Determining how much processing power a computer offers is a lot more complicated than it used to be. Nowadays multiple benchmark tests are used to determine the power of your machines -- this video shows the differences between the three most common types of tests out there.
Because current PCs are comprised of so many different parts of varying power and quality, benchmark tests can help you tell the difference between two setups. In this video, Linus from the Techquickie YouTube channel gives a quick and simple explanation for each of the three types and what type of systems they're best suited for:
- Synthetic benchmarks: These tests are designed to have easily repeatable results for accurate comparisons and minimal bottlenecks across different isolated tests. This makes it easier to test individual parts -- like an SSD or processor -- without other factors affecting results, but it also may not reflect a user's actual experience with the machine since it doesn't test everything working together.
- Real world benchmarks: Your entire system is given a real workload -- such as file compression or 3D rendering -- and tested on how fast it can complete the task. This benchmarks your machine as a whole, but doesn't give you any results regarding individual components.
- Hybrid benchmarks: These benchmarks use a combination of isolated synthetic-style tests in combination with more generalised real world benchmarks.
So which one should you use? If a real world benchmark is available for your exact type of workload, that's probably your best bet. For example: a real world benchmark specifically designed for high-end gaming is only ideal for testing a gaming PC. Otherwise, look into synthetic benchmarks for specific aspects of your PC and consider hybrid benchmarks for overall performance.
Benchmarks as Fast As Possible [YouTube]