But keep in mind that other developers have offset this limitation and can offer very effective glasshields effects in FSX.
Other developers...I believe you might know we could even easily switch dynamically between two different versions of the object, depending on the sim used, and as I've said, WE did custom reflection maps years ago, because there wasn't such thing as dynamic reflections back then, they were added with P3D 3.
So yes, we *could*, in theory, offer a better solution for FSX. But why would we ? FSX is already crushing under its own weight so, why should we spend time=effort=money, just to keep it alive for a short while more ?
The same time=effort=money would be much better spent doing more and more P3D-only stuff, gradually, the more and more the platform gets more widespread use.
See, with P3D V3, there's wasn't much incentive to work natively on it, to support all the *amazing* thing it can do in with its own SDK (not backward compatible with FSX), because they might not have fit in memory just the same.
But V4 changes everything, because we know it could handle so much more, that we won't risk killing the system trying new things.
Now i should go directly to P3D V4. I understand that as well but it's taking a lot of money to do so. (remember that PMDG requires 2 separate licences for its aircrafts).
I can only say we are doing our part.
Take something like KLAS. Released in 2009, when FS9 was still around. Former FS9 users that bought KLAS for FS9, were able to bring it over FSX, FSX:SE, P3D 1, 2 and 3, and now even P3D4, and we never charge A PENNY for those 8 years of upgrades...
So in clear, if you do release Louisville only for P3D, i will not follow. And it will be sad cause i enjoy your products.
No, KSDF is almost done, and it will be released for FSX.
But we are seriously considering making the new O'Hare (total remake) P3D4-only, because there are new things that has been added in the V4 SDK, that are doing all the difference in the world in relationship with balancing visual quality and performances, and would require an entirely different approach to modeling, that if we wanted to do an FSX version, it will be like doing an entirely new product.