3DMark 11 is a synthetic benchmark designed to test a GPU's performance at various aspects of DirectX 11 such as tessellation and DirectCompute. At both presets, ASUS's GTX 770 was only slightly higher than the reference card, due to its underwhelming overclock.
Our other three custom cards performed much better, each outscoring the reference card by about 6 - 6.5% at both presets. Palit was able to pull slightly ahead at both presets, but the difference was less than 1%.
For an idea of in-game performance, we chose Crysis 3, the toughest game in our benchmark set due to its extreme amounts of tessellation, per-pixel per-object motion blur, Bokeh Depth of Field, displacement mapping, particle and volumetric lighting and fog shadows.
As expected, ASUS was only slightly better than the reference card, while our remaining three custom cards fared much better, gaining as much as 11% over the reference card at 1920 x 1200 with 8xAA. Overall Palit maintained a slight edge in three out of four settings, but once again, the difference between all three was less than 1%.
For our overclocking efforts, in terms of pure clock speeds, we had the most success with MSI, which we were able to push to 1240MHz on the core. Despite this, we saw a bigger performance increase from Gigabyte, which we raised to 1237MHz, which gave it a slight 1% performance lead over its competition. Overall, all of our cards performed quite well here, gaining about 5 - 6% in performance, with ASUS performing slightly better, gaining about 9% in performance (due to its lower starting point), but it was still trailing in absolute results.
As an interesting side note, it seems that 7412MHz DDR is the sweet spot for memory clocks with the GTX 770 as all of our cards (including the reference card) performed best at this setting. Your own endeavors might vary.