We don't know anything about the composition that you're working with.
Are you using the ray-traced 3D renderer? What effects? What kinds of footage?
It helps if you provide answers to the questions listed here:
My bad. I don't use any GPU accelerated effects or raytracing. I often work with psds as source footage or .movs. I use a lot of 3rd party plugins (sapphire glows, frischluft volumetrics, trapcode particular).
The compositions i am working in i have worked on (within one day) on both my i7 2600k rig and my dual xeon rig so i was directly able to compare the speed and cpu usage of the two systems.
Edit: I have just created a 4k comp with massive radial and lens blurs. After effects is using one thread at 100%, the 31 other threads are sitting idle.
Hey James, I just started doing some direct comparisons as well and looks like i'm getting similar results. I'll keep digging further, but yes my i7 rig is substantially faster than my dual xeon, in just about every aspect. I'm seeing the same 7-9% on the Xeon when trying to preview/RAM preview whatever we call it now.
However, If I take a 4k 16bit tiff sequence and ONLY apply radial blur to it and export. I can get after effects to use all cores to render the frame... which is exactly what I see on my i7. So I suspect it's something else that's causing the issue....
But I certainly agree with you that the i7 is much faster with the most recent update to AE (2015.3)