If you don't want to install separate software Shmi, and don't mind waiting until the end for the results, you can also hit the "Save Score" button once the benchmark has completed a run, and it will dump a log into the benchmark's directory with your average FPS (among other relevant info that can be helpful in comparing multiple runs with different settings).
As for the benchmarks reliability, I'd have to respectfully disagree and say that in my case, I'm finding it to be a fairly accurate representation of performance (in terms of FPS, not the Score).
When comparing the average FPS in bench logs to the average FPS I had in Beta Phase 3, it's nearly identical (though the benchmark feels slightly more fluid).
The caveat to my disagreement being that I would be wary of comparing your results with other players' as:
- There are so many variables to take into account when it comes to client performance (don't think that just because you simply share the same CPU/GPU, you will have the same results).
- Knowing people can up their score just by minimizing the benchmark...you are trusting that their scores were gotten fairly (which may very well have not been the case where the internet is involved).
I would also very much disagree (again, respectfully) that SSAO provides no discernible difference in graphics, as it adds a graphical level of depth to a scene that is both very noticeable, and very enjoyable for me, both in XIV and many other games. When turned off, objects tend to look very flat by comparison, as if they were simply Photoshopped over a scene, rather than actually occupying it. The effect can be subtle at times, though most times I notice the difference to be quite glaring. However, one thing I will most certainly agree with is that Ambient Occlusion is a massive performance hog for most systems (except Connor with his quad-SLI Titans), and it is one of the first things I recommend turning off for substantial FPS gains in any game that supports it (as we discussed briefly in RaidCall the other day).
As with any setting, you will likely have to sacrifice some measure of graphical fidelity for performance gains. Whether you notice the difference or not, or whether you mind the sacrifice or not, comes down to each player - for example: I'll be playing with LOD options enabled. While I can notice the difference with those options off if I specifically look for them, in practice I usually focus so much on the immediate scene that I don't mind losing some detail in distant objects in exchange for the small FPS bump I get - especially when I play with Depth of Field (another "controversial" setting) on in cut scenes (where I would be more likely to notice those missing details, if not for the blur effect). As for Occlusion Culling, I just don't see the point of rendering objects that aren't in my immediate field of view, outside of overclocking/benchmark trials to push your system to its limits. So for actual gaming, that option stays on for me.
If the game feels and looks acceptable to you though, that's all that matters in the end (don't let "random internet person #4,739's" score/fps bum you out)!
» Edited on: 2013-08-03 03:18:22