Mine is printing right how, with the owl. no support. we shall see... its a KS Printrbot+ with many upgrades. im printing 150µm layers on a slowish 70mm/s and 50mm/s 25% rectilinear infill. I disagree about using manufacturers defaults. in many cases there are no manufacturer defaults. i suppose whatever was default with Slic3r but then what are the defaults? maybe they are completely useless. rather the same print parameters should have been used across all the printers in order to compare the actual quality of the hardware and control software - the printer. using different parameters would wreck the test. perhaps a default set of parameters and then tweek them to get the best results. post both results. that would give a better sense of the quality of the actual machine.

did the replicator 1 pass the test?

Arrrrrhhhh, I have knocked over the little par of the arc 2x now.  this time I was about 10mm hogh on it.  50% infill, no support material, having the z lift .2mm.  I have lots of jitter and I think thats what is knocking it loose. The jitter happens when it fills the hollow part of the box walls. 

Is it meant to be printed with or without support material?

Link to Cliff's Thingiverse page, since he made this object:  http://www.thingiverse.com/cbi...

It's not an impossible print, but it's really hard.   Even the one in that photo is missing the thin section.    Getting the arch and overhang right is a good test, but getting the holes / wall thicknesses right was the harder test. 

I tried printing hollow it was coming out really good until about 10 mm where because I was printing hollow the arch curled up and grabbed the print head.   That overhand under the arch is nasty too.  Cool print and definitely a good torture test.

Print solid hollow or with some infill?