but the device itself is useless = http://www.drycreekphoto.com/Learn/Calibration/MonitorCalibrationHardware.html
I would take that report with a grain of salt. iMO at the very least, it is misleading. But perhaps in a comparative fashion, not taking the numbers and the statements made as absolutes, the report can have some value.
One thing that I think accurately comes out of that report is how some instruments can be highly variable between copies. For instance, this can explain the high variability of reports on success with the Datacolor solution. And I do not think the software is responsible for this. Also I do understand that the Discus is a very accurate solution, probably leading the rest. You do get what you pay for. This observation also is made by the report.
According to the NEC Spectraview II monitor calibration software that I use, the Colormunki Photo does a very respectable job, just as it does for printer profiles. I did not get this observation from that report. I do understand that colorimeters calibrated to use a specific display can perform better than a more general purpose spectrophotometers. This is mentioned in that report too. But notice I am not basing these observations from the report on the numbers. And I avoid the flat statement they sometimes make, like the i1Pro not being able to provide a critical result with its printer profile. IMO with the right software, it can, at least for my purposes.
FWIW
Bob
Bob Graham