Ok, so I've installed the new beta patch, and I did some benchmarking with Fraps. I have made some chart primarily comparing Skyrim 1.3 (named Vanilla in charts) vs. Skyboost r3 vs. Skyrim 1.4beta. I did not include TESVAL in these comparisons as previous tests on my system showed that it is inferior to SkyBoost (not by much, though). I also tested how increasing ugrid from 5 to 7 had impact on the framerate. The last point, especially turned up some interesting results. The benchmark ran from the start of the intro sequence until the driver of the cart stands up after reaching Helgen. System specs: i7
[email protected], 8GB RAM, GTX590. Operating system on OCC Vertex 2 SSD, game is installed on old 5400rpm HDD. TESV 1.3 vs. SBr3 vs. TESV 1.4beta. UGRID=5 TESV 1.3 vs. SBr3 vs. TESV1.4beta. UGRID=7 SBr3 UGRID=5 vs UGRID=7 TESV1.4beta UGRID=5 vs UGRID=7 What's interesting in these graphs are that the performance before reaching Helgen (ie. outside town) is pretty much equal when ugrid=5, the new beta patch then takes lead as the player enters the town. There is an increase in FPS compared to SkyBoost r3, allthough this is only present inside the town. Now, when running ugrid=7, both the old 1.3 and the SkyBoost take significant drops in FPS, but there is almost no drop when running the 1.4 beta patch. This is pretty good, and there is now no reason for me to keep playing with ugrid=5 at all. Another way to look at it is to see how much time the FPS stays above 59 (v-sync cap at 60). This is a more real life test, allthough it doesn't take into account large and rapid variations in framerate which could create stutter. Comparing all measurements that I have done, the results looks very good for the 1.4 patch. In the table below, I have split the average FPS between the first half of the intro scene (first column) and the last half (second column). I have also counted number of measurements where FPS >59 (third column) and finally how much of the intro scene is rendered with an FPS >59.
