Far Cry Primal: PC graphics performance benchmark review - CPU Scaling - GFX perf vs Normal - Very High and Ultra quality settings

Graphics card performance vs other quality settings

What if you we play around with image quality modes ?  Well, you will have several presets ranging from Low to Ultra High. Aside from anti-aliasing, all of these image quality levels feel well-balanced. Anti-aliasing remains  a bit of a conundrum though and already activated on the Medium and high presets. What kind of effect will choosing lowering your quality rendering mode ? Have a look:

Above a GeForce GTX 980, I did not opt the Ti model as I wanted a 4GB graphics card (which most high end users use). Take a good look at our settings and let's focus on 2560x1440 as I assume that's the monitor resolution you would use. Yes even Ultra quality mode remains playable with a GTX 980. Now I included Full HD and Ultra HD as well. For mainstream and upwards we recommend you to use the High quality level. Ultra settings you can apply up-to 2560x1440 but Ultra quality in Ultra HD becomes an issue VRAM wise:

At Ultra HD the GTX 980 ran into stuttering (and overall it starts lacking performance) with Ultra quality though, it's VRAM limited at thatparticular resolution with Ultra quality mode. All of the 4GB is eaten away, making it GTX 980 Ti and Titan X gaming ready only. So for most high-end cards UHD at very high quality settings is recommended as maximum. If you have more VRAM, Ultra quality becomes a reality at 3840x2160 pixels.

CPU scaling

We received the question if performance would differ based on the number of CPU cores. Our X99 based motherboard can independently control the number of CPU cores of our Core i7 5960X in the sense they can be disabled / enabled as well as Hyper-Threading.

On 2 cores with hyper-threading (SMT) disabled we did notice a HUGE increase in game load time, after ten minutes the game still had not finished so we aborted. Enabling Hyper-threading (two cores + 2 HT threads) solved this and the game started working again. With Very High image quality settings we only slowly see a bit of an effect in the average framerate, yet we are talking 2 to 3 FPS differences here in-between two and eight cores enabled in Full HD. Our processor is overclocked to a fixed 4.4 GHz per and for each CPU core. Short conclusion, running the game with a modern quad-core excellent. Interesting to see is that often HT disabled increased the framerate over HT enabled. I must note that the internal benchmark itself is not the best test for CPU scale testing being a scened loop. But overall, anything four core with a decent clock frequency will make this game run fine.