Just for fun, I put on my science hat and checked the paper (I have access through work). The 2 devices they are using are the Sony PRS 600 and the Apple iPad (1st generation). I only looked at Figure 1 (because the rest of the paper contains tables and they bore me). Particularly for General Fatigue, but also slightly for visual fatigue, the LCD device was subjectively worse for all trials. However
, the differences were not statistically significant. HOWEVER
, they only had 10 participants (5 male, 5 female, all right-handed). I would be very interested to see the statistics on a larger population, for example 100 people. The measurements of reading speed look very similar for the 2 devices.
ETA: Of course I should have said that statistically significant does not necessarily mean biologically relevant. In other words, some differences don't matter much.