More on FSX-SP2(DX10) delay factors

I will try one more time to explain why SP2 is taking longer, since there are persistent doubts as to Aces intentions in general and my truthfulness in particular.

While the features may have been final, they have not been tested. What do you mean, they haven't been tested - you said there are no new fixes right? Yes, but read on.

Since SP2 as a web download is a separate product we have treated it as such. SP2 is in a separate source tree. That is really the only way to be sure about what you are developing and building. Trying to build 2 products out of 1 tree is inviting disaster.

Because it is in a separate source tree, it requires a separate build process. Any separate build requires separate testing by definition. That is how the "form" of SP2, in that it is a separate product in a separate tree with a separate build process, dictates future actions. Because it is a separate product and treated as such, the bits are most decidely not "just sitting on the shelf". The bits need to be rebuilt and retested.

But that is not sufficient to ship them, that is only part of the story. On to the rest of what has to happen before we can ship.

Once all Acceleration languages were finalized ( and it took a couple weeks to do that after Acceleration RTM for all additional languages ), we could then turn our focus onto SP2. The single setup developer we have was busy with the extra language setups for Acceleration and could only move to SP2 once all the languages for Acceleration had shipped.

One he had done that, he then had to then write the setup for SP2. It did not exist before he finished with Acceleration. Without a setup, we have nothing to ship.

That still is not sufficient.

We stillĀ have to validate the contents of SP2. To do this, we then had to go back thru the bug database and checkin mails and validate all fixes that should be in both trees were. And we found a couple that, in the heat of trying to finalize Acceleration, were indeed not checked in both trees.

These then had to be integrated. Once that process was complete, we then could kick off "all-in" SP2 builds.

And start the localization process for SP2. Since we have a shared Loc team in geographically dispersed areas ( Dublin, Tokyo, Taiwan, Korea ) we direct that process but do not own it and have our hands tied as to how fast things progress.

Once the SP2 builds got to a certain stability, we could then post an SP2 beta to validate:
1) setup,
2) the required fixes are indeed in, and
3) that free flight multiplayer ( modulo shared cockpit ) between SP2 and Acceleration is indeed compatible.

These are all important, required engineering steps. This testing is critical testing and cannot be skipped.

So reaching SP2 beta is an important milestone and one which justified my blogging about it to inform the community as to progress.

From this much more detailed discussion, perhaps skeptics can be convinced. Even if not, let me remind folks that back-seat engineering is just as useful as back-seat driving.