Ok here we go. Sorry for the long delay, various Issues. First, Chris let me know that what I had was not XD and happily sent along a sample of the XD. From a customer service and value stand point, I really do like Seymour AV and Chris. That alone is sure worth a lot. None the less, I'm sorry to say that, acoustically, there was no real difference between the materials.
Below is a graph of the EN2, standard Seymour AV material, XD material, and a reference response of the speaker tested. The response is cut off at 1.5 khz as variations below that point aren't being caused by the screen (and largely didn't exist). This is a very zoomed in completely unsmoothed measurement.
Below is a graph showing 1/16th smoothing with a more normal scale. This is more inline with the type of measurement a magazine would publish, not magnified to show differences.
What you should notice from these graphs is that the EN2 is better, but that all three materials are considerably worse than nothing at all. At the same time, looking at things on a more normal scale, or even applying 1/3 smoothing, which is arguable more inline with how we would perceive things, it will look a lot better, and isn't make for all that great a difference. None the less, all of them are causing comb filtering, and I think it's important to realize this. From a value stand point, the Seymour AV materials have my vote, as the measurements aren't bad, and the image looks great, the response isn't all that bad, and the service is quite impressive. If perfection is your gain, don't buy any of these.
Ok so you are probably wondering where those percentages are and measurements of the 4K material. Well, it takes a lot of time to do this right, and I don't have a lot of it. Turns out that while my little excel program is great in concept, differences in the raw data can drastically throw things off. The measurement program I use doesn't use discrete frequency test tones, and as such, the density of the data can vary depending on whats happening. From a graph stand point, thats fine, if trying to mathematically use the raw data, it does create a problem, as I need to search for data points in two different files which are "near" each other (~), discard repetitive data, etc. If I get it working right or find a way to restructure the files raw data to get a percentage calculation, I will give you guys a sort of distortion figure for coherence with these screens.
As for the 4K, the sample is too small to test in the same way as the others. The other way of testing is to put the material directly infront of the mic, instead of the speaker, but then it's not comparable to the other measurements. Doing so gives a measurement nearly identical to the reference, so I really do think that this is a truely transparent material with no comb filtering.