Here is a copy of my yet results from tonight, in relation to retention of circular polarisation for a passive 3D system, with select screen specifically designed for polarized light.
Well - started the tests a bit late, on account of da Superbowl
Good and bad results tonight. But the main summary points are: (1) circular is definitely not the way to go (2) the Solar HD 1.3 I just got from SI that was supposed to be circular compatible retains almost zero polarisation.
Also, these results are not complete, as it seems I left the Harkness sample in the office - and I neglected to run the "baseline" measurement with just the pure filters, without screen interplay, to see what the theoretical maximum retention COULD be.
I will edit this post tomorrow night with the additional numbers.
That said, here is what I found tonight. Remember, these are not precise, as they are based on eyeballing the test patterns.
In order of performance on retaining circular polarisation:
Stewart 3D - 80 to 81 %
Stewart 5D - 77%
Dalite Silver Matte - 77%
Dalite 3D Virtual Grey - 70%
SI Solar HD 1.3 - almost 0%
Overall, I am very disappointed in the performance seen in these tests. Even the very best screen in the tests barely retained more than 80% of the circular polarized light, meaning that there will always be some situations where ghosting may be an issue, no matter what you throw at it.
This would explain why RealD has to go to such technological lengths to pre-process movie material, estimate the ghosting, then cancel an appropriate amount of light from the other image. They certainly make it work well in the theaters, but that is not a realistic approach for a home theater.
The Harkness screen may still outperform the Stewart when I test it tomorrow, but it is still likely to not be able to retain a perfect score with circular.
So, while I had originally thought that circular would be the perfect solution, as it allows me to move my head more, my testing has shown me the error of my pre-conceptions.
So it is looking extremely likely that Linear polarized is the way to go.
Btw - until you have seen 3D at 48 frames per second, using frame interpolation, and flashed at 96 hz - you just cannot understand what you are missing.
I wish that I could have shown James Cameron how amazing Avatar looked on this system tonight. It was never so clear and easy to follow the movement in the theater.
The scenes where they are flying through the valleys and floating mountains were absolutely spectacular. And the action scenes just didn't break up at all, like the way they did in the theater.
And How to Train Your Dragon was amazing also. Particularly the fight scenes, such as at the beginning, where they introduce Astrid, and the huge dramatic battle with the Uber-monster at the end. The intro to the battle in the ring, where the camera is flying, circling the ring with all the people just highlights how crystal clear and smooth this system is with the FI. I promise you you've never seen anything like it.
Anyway. More results tomorrow. But I think we are now drawing close to the end of all the investigation for the ultimate system. And it looks more and more likely that Linear will win this by a longshot.
I'm hoping that the end result will be a screen that allows good 2D qualities, not just 3D. I'm holding hopes out for my trusty BD II to come through in the linear tests next weekend.