Dual Xeon Processors @2.2 GHZ vs core i7 4770HQ @2.2 GHZ

[MrBrN197]

Hello, I was recently testing a flip simulation on a workstation with dual xeon 2630 v4 processors 10 cores each clocked at 2.2 GHZ, to my astonishment my laptops quadcore 4770HQ @2.2GHZ delivered ,not the same, but better FLIP simulation times. Can anyone explain to me why this is. I heard someone tell me something about threading.  Can someone give me some insight on which nodes are multithreaded. with the smoke solver its stranger when my division size is higher like 0.2 the quadcore processor absolutely destroys the dual xeon, flying past the simulation. but when I start reducing that division size the xeon processors start to catch up and eventually when I am at a final render division size the xeons are ahead of the quadcore. Also when I cache out a simulation and check task manager when the cache is somewhere in the middle of the simulation(presumably where the flip simulation has the most particles to solve), task manager shows the cpu hovering around 2-3 %. 

[Guest tar]

the issue may the dual cpu need to send memory between each other, look up Numa. Also the 4770HQ turbos a bit more when single threaded.

[MrBrN197]

So why would a dual CPU ever be efficient if the time it takes to communicate between each other is a bottleneck. considering how expensive Xeons are nothing seems to justify there expense other than  marketing. 

[Guest tar]

That's too simplistic; Xeons are server chips built for multithreaded applications. It's up to the OS and the App to utilize the chip effectively. 

[MrBrN197]

That makes sense though. according to the specs the xeons should be 5 times faster and I can see it depends on the software because I am just doing some rendering comparisons and it seems to be always faster (not 5 times faster around 2-3 times). So i guess your right that its up to the software to take advantage. If you were to recommend a perfect processor for houdini, no budget. what would the specs look like; would it ever need to have more than 4 cores?

[Guest tar]

It depends on the part of Houdini you'll be using, and each Houdini version gets better at threading; so you want a high single thread speed for the UI interaction and parts of Houdini that aren't threaded yet, and you want more cores, i.e. total GHz, for things like rendering.  You have to pick which part you'll be using more. 

I'm not currently up to speed with the range of CPUs but there are quite a few threads here to search for, usually under Hardware.