Benchmark changes?

[margrave]

It seems I'm not the first to notice the radical changes in the ratings. Doesn't this reduce the usefulness of these ratings?
This needs to be fixed.

If an ounce today is different from what it was yesterday, all weight measures become meaningless.
If an mile today is different from what it was yesterday, travel would be difficult!
I such things had been changed arbitrarily a half century ago ... Neil Armstrong would have landed in the Pacific instead of on the moon.
So why screw up the CPU ratings system? Somebody has really goofed up!

[Tim (PassMark)]

The CPU tests in PerformanceTest hadn't had any major changes in years, CPUs have changed a lot in that time so the tests were becoming less relevant, the tests needed to be updated in order to be useful on new CPUs. Changes were made to the 2D and 3D test for the same reason.

The overall mark values have been scaled down to be similar to PT9 but it isn't possible to get them to match exactly (older CPUs will generally be worse off now and newer CPUs will perform better).

See this thread for a more in depth explanation.

[David (PassMark)]

If an mile today is different from what it was yesterday, travel would be difficult!

Your comment implies 1) there is a standard unit of measurement for computing power and 2) there is only one characteristic worth measuring.
But neither is true.

A better analogy would be measuring the performance of all cars based on 'litres of 98 octane fuel used per kilometre', and finding out that there was an new electric car that didn't use 98 octane. Then instead of updating the performance measurement to use current technology, you continue to use the old L/km measurement, even though it makes no sense anymore.