@ Kju you’re welcome BUT remember this is what the developer wrote
"> CPU Benchmark uses pystone a single core python module to benchmark native python implementation.
Brut python performance is calculated using a standard metric “pystone” . Users can formulate real word expectations comparing Kodi plugin performance.
The pystone score does not reflect hardware capabilities, nor does it benchmark them. pystone does not take into account multi-core processing, which can greatly impact the performance of plugins that supports it … Enjoy!!."
the single core python scores will have a big play in addons but some addons might utilise multicore more etc.
I still think the scores do really well to show relative performance for addons.
the developer has an I/O Benchmark addon
https://kodi.wiki/view/Add-on:I/O_Benchmark
I’ve ran it and got results that are all over the place, so I’m asking the developer to have a look.
once it’s ready and we combine these REA/WRITE scrores with the single core python scores, we’ll get a slightly better picture,
so boxes with the same SOC, might give you different READ speeds. these read speeds have an impact on how fast artwork will be displayed, I also want random read speeds if the author can manage.
so user can run this and report back so I can add to the collumn. will be fun 
if you have any results you want to add to the community table, I’d be happy to take them
So the “somewhat faster” from the first post is a bit misleading. When I replaced my N2+ with a Homatics R 4K Plus I felt a big performance degradation. But I thought, it was due to the EMMC vs USB. But now I think it’s more likely that it’s the CPU itself.
I mentioned this countless times to ht2tweak that saying something like the N100 is same speed as AM6B+ is not fair representation. big reason why I looked around for an addon to give us better objective measurement.
the SK1 is a huge surprise though, it’s 90% faster single core python speed. THX