You cannot compare scores from different benchmarks. The aim of benchmarking is to run the same test on every system. This is the only way you can compare results fairly.
Each UL benchmark has unique workloads designed to test specific combinations of hardware and software. This means you cannot compare 3DMark and PCMark scores, for example, nor should you compare scores from Time Spy, Fire Strike or other 3DMark tests.
In our Advanced Edition and Professional Edition benchmarks, you can choose to run a test with custom settings. You should not compare results from custom runs with scores from default runs since the amount of work is no longer the same.
Some legacy benchmarks, like 3DMark 11, have presets such as Entry and Extreme for different levels of hardware performance. Scores from different presets cannot be compared directly.