Den Eindruck kann man gewinnen, wenn man sich die Erklärung von FS Dreamteam, 29Palms, Fly Tampa und Flightbeam durchliest. Dort erklären die Entwickler mit welchen Problemen sie zu kämpfen haben und welche Limitierungen sie einengen. Die komplette Erklärung findet ihr unter
Dear Flight Simmers,
For the past 11 years, FSX has been our simulator platform of choice. It served us well, and we saw amazing things being done with it by 3rd party developers, offering us very complex airplane simulations, very detailed sceneries, and clever utilities that extended the simulator’s capabilities far beyond what we thought to be possible years ago.
Lately we, developers, start feeling more and more restrained by the underlying platform limitations. While 4GB of maximum ram would seem adequate in 2006, when FSX was originally released, and no PC could handle the detail that consumed that amount of memory, in 2017 we have hardware far more capable, but we just cannot use it because of the software limitations.
The most aggravating problem is the 32bit memory limit, which no hardware can fix, since it’s a core software problem. This not only is the cause of OOMs, but it also forces developers to design a product taking into account this limitation. Some optimizations that might be good for fps, like LOD levels, have a side effect of taking more RAM, so we are often forced to choose between the lesser of two evils, either having a good frame rate and smoothness (risking OOMs), or losing fps and smoothness, trying to save up to the last byte of memory, to prevent crashes.
FSX was born with issues that, after the latest patch in 2007, nobody could have addressed, because the traditional cycle of a new sim every 2-3 years has been interrupted, so we had to live with it, and found ways to overcome it. We found very creative ways to bypass several underlying limitations of the sim, a common example of which is the complexity of creating convincing ground textures, by layering lots of different materials over the FSX round earth model, and other ‘hacks’ which wasted lots of our time in developing and offering support for, and that would have been surely better spent perfecting the art, instead of finding solutions to problems that shouldn’t exist. Airplane developers have been also very limited in their options, and many of them had to reverse-engineer the sim, to hack into it at the deepest level, for example to simply get a more realistic lighting of the cockpit.
But we got used to it, and made good products, which are loved by users.
Recently, the discontinuity between what the FSX engine can do, and what other game engines are capable of, has become a real burden, and it’s hampering our chances for future advancements.
In the past years, we have seen the rise of Prepar3D, which was initially based on the FSX graphic engine, but it has now evolved on its own path, while still maintaining a surprisingly high rate of backward compatibility with FSX.
Prepar3D has a lot more to offer to developers, and therefore to its users, than just backward compatibility. The native SDK is being constantly improved, and it has grown so much, that we are more and more frustrated by the FSX limitations.
But until today, Prepar3D was still a 32 bit application, which means it might have not been possible to look at the native SDK, because many of the things it has to offer, would likely not be possible due to memory limitations. Today, the release of Prepar3D 4 changes everything. Not only we might not be constrained by an SDK made in 2006, but the better features that Prepar3D always offered to us, become now really feasible, for the first time.
Take some of the examples below:
– No “hard” Memory limitations which might cause OOM errors.
– Dynamic Lighting
– Improved human Animations
– Creation of more advanced SimObjects
– Conditional display of objects/textures
– More realistic and reliable ground models
– Improved compatibility between different add-ons (and autogen), due to the new installation method.
Today, if you compare FSX with Prepar3D V4, you have the new sim which offers a solution to a problem (memory) that can only get worse as add-ons get more complex and the hardware more powerful to support them and you also get a generally smoother flying experience and better fps too. And if you buy more hardware, you will get your money’s worth of it, because buying a faster video card or adding more memory will actually be reflected on your sim’s performance.
We believe the platform that will stay for us as the main sim for the next years, is Prepar3D, so we decided to make public a decision that we have been considering for several months already, since the first time we got to see the new sim:
In the next months, we’ll all stop developing new products for FSX.
Some of our products that are already very close to release, such as Flightbeam KMSP or Fsdreameam KSDF, will still be offered for FSX too, and we’ll of course continue to support all our existing releases with bug fixes, but the new products we’ll work from now on, will be designed for Prepar3D 4 and future versions.
We all agree this is the best decision for the future of the community moving forward, and we can only hope that our stance would result in other developers considering the same move. But we also need you, the users, to make this work.
Umberto Colapicchioni – Fsdreamteam
Alessandro Cucinotta – Fsdreamteam
Amir Salehi – Flightbeam
Lars Pinkenburg – 29Palms
George Grimshaw – FlyTampa
Martin Brunken – FlyTampa
Emilios Gemenetzidis – FlyTampa