More a curiosity than a question, is there a sort of benchmark of 3D printers? I mean, I know that you can perform torture test or objects that can prove the quality of a printer, but for example to compare two printers on speed i think it is not sufficient to look simply at max linear speed of carriage, but as overall speed in a specific task. For example with specific parameters of quality and so on which printer first complete the object respect to other or, considering the same object how small can a printer go with layer height. I’m asking because now The BadPrinter2 is complete and I’m very happy with the quality and speed, but I’m interested in knowing which can be the area of improvement respect more consolidated printers like ultimaker, replicator2 and so on… but also which can be the good point of my project again compared with others to understand where I can freeze my project and where I have to work. What do you think?
If you haven’t already, see last year’s Make: Ultimate Guide - http://makezine.com/volume/make-ultimate-guide-to-3d-printing/
It’s very hard to directly compare printers because there are so many external variables (filament quality, slicer parameters and version , ect), and each printer needs calibration and experience to produce the best results. This year’s Make 3D printer issue will focus solely on pre-assembled printers because they simply do not have the resources to build,calibrate and test kits.
Thank you for the answers and the link, it’s very interesting. I imagine such thing, this why I say curiosity at first to know if someone already studied a sort of fixed parameters to directly compare the machines like perfomance for cars 
but in fact there are a lot of variables that can’t be directly compared. I will wait for this year issue of Make’s guide, hope to be in there one day with ours BadPrinter2 
I belive that as long as the mechanics are accurate (so no lash or wobble and a decent hotend), every printer can achieve a superb print quality with the right firmware and slicer settings.
A real benchmark would be quite hard to setup since there are so many properties that describe a print’s quality, ranging from layer adhesion to retract strings.