New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
How accurate are ServerBear's UnixBench scores?
I'm concerned about the UnixBench scores that show up in sortable tables on ServerBear.
Is there any reason to believe that the scores posted in their database are going to at all representative of the performance I will see on any of these providers?
Because when you run UnixBench, you run it given whatever all the other users of the physical box happen to be doing at the time. I bought the EDIS Smart VRS because it had almost the highest UnixBench score on the list, but then I realized the above...
What if anything can give me some idea of real-world performance I will see on any given provider?
Comments
Of course not. If you get a VPS on some overloaded / oversold node you are not going to get any performance out of the box.
Providers can also run them on new nodes to boost their results though that's usually tempered by customers doing the benchmarks themselves. Not that there's something wrong with that because I think we all want to put up a good look, but gaming your results on empty nodes will just incur backlash. Just look at all the doctored Android benchmarks discovered recently.
They are using vserver, that has access to all server resources, in most cases.
If the script is still as it was before, you don't need to run any benches at all.
Just go post arbitrary values to their collector URL, it has no way to verify your claims.