Dual Xeon's vs Single i7 Desktop CPUs in Rendering.

Below is a link to a page where Adobe Premiere Pro was benchmarked by many different users. Unfortunately they have stopped updating since 2014, but the test results are still relevant.
http://ppbm5.com/index.html

Looking a bit closer into the results and ignoring the CPU's marked as OC (overclocked)
(Bear in mind that overclocking is great when benchmarking a system and trying to beat a high score, but in reality these systems are not 100% stable and I certainly would not recommend it.)

According to the above test results we find the i7 3930K comes in at number 15 while the DUAL X5680 Xeons lag behind and come in at number 35

To give you an idea of the speed difference, using the following as a guide: http://www.cpubenchmark.net/high_end_cpus.html
the i7 3930K scores 12,061
while a single X5680 scores 8,983, so therefore 2 x X5680 should score 17,966

One would expect the Dual Xeon system should do a lot better than a single i7 CPU, but unfortunately this is not the case. The reason for this I believe is quite simple. XEONs are designed for SERVERS and Server software which are meant to deliver many hundreds of simultaneous requests for Data, but are NOT designed for everyday computing with software such as Adobe, Solidworks, Autocad etc.

 

16 Cores vs 32 Cores

BENCHWELL V3 is a benchmarking utlity for Maxwell render. http://www.maxwellrender.com/benchwell

In the above screenshot it is interesting to note that the SINGLE i7 5960X CPU (8 Cores - 16 threads) is on par with the DUAL XEON E5-2650 V2 System (16 Cores - 32 Threads). Remember the XEON system is 2 expensive XEONs on a very expensive motherboard that supports dual Xeon's while the i7 system is just a single 5960x CPU on a desktop Motherboard.

Another interesting point to note is the speed difference with increased amounts of RAM (memory)

In the above example we have the same i7-5960X CPU with 3 different amounts of available RAM, 14 GB, 31 GB and 63 GB. As you can clearly see a lot more RAM does not increase the performance of rendering.

 

 

Another interesting article found on the web is "single i7 MUCH faster than dual xeon E5-2650 v3 !!! "
http://www.cfd-online.com/Forums/hardware/144936-single-i7-much-faster-than-dual-xeon-e5-2650-v3.html

The post starts off with a member stating "....I'm posting this info in here so no one throw away they money as I've done..... I did perform the same analysis over a SINGLE i7 3820 and over a DUAL xeon E5 2650 v3 , and the SINGLE i7 is 2 to 3 time much faster !!!!! ....."

There were a few self professed experts who replied saying that his problem was RAM and so on.
After making some of the changes that they suggested his final post was "Thank´s Roberto, I do totally agree with you, XEON is not worth its price at all "

Another member wrote....
"I work on either a cluster computing with four nodes with 2 x 8-cores Intel Xeon E5-2650v2 2,6GHz each and on PCs with i7-3770 and i7-4790, and I can tell confirm you that already a single i7-3770 (a little old in this moment) is MUCH faster than two aforementioned xeons... when comparing them with the 4790, the ratio is almost 32 xeon cores to 4 i7-4790 cores, for Matlab, Ansys HFSS and CST. Maybe the xeon processors are made for servers, but for computing they just suck!.... People, if your scope is computing, don't buy XEON processors!!!!"