11 Replies Latest reply on Apr 29, 2018 10:50 AM by Richard Gergely

    Poor performance benchmark test

    Tim Pangburn

      I recently upgraded my desktop. Previous was a 2012 xeon processor, 16mb DDR 3, Quadro K4000 processor.
      new desktop is a beast. HP Z6 workstation. Dual Bronze 3106 xeon processors. 32gb DDR4,  Quadro P4000 processor.

       

      It's entirely possible I'm not understanding the performance test, but to me it looks like the new computer is significantly slower. If that is the case, is there something I'm doing wrong? Drivers recently updated, nothing running in the background.  New computer is on the left.

       

      test.JPG

        • Re: Poor performance benchmark test
          Francisco Martínez

          Solid Works is a single threaded application, it will only use one core of your dual processors. ( so 1 out of 12 or 16 cores, how ever many you have)

           

          The only possible performance increase is at the clock rate, I went from 4.0ghz to 4.8ghz and saw a small performance increase.

            • Re: Poor performance benchmark test
              Tim Pangburn

              Every search on the same topic tells me solidworks is not a single threaded application, so I'm not sure if that's correct. Also, I'm not comparing apples to oranges. Both computers were multi core. Memory has gone up significantly, and graphics card is a significant upgrade. I have 0 understanding why the graphics time would double.

              • Re: Poor performance benchmark test
                Richard Gergely

                Francisco Martínez wrote:

                 

                Solid Works is a single threaded application, it will only use one core of your dual processors. ( so 1 out of 12 or 16 cores, how ever many you have)

                 

                The only possible performance increase is at the clock rate, I went from 4.0ghz to 4.8ghz and saw a small performance increase.

                That's not correct. Clock rate is only relevant on the same generation of processor/MB/RAM or we would all still be using 4.0ghz processors from 15-20 years ago.

                 

                At a very basic level say intel brought out a chip which was exactly the same architecture as yours tomorrow but the only difference was semiconductor technology of 10nm instead of 14nm in all likelihood a 3.5ghz processor would be faster than your current one.

                  • Re: Poor performance benchmark test
                    Francisco Martínez

                    Richard Gergely wrote:

                     

                    Francisco Martínez wrote:

                     

                    Solid Works is a single threaded application, it will only use one core of your dual processors. ( so 1 out of 12 or 16 cores, how ever many you have)

                     

                    The only possible performance increase is at the clock rate, I went from 4.0ghz to 4.8ghz and saw a small performance increase.

                    That's not correct. Clock rate is only relevant on the same generation of processor/MB/RAM or we would all still be using 4.0ghz processors from 15-20 years ago.

                     

                    At a very basic level say intel brought out a chip which was exactly the same architecture as yours tomorrow but the only difference was semiconductor technology of 10nm instead of 14nm in all likelihood a 3.5ghz processor would be faster than your current one.

                    Yes you are correct,

                    I should have said with in the most current generation or vs a similar generation chip.

                     

                    ohh by the way where can I buy a intel 10nm chip?

                    Last I read they were delayed again. I mean samsung was at 10 nm last year this time(Exynos 9 Series 8895)

                      • Re: Poor performance benchmark test
                        Richard Gergely

                        Who knows it's like watching paint dry. I hear cost is the biggest problem for a new production plant.

                         

                        I gave up waiting TBH for big increases in processors and have a new workstation delivered last week waiting to be turned on - in the box still. Went with the new w-2125 processor, the top 4 core i7 maybe at a push would be a tiny bit faster if it's overclocked but even then it wouldn't be anything you could notice in practice.

                        My existing rig is 6 years old and the worry it will go pop grows by the day. You can bet it would go pop at the least convenient time. So I gave in.

                         

                        If nothing else at least all the rest of the tech is moving. M.2 nvme are crazy fast now.

                  • Re: Poor performance benchmark test
                    Francisco Martínez

                    To be honest never heard it from the horses mouth, But I have seen it visually and also did extensive research.

                    This is one of the best reviews I have seen, they ran tests and posted results.

                    https://www.pugetsystems.com/labs/articles/Solidworks-2016-Multi-Core-Performance-741/

                    "

                    Conclusion

                     

                    One thing we want to make sure is very clear is that the majority of tasks (including creating/editing/rotating/viewing models) in Solidworks are single-threaded so there is no performance gain by having a higher core count CPU. The purpose of this article is not to examine those common tasks at all, but to look at the tasks that are often claimed to be multi-threaded. Overall, we can come to four conclusions:

                    1) Opening and saving files (both assembly and drawings) is multi-threaded, but is limited to the point that there is no actual benefit to having more than a single CPU core. Simply put, something else (whether it is hardware or software) is the bottleneck in our test system, not the CPU. Considering that the CPU we tested with only has an operating frequency of 3.1GHz and we are using a fast SSD storage drive, we can fairly confidently state that unless you have a very low-end CPU your processor is not going to be the limiting factor when it comes to opening and saving files.

                    2) Running simulations only sees a very moderate benefit to having more CPU cores, but there is a drop in performance with multiple physical CPUs. For a single CPU, we saw about a 70% multi core efficiency for FEA simulations and about a 75% multi core efficiency for flow simulations which is bad enough that a high frequency quad core CPU will match the performance of even the most expensive Xeon E5 CPU. In addition, for all three simulations we saw a significant drop in performance with multiple physical CPUs so a dual Xeon system will actually be slower than a system with a single CPU.

                    3) Mesh creation sees a small benefit to having more CPU cores, but is extremely limited. At worse we saw no benefit to having more than a single CPU core and at best we saw a 60% multi core efficiency for the first 8 cores. In two of the three tests we saw a point where more cores actually reduced performance, while the third simply showed no benefit to having more cores.

                    4) Rendering is extremely good at utilizing multiple CPU cores. Rendering is really the one place in Solidworks where a higher core count CPU (or multiple physical CPUs) would actually show a significant improvement in performance. For a single CPU, the efficiency is pretty much perfect with a ~100% multi core efficiency. The efficiency did drop when using multiple physical CPUs (down to about 95.5%), but that is somewhat to be expected given the overhead associated with multiple CPUs.

                    What all this means is that the choice of CPU for Solidworks is actually fairly easy:

                    Even for the multi-threaded portions of Solidworks (including simulations), we would recommend getting the latest generation, highest frequency quad core CPU you can. At the time of this article, this would be the Intel Core i7 6700K 4.0Ghz Quad Core.

                    "

                    • Re: Poor performance benchmark test
                      Matt Furches

                      Do both computers have the same kind of hard drive?

                      • Re: Poor performance benchmark test
                        Jim Sculley

                        Tim Pangburn wrote:

                        Drivers recently updated

                        Recently updated to what?  Make sure you are using the SOLIDWORKS certified drivers for your card.  The newest drivers are not always the best.

                          • Re: Poor performance benchmark test
                            Tim Pangburn

                            I use the drivers from solidworks Rx.

                             

                            For comparison, previous processor was a Intel® Xeon® Processor E5-1620. I would expect a 6 year old processor to be significantly better. right? Graphics card is also a huge upgrade. it's not adding up.


                            Both computers had SSD.

                              • Re: Poor performance benchmark test
                                Richard Gergely

                                Tim

                                You bought the wrong machine/processor.

                                 

                                HP Z4 G4 is what you needed (cheaper)

                                 

                                Then the direct replacement for a E5-1620 is the W-2123 which would see decent performance increase.

                                Even better still the W-2125 which at this time has the fastest passmark result for any 4 core processor which hasn't been fiddled with.

                                 

                                Edit - The other problem is that the processor is the bottleneck outside rendering in Solidworks. So you could a even better put a better graphics card on still and see no better result.

                                 

                                And yes processors have not come on as fast as we would normally expect. OK the E5-1620 when it came out 6 years ago was really very decent for a 4 core. Over the years in it's various v2,v3 etc forms it hasn't improved much. Hence when even today the original first version passmark result isn't socking slow by todays standards.