PC System Benchmarks – Software Testing Weekly #2

Benchmark Tests / / 10 minutes of reading

Introducing PC System Benchmarks

PC system benchmarks are a good method to identify potential bottlenecks in your computing performance and choose effective system upgrades. As there are several types of different PC system benchmarks, our focus has always been on those that provide scripted test scenarios inspired from real world use cases.

In March of 2015 we set off to test what various software products were doing to our computers. In this study, 133 products were tested. The inclusion process was simple and straightforward: few software download sites were manually scanned for any products claiming to make your computer faster by cleaning it or by fixing system errors.

In October of 2016 we tested the top ten registry cleaners. The products tested were chosen simply with a Google search of Windows registry cleaners and the top ten results were picked.

In January of 2017 PC system benchmarks were performed to determine the best registry cleaner from a selection of the top three which previously ranked best.

This week, as we have just released jv16 PowerTools 2017 version 4.1.0.1728, our interest for PC system benchmarks was reborn.

Why this test now?

Today is supposed that PCMark 10 will be released for public use.

As it has been quite a while since PCMark 8 has been the standard in PC system benchmarking, we wish to celebrate the upcoming public release of PCMark 10 this week with a Windows 10 benchmarking test report done using PCMark 8, before and after using popular PC system utilities.

Since PCMark 8 is not necessarily fully compatible with Windows 10 benchmarking methods, we expect surprising results and some enticing conclusions that may lead us to further test cases for next week’s lab report.

Once PCMark 10 will be released, the conclusions of this micro test will be used to compose the test cases for a much wider study involving more software products that promise to clean and speed-up your computer.

If at the time of reading this report PCMark 10 has been already released, then you can freely benchmark your computer using this software for free in the same way that we did in our lab with PCMark 8.

To download PCMark 10, visit: https://www.futuremark.com/benchmarks/pcmark10/. We would love to know your results, so don’t hesitate to join the discussion about this topic on the forum.

Tested Products

To start, we looked at the top competitor (according to Google.com search results for “registry cleaner”) and decided to see the PC system benchmarks score impact (if any) before and after using the said product: CCleaner.

In order to learn the most from this experience, we downloaded the free, professional trial of CCleaner in order to evaluate this software from the best angle possible.

As far as we know, this version is supposed to be fully featured, including an active background system monitoring service – which may or may not have an impact on the overall PC system benchmarks score given by PCMark 8 – to be determined.

Test Cases involving PC System Benchmarks

The objective of this test was set to record the virtual PC system benchmarks score given by PCMark 8 on Windows 10 with some of the most common 3rd party applications installed:

  • Before installing and using any of the tested PC system utilities
  • After installing and using CCleaner – Professional
  • After installing and using jv16 PowerTools 2017

PC System Benchmarks Test Environment

The current benchmark test was performed using an Oracle VirtualBox (version: 5.1.22 r115126) virtual environment running Windows 10 with 4 CPU cores and 8 GB of RAM assigned.

In order to ensure that no system updates or other updates are downloaded during the testing procedure, the system was disconnected from the Internet before starting each PC system benchmarks test set.

The virtual machine was run on a host computer running Ubuntu Linux 17.04 LTS (64 bit) with an Intel-i5 processor, 20 GB of DDR3 virtual memory and a SSD system disk.

Such an ordinary test environment was created and used in order to simulate the easiest conditions that can be reproduced by any consumer to verify the validity of these tests.

Using high-end hardware or complex distributed system may result in more professional test methods but may also return test results that are only applicable in an uncommon environment (test lab which does not reflect a real world use case).

For the future, we will also approach this from the ISO point of view. For now, trying to keep it simple while educating consumers about their options is our primary objective.

PC System Benchmarks: Verifiability

All of the past and current test results are accompanied by screenshots and video recordings which have been saved during various testing phases in order to address some of the most common test result credibility concerns for this type of studies.

Test Case #1: Summary

In the first test case we measured the PCMark 8 PC system benchmarks score for a virtual Windows 10 system that was freshly installed and updated for this purpose.

In this environment, we installed the most common software used today, according to ninite.com. This wonderful tool allows you to quickly install or update multiple commonly used software at the same time, either on a local end-point or even across the network.

Once all the software has been installed, the virtual machine state was saved and we used this state of the operating system as a benchmark point of comparison (without any PC system utilities installed or used on the machine before measuring its baseline PC Mark 8 score).

In order to measure baseline performance, we run the free, conventional test offered by PCMark 8. This test is advertised as being the standard for common home computing tests.

The first of a total of six tests that PCMark 8 does is a scripted scenario in which the PC system benchmark score is measured while simulated social networks are automatically browsed by the benchmarking software.

During this automated test, no user interaction is needed as it is not recommended to use the system which is being benchmarked at the same time that tests are running.

In the second test system performance is measured while browsing for items to buy from an online retailer.

The third test addresses the common use case scenario of editing content using a word processor.

In the following test PCMark 8 measures overall system performance during a simulated video conferencing session.

Finally PCMark 8 helps to determine the PC system benchmark score while the computer is being used for photo editing.

PC System Benchmark Results: Test Case #1

On Windows 10, without cleaning or optimizing the system with any PC system utilities, PCMark 8 has calculated a PC system benchmark score of 1060.

Here are the individual score values allocated to each of the tests:

  • Writing 5.51557 s
  • Video Chat encoding v2 Conventional 237.00000 ms
  • Web Browsing – JunglePin 0.44450 s
  • Casual Gaming 0.70 fps
  • Advanced Photo Editing1 Conventional 0.98143 s
  • Web Browsing – Amazonia 0.16620 s
  • Video Chat playback 1 v2 Conventional 20.01 fps

The total PCMark 8 Home Conventional Score for this test case was 1060.

Test Case #2: Summary

In the second test case, the previously saved Windows 10 snapshot (with most common 3rd party software installed) has been cleaned and optimized using this week’s tested PC system utilities.

PC System Benchmark Results: Test Case #2.1 (CCleaner – Professional Edition v.5.31.6105)

After cleaning the Windows 10 test system with CCleaner – Professional Edition (Registry Cleaner) a total of 53 registry problems were fixed.

After cleaning the Windows 10 test system with CCleaner – Professional Edition (File Cleaner) a total of 400 megabytes of disk space were cleaned.

On Windows 10, after cleaning and optimizing the operating system using CCleaner – Professional Edition, PCMark 8 has calculated a PC system benchmark score of 1016.

Here are the individual score values allocated to each of the tests:

  • Writing 5.72957 s
  • Video Chat encoding v2 Conventional 284.00000 ms
  • Web Browsing – JunglePin 0.45689 s
  • Casual Gaming 0.69 fps
  • Advanced Photo Editing1 Conventional 1.02341 s
  • Web Browsing – Amazonia 0.17674 s
  • Video Chat playback 1 v2 Conventional 18.03 fps

The total PCMark 8 Home Conventional Score for this test case was 1016.

PC System Benchmark Results: Test Case #2.2 (jv16 PowerTools 2017 v.4.1.0.1728)

After cleaning the Windows 10 test system with jv16 PowerTools 2017 a total of 83 file and registry related problems were fixed. More detailed statistics are available when using the software’s dedicated tools. For the purpose of this test, we used the Clean and SpeedUp My Computer tool.

This tool was designed to automatically use multiple other tools packed into jv16 PowerTools (such as the Startup Optimizer and Registry Compactor) in order to both clean and optimize a system’s performance.

On Windows 10, after cleaning and optimizing the operating system using jv16 PowerTools, PCMark 8 has calculated a PC system benchmark score of 1030.

Here are the individual score values allocated to each of the tests:

  • Writing 5.57387 s
  • Video Chat encoding v2 Conventional 277.33333 ms
  • Web Browsing – JunglePin 0.46557 s
  • Casual Gaming 0.70 fps
  • Advanced Photo Editing1 Conventional 0.97845 s
  • Web Browsing – Amazonia 0.17018 s
  • Video Chat playback 1 v2 Conventional 17.01 fps

The total PCMark 8 Home Conventional Score for this test case was 1030.

PC System Benchmarks (micro testing): conclusions lead to wider tests

After cleaning the Windows 10 operating system with either CCleaner – Professional Edition or jv16 PowerTools 2017, the PCMark 8 PC system benchmark score was lower than before using any PC system utilities to clean the operating system in the first place.

This result is not surprising for us at all, considering that both of the tested registry cleaners attach a resident process to the system in order to monitor the operating system’s health and overall performance.

How can the operating system get a lower PCMark 8 benchmark score after it has been cleaned?

Most likely, the reason for the lower score obtained by the operating system after being cleaned with each of the tested products in comparison to the baseline test score is because of an extra running background process in both cases, however, this is just an assumption at this level.

In order to clarify this assumption, more extensive tests will be carried out in the near future, with and without background monitoring services enabled for either jv16 PowerTools and any other similar interest software product that we can easily find using a simple search on Google.com

With PCMark 10 pending release later today, we wish to do a similar test case scenario next week and hopefully compare the results between PCMark 8 and PCMark 10.

Results should vary as PCMark 10 has been specifically designed to become the industry standard for benchmarking Windows 10 operating systems, as PCMark 8 has been for a long time in regards to Windows 8 and Windows 8.1.

What do we expect to see in the next PC system benchmarks done on Windows 10 with PCMark 10?

First of all, we expect PCMark 10 to be far more compatible and capable to measure small differences of Windows 10 performance before and after certain actions have been performed in the test environment.

As PCMark 8 was not necessarily designed to manage Windows 10 after its latest updates, we expect this to improve with the latest release of PCMark later today.

For the start of the next week, our weekly software testing report will most likely include a preview of PCMark 10 and the scores obtained by a Windows 10 virtual environment before and after cleaning it with popular pc system utilities.

Related weekly benchmark test reports:

Other historical benchmark test reports:

Since February of 2009 (when the first test of this kind was done in our lab), Macecraft Software has published a total of seven registry cleaner product comparison studies: