A little more information on the eyepiece transmission test report from Markus Ludes I and others have posted:
Lack of description of methodology was brought up by BillP in the Abbe vs Plossl thread so I went to the source and received the following info. The testing was performed by Konrad Horn Optik Messungen (Konrad Horn Optical Measurements), a German optics testing institute that specializes in UV-VIS-NIR spectroscopy in the range 185-3200 nm and provides measuring services for a variety of optics including optical fiber, prisms, lenses, etc. The equipment used is regularly tested and certified. I don't think we need to worry about the accuracy of the figures shown in the above link.
Actually you do. Before buying into this test we should all ask how it was performed which can have a drastic effect on the results. And it they just tested one eyepiece of each type, than at best, no generalizations can be made to that type of eyepiece. In other words, if the tested just one 19mm Panoptic, it is not appropriate to generalize that transmission level to all 19mm Panoptics. The sample size is obviously too small. Testing 101.
It is what it is until someone forks over the money to fund measuring multiple samples of each. Any volunteers?
Since there's no evidence that there is visually detectable (and since these are eyepieces that's all that matters) unit to unit variation in eyepieces from Pentax, TeleVue, Zeiss and others in that test, I'm not put off by a sample size of one. And it's not like this is some difficult task, these are routine measurements like length or width for a company that specializes in measuring such things with calibrated equipment made specifically for that task, no need to treat it like it's a scientific experiment. Common Sense 101.
If you can show evidence that there is visible unit to unit variation in eyepieces from these companies then sample size would be more important in drawing conclusions.