of the total video content that Dibyajyoti Ghosal has uploaded to YouTube.