...have always been one of the most controversial topics in the tech world.

 

No matter which benchmark you publish, someone will almost certainly have something negative—or outright obnoxious—to say about it.

If you don't believe that, write an article, include a few benchmarks, get it published, and then sit back and enjoy the flood of argumentative

emails and hostile comments.

 

I think it comes down to two main factors. First, people become emotionally attached to their favorite operating systems, applications, hardware,

databases, and other technologies. When their preferred platform ends up on the losing side of a benchmark, it can feel surprisingly personal.

For some, benchmark results are interpreted as an attack on their own choices rather than simply a set of measurements.

 

Second, benchmarks are, by design, limited in scope. They cannot possibly capture every aspect of an operating system, application, or hardware

platform. Every benchmark measures a specific workload under specific conditions—nothing more, nothing less.

 

Because of that, it's often easy for someone with a reasonable understanding of the technology to point out limitations or identify scenarios where

different results might occur. That doesn't necessarily make the benchmark wrong. It simply highlights that no benchmark can answer every possible

question. Benchmarks are tools for exploration. Some are broad and comprehensive, while others focus on a single aspect of performance. Their purpose

is to answer the questions they were designed to ask—and, ideally, to inspire new questions and further investigation.

 

If I publish a review and you believe there's something worth improving or testing differently, I'd genuinely like to hear from you. Constructive feedback is

always welcome, and I'm happy to discuss methodologies or consider additional benchmark suggestions.

On the other hand, if your only contribution is to leave rude or dismissive comments, I'd encourage you to conduct your own testing, document your

methodology, publish your findings, and let the community evaluate your work. No one is stopping you, and creating meaningful benchmarks is botheducational

and rewarding. So, as you read and use my reviews—or anyone else's—keep in mind what benchmarks are meant to do, and just as importantly, what they are not.

 

Finally, many thanks to Tony for helping shape these thoughts. You know exactly what I mean, my friend.