Reading the length of this thread, there is much to consider. I too use H-a filters with night vision (NV) and have produced images with both my Astronomik 12nm and Optolong 7nm filters. The visual image at the NV ocular is less impressive than my phone photos (Phonetography) in terms of resolution and visible detail. The dynamics of NV require a narrower band for visual use, so the Orion Extra Narrow Band filter previously mentioned, may be working at a narrower bandwidth than 5nm. And, the degree of band shift experienced with only the center 15-20% of FoV revealing the H-a subject, is a strong indicator that the filter is out of phase on the f:1.2 Envis lens. Although I see band shift on my Envis lens with all of my H-a filters, including an Astrodon 5nm, I do not experience band shift when they are used with a slower prime lens, either in front of or behind the objective.
To keep this post on topic though, I have two 7nm Optolong filters, both purchased during 2017, one 1.25" and one 2", from different retail vendors. I felt compelled after reading this thread to test each of them along with a 12nm Astronomik filter. So here (below) are results of my tests, which seem to reveal no color bleed-over. My Optolong filters do look different from one another though. The 1.25" has a very distinct gold tone reflective coating on the incoming side and a silver/blue tone reflective coating on the outgoing side. The 2" Optolong has a bright silver reflective coating on both sides. I have used them both many times and do not notice any distinguishing characteristics between them in use.
Here is a link to my gallery of images taken with all three filters; most images taken with an iPhone 6+ but a few of the recent ones with a newer XR; these are single images of no more than 1/2s, averaged for 5s-30s in NightCap Camera App on the phone:
The below images were taken with my iPhone using a RGB scale on a MacBook Pro (old 2010 mdl.)
Optolong 1.25" 7nm (gold tone reflective coating)
Optolong 2" 7nm (silver tone reflective coating)
I cannot address some of the technical issues that have been discussed in this thread, but relating to NV, I can say that I have tested my filters against one another for visual use. I find that a narrower band filter helps to define the H-a subject with greater contrast between H-a and sky background, but at the cost of attenuating background stars. And, with a very fast optical system like the f:1.2 Envis prime lens, band shift does interfere with a full FoV transmittance. At 12nm, there is +/- 5% loss at the perimeter; at 7nm it is +/- 10-15%; and at 5nm, it is 20-30% loss... meaning that the H-a subject only appears in the center 60-80% FoV. When the optical system is a bit slower as in most telescopes, band shift is not an issue when using NV.
In extensive visual testing using NV between my 7nm Optolong and 5nm Astrodon filters, another CNer on the East Coast and I on the West Coast, using different telescopes, found subtle differences between these two filters, sometimes preferring one over the other for specific targets. If interested, our test results can be found in this article:
NV does change the parameter of this thread, but I use narrow band H-a filters to achieve realtime visual results and the subjective interpretation of those results have some merit, even if from this unique perspective. I have no personal interest of any kind in ANY company making filters. Actually I like all of my H-a filters for different reasons. But I will say that I do not use the Astrodon 5nm for images; I use the 12nm most of the time as it permits more light with a wider pass band which does not stress the NV or my phone camera. For visual use with NV, I use the 7nm Optolong a little more often than the 5nm Astrodon; a personal choice based on the limitations of each filter.