Compare systems based on baseline *averages*

[David.P]

Hi forum & Passmark team,

I'd like to make baseline comparisons using the PerformanceTest software. For example I'd like to compare three certain notebook models.

However, the problem is that (although I can find many baselines for those notebooks in the online baseline collection) I only can compare single specimens of baselines to each other, which usually however do not represent the typical (i.e. average) performance of the respective system, e.g. noteboook. Instead, I even might pick baselines that could be off that average by a wide margin for some reason, and therefore not get meaningful comparison results.

Therefore I'd like to ask if there is already a way to compare certain systems baseline averages to each other, instead of having to pick only single baselines for every PC system, in a comparison?

Thanks for any insight
David.P

[David (PassMark)]

Have a look at the components instead of the whole machine.

For example you can find the Laptop CPU chart here,
https://www.cpubenchmark.net/laptop.html

[David.P]

Thank you David, though this is only the component benchmark then.

I'd however rather compare the entire system using all the informative tabs and beautiful bargraphs inside PerformanceTest -- and it would be just SO amazing if this were possible using averages of, say, 5-10 systems of the same type and build!

Thank you for possibly considering such a feature,

Regards
David.P

[David (PassMark)]

There are 1000s of machines on the market at any one time and they turn over fairly quick, and come in multiple configurations. So getting multiple samples of the rarer units isn't always feasible.

[David.P]

I know, but I still could choose like 3-5 samples of the same units manually and then choose "build average results from selection".

This way, I would be able to get much more reliable results for a system benchmark comparison. Currently, for a comparison I can only pick one unit of each type -- which however usually has a couple of runaway values here and there, such that you never know if the comparison is distorted or not, and if it is, where.

Can you see what I mean?