Jump to content

Apple macOS 10.15 vs. Windows 10 vs. Ubuntu 19.10 Performance Benchmarks


steven36

Recommended Posts

In addition to this month bringing the release of the Ubuntu 19.10 "Eoan Ermine", Apple also shipped macOS 10.15 "Catalina" as the sixteenth major release of their macOS operating system. So with that it makes for an interesting time seeing how macOS 10.15 competes against both Ubuntu 19.10 and Windows 10 on an Apple MacBook Pro. Here are those results from dozens of benchmarks.

 

125760869_157245043644524452.jpg

 

Using an Apple MacBook Pro with Core i7-6700HQ Skylake CPU, 2 x 8GB RAM, 250GB Apple SSD, and Radeon Pro 450 graphics, macOS 10.15, Windows 10, and Ubuntu 19.10 were all benchmarked off this same system. All three operating systems were tested with their latest software updates as of testing.

 

125760898_157245043644524452.jpg

 

Normally we also include Intel's Clear Linux too for looking at the performance of a very optimized rolling-release Linux distribution, but unfortunately Clear Linux was having issues with its UEFI boot configuration in being able to boot on this MacBook Pro.

 

125760909_157245043644524452.jpg

 

It's worth noting that finally in Ubuntu 19.10 is good support for recent MacBook Pro laptops and other Linux distributions using the Linux 5.3 kernel or newer. The Linux 5.3 mainline kernel brought a driver for handling the Apple keyboards and trackpads for the MacBook/MacBookPro devices over the past couple of years. So with the newly released Ubuntu using Linux 5.3, the 2016 era MacBook Pro and newer are finally working decently out-of-the-box.

 

Via the Phoronix Test Suite more than 80 benchmarks were run on macOS 10.15, Windows 10, and Ubuntu 19.10 from this MacBook Pro.

 

 

First up was Valve's Dota 2 game that runs well across Windows/Linux/macOS. With Dota 2 using the OpenGL renderer, both Ubuntu 19.10 and Windows 10 were much faster than macOS 10.15. No surprise really considering Apple has deprecated OpenGL on macOS in favor of their own Metal graphics API.

 

When running Dota 2 with Vulkan, the performance was now similar across all three platforms. It's worth noting the macOS 10.15 performance was much better with Vulkan -- which Valve is achieving through the use of MoltenVK to map to the Metal drivers. Windows did still have a slight lead, but for all key purposes the performance was similar.

 

When running Xonotic that is OpenGL rendered and has similar quality builds across all three operating systems, Windows 10 was delivering much better performance with the Radeon Pro 450 than under Ubuntu 19.10 with its default drivers and macOS 10.15 with their lousy OpenGL driver coverage.

 

Switching from gaming to the ParaView scientific visualization software that makes use of OpenGL but is also CPU heavy with MPI, Ubuntu 19.10 was tending to be the fastest from this MacBook Pro followed by macOS and then Windows. Windows did lead in the wavelet sub-test of ParaView.

 

And for Java 2D rendering performance, with all three operating systems using OpenJDK 11, Ubuntu 19.10 was much faster with the exception of macOS 10.15 having better results for text rendering.

 

When it came to running a variety of browser benchmarks on the MacBook Pro, macOS 10.15 did tend to deliver noticeable performance advantages over the same Firefox 69 build on Windows 10 and Ubuntu 19.10.

 

When looking at the Google Chrome 77 performance across the three operating systems, macOS 10.15 still led in some of the tests but Windows 10 did pick up a few first place finishes. Ubuntu did pick up a victory in the graphics-heavy MotionMark benchmark.

 

When running NAMD across the three operating systems, Ubuntu 19.10 did deliver the best performance.

 

Ubuntu 19.10 also delivered the best Golang performance aside from the JSON parsing test where macOS Catalina was the victor.

 

Running some Java language micro-benchmarks also allowed Ubuntu to rack up some more wins.

 

In the Java-written DaCapo and Renaissance benchmarks, Ubuntu 19.10 was generally the front-runner here as well with all these tests on OpenJDK 11.

 

When running Intel's OSPray ray-tracing tests across the three operating systems, the performance was similar.

 

Intel Open Image Denoise saw slight leads with Ubuntu 19.10 and Windows 10.

 

OIDn and OSPray are among the components to be making up part of the oneAPI graphics rendering toolkit and thus will become more important with time.

 

C-Ray ray-tracing was the fastest on Ubuntu 19.10.

 

XZ and 7-Zip compression was the fastest on Windows 10.

 

Surprisingly, Windows 10 had the fastest Zstd compression results.

 

Python scripts ran the fastest on Ubuntu 19.10.

 

Ubuntu 19.10 also faired the best with the proprietary NeatBench video editor test.

 

With the Appleseed production renderer, macOS 10.15 was able to yield the best performance in this creative workload.

 

PHP performance was the fastest on Ubuntu 19.10 for those doing any server-side web development from your laptop.

 

Git also ran the fastest on Ubuntu 19.10.

 

Of 86 benchmarks carried out across all three operating systems, Ubuntu 19.10 was the fastest 48% of the time followed by Windows 10 with leads 31% of the time and macOS 10.15 at just under 20%.

Of the three operating systems, macOS 10.15 was in last place about half the time.

 

If taking the geometric mean of all the benchmark results, Windows 10 had an 18% advantage over macOS 10.15 Catalina. Ubuntu 19.10 meanwhile had a 29.5% advantage over Apple macOS and 9% over Windows 10 for this tests from the same MacBook Pro.

 

Where macOS tended to perform the best was with the Firefox web browser benchmarks, the results on Google Chrome were mixed under Windows and macOS, and macOS also did well in some of the creative workloads like the Appleseed renderer. Windows 10 picked up some wins in gaming tests and other select workloads while Ubuntu 19.10 showed its best under the heavy CPU/system benchmarks.

 

Source

 

 

 

Link to comment
Share on other sites


  • Views 772
  • Created
  • Last Reply

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...