how much of a speed up are you looking for relative to what you're going to spend on the hardware?
how big are the problems that you are solving right now?
what physics are part of your model?
have you already optimized the analysis?
there are some basic benchmark results in the solidworks kB that will probably point you towards not spending your money on a new processor. I'd expect less than 5-10% improvement just off the top of my head. solve time decrease will come from optimizing your analysis, not throwing hardware at it. But if you do want to go down this route, i would ask your hardware var for some test units and do the test directly. That is the only true way to determine what the advantage will be because of the variables I list above.
I order one should have it next week. Currently have the 3930k at 4.6.
I have put up some results for the FEA Bench. Do you have a model for flow simulation yet? I have been pretty lucky and have a stable 4.6 overclock.
I guess if you have a look at the article you found with reference to the Euler 3D Fluid Dynamics they recorded a 24% increase in performance. There is 25% more cores in the 5960x and the result from solidworks is 120/163= 26.3% faster. I also have a higher clock speed so not really liner compared to your result.
Ok Hayden I just changed my mind. I thought I would run a study that I did last week and it was quite a big job. It is only a structural static analysis but just meshed and solved in 50% of the time. The mesh has 1.3mil elements and rotates like its a extruded cube. I was even able to stop the solver without it going into meltdown. It just asked if I wanted to cancel and that was it. So if you have a need to solve large problems then go for it upgrade.