Benchmarking my PC’s


So this post has been coming for a while. In between working, and setting up a new office, and just getting things off the ground, I’ve finally got time to sit down and write a post about the PC’s that I use, and the benchmarks they have, with different software, both CAD and non-CAD.

My main PC, is the WS63 Workstation class laptop from MSI. Its black, slim, and very very light for the kind of power it has. In short, it contains:

  • Intel Core i7-8850H running at 2.6GHz
  • 32GB of DDR4 RAM
  • Nvidia Quadro P4200 GPU with 8GB RAM

The machine that I’m going to compare to is an old HP Z800 that I have, that still gets used for some minor daily duties. I resurrected it from Gumtree around 12 months ago and just threw in an SSD and a AMD Radeon RX580 GPU, also with 8GB of RAM. The specs are:

  • Intel Xeon X5675 running at 3.07GHz (Dual 6-core Processors, 12 in total, 24 threads)
  • 64GB of DDR3 RAM
  • AMD Radeon RX580 GPU also with 8GB of RAM
MSI WS63 with AutoCAD Plant3D 2019

There is a variety of software to use for benchmarking. For CAD, I couldn’t really find anything that could benchmark the CAD software itself. So I just decided for AutoCAD and Revit, I figured Rendering would be the best option. There is a benchmarking tool for Inventor, but I primarily use the AEC tools, so AutoCAD and Revit are my main tools of choice for now. For non-CAD benchmarking, I used:

For the MSI, I did them all on the laptop, with no screen connected, and I did both 1080 and 4K tests. With the HP, I don’t have a 4K screen, and after seeing the results on the 4K tests with the MSI, it’s very taxing on the system, and I don’t think gives real-world CAD performance, so therefore the HP was just done with 1080 resolution on a Philips 32″ monitor.

Lets get started with Cinebench. As per the Maxon website, CINEBENCH is a real-world cross platform test suite that evaluates your computer’s performance capabilities. CINEBENCH is based on MAXON’s award-winning animation software Cinema 4D, which is used extensively by studios and production houses worldwide for 3D content creation.

What does the test involve? The fps score is based on what, according to Cinebench?

This procedure uses a complex 3D scene depicting a car chase (created by renderbaron) which measures the performance of your graphics card in OpenGL mode.The performance depends on various factors, such as the GPU processor on your hardware, on the drivers used. The graphics card has to display a huge amount of geometry (nearly 1 million polygons) and textures, as well as a variety of effects, such as environments, bump maps, transparency, lighting and more to evaluate the performance across different disciplines and give a good average overview of the capabilities of your graphics hardware. The result is measured in frames per second (fps). The higher the number, the faster your graphics card is.

Cinebench fps Score (higher is better)

The cb test in Cinebench is defined as:

The test scenario uses all of your system’s processing power to render a photorealistic 3D scene (from the viral “No Keyframes” animation by AixSponza). This scene makes use of various algorithms to stress all available processor cores. In fact, CINEBENCH can measure systems with up to 256 processor threads. This test scene contains approximately 2,000 objects which in turn contain more than 300,000 polygons in total, and uses sharp and blurred reflections, area lights, shadows, procedural shaders, antialiasing, and much more. The result is displayed in points (pts). The higher the number, the faster your processor.

Cinebench cb Score (higher is better)

So the result is pretty obvious. The MSI, being a later machine with a later more up to date processor, doesn’t handle as good a cb score as good as the Xeon, but the GPU in the HP is lagging, being the RX580.

Next for benchmarking is Heaven. According to the website, Heaven is defined as a GPU-intensive benchmark that hammers graphics cards to the limits. This powerful tool can be effectively used to determine the stability of a GPU under extremely stressful conditions, as well as check the cooling system’s potential under maximum heat output.

As per the Cinebench test, the higher the fps score, the better. As you can see in the image below, running the WS63 at 4K really kills the performance.

Unigine Heaven benchmark (higher is better)

Heaven also gives a ‘score’, similar to the cb score with Cinebench, higher is better

Heaven Scores (higher is better)

So again, its pretty obvious that the later machine, is going to win all around, but I think this is more on the GPU than anything.

Next set of tests were done with Geekbench. Geekbench is an app that needs to be installed and does CPU and Compute Benchmarks. The CPU Benchmark models a real-world task or application, ensuring meaningful results. These tests are complex, avoiding simple problems with straightforward memory-access patterns, and push the limits of your system and the Compute Benchmark measures the compute performance of your GPU. From image processing to computer vision to number crunching, Geekbench 4 tests your GPU using relevant and complex challenges.

Firstly, the CPU Benchmarks

CPU Single Core Benchmarks (higher is better)

Even at Single Core scores, the WS63 walks all over the HP Z800, by about double. Looking at the Multi-Core scores, it’s a bit closer. The graph doesn’t look like it, but there’s only around 2000 points between the WS63 and the Z800.

CPU Multi Core Benchmarks (higher is better)

In Geekbench’s Compute, we test the GPU’s and their scores are:

Compute Scores (higher is better)

Again, the Z800 isn’t that far behind, considering the difference in GPU’s. The AMD RX580 isn’t a CUDA enabled card, so there are no scores for it, against the Quadro P4200.

Now let’s look at the last two products, Revit and AutoCAD. Using both products, in general day to day use, there really isn’t any difference. Even with InfraWorks on the highest settings, you couldn’t pick the difference between the two. AutoCAD and Revit don’t really get into the GPU for Rendering, they used the CPU, but when using InfraWorks, the GPU is used extensively. I would tend to think that any half decent video card is going to suit CAD (previous to the WS63 I had the MSI GT75VR with the GTX 1080 GPU and it never missed a beat).

Revit Render on the HP Z800

Here’s the results for the Revit rendering. The model used was the standard Revit House, with the Render settings being

  • Settings: Best
  • Resolution: Screen
  • Scheme: Exterior, Sun Only
  • Sky: Few Clouds
  • Screen Resolution: 1920×1080
Revit Render @ 1080 (lower is better)

So you can see this is where the Z800 shines, but it is a 12-core machine, compared to the 6 cores of the MSI. So in CAD terms, the better the CPU, the quicker the render is going to be. Next on the list was the AutoCAD render.

I tried to look around for something that would be a ‘standard’ model for users to test, if they wanted to do something similar. I use Plant3D a lot, and the standard Plant3D Sample Project has an overall master model that looked good, to test the render time. Done again in 1080, the settings were defined as:

  • High Quality
  • Screen size output of 1920×1080
AutoCAD Plant3D Render @ 1080 (lower is better)
AutoCAD Plant3D Render on the Z800

Again, the CPU on the Z800 comes into play here. Its not by much though, is an extra 20 seconds worth the wait? Only you can decide the answer to that.

All in all, the WS63 is a very capable machine. I have a couple of gripes with it, moreso the port locations, they are on the sides of the machine, and just sometimes they are a pain when you have to plug in and out when travelling. I like the ports at the back, like I used to have on the GT75 (with some USB ports on the sides).

If you want more information on the MSI range of workstations, I’d be happy to load a model, test it, and see what its performance is like for your real world models, no obligation.

Thanks for reading!