Wavefront OBJ

One of the new features I have started working on since the release of ModelConverterX 1.0 is adding support for the Wavefront OBJ format. One of the reasons for this is that we use it at work for some files and I would like to try the LOD Creator functionality on those. But I think it could also be an alternative to the 3DS format for some people who try to import their old work into other tools. The 3DS format has some annoying limitations, like texture names having the 8.3 DOS naming convention. At the moment it seems to me that the OBJ format has less of such limitations.

In the current development build of ModelConverterX the OBJ reader is already implemented and until now it seems to work well on all test objects I have found until now. At the moment I am programming the OBJ writer, so this should be available for testing soon as well.

At first the writer will only support the object geometry, I will have to see if I can implement advanced features like animations or attachpoints. I will do some more study on the format for that later.

FSDeveloper not reachable

At the moment the FSDeveloper site seems not reachable. I have asked the webhost to look at this problem, so hopefully the site is up again soon.

EDIT: By now the problem is solved and the site is up and running again. Sorry for the inconvenience.

We almost got a release candidate

I almost finished the release candidate for ModelConverterX 1.0. I have decided to stop trying to make the LOD Creator more complex (and hopefully better) for the moment. There are some mathematical challanges in that, which I will take on again later. But compared to version 0.55 so many things have been improved already now that a new release is deserved.

So at the moment I am doing some final checks with my collection of test objects to see if no new bugs have slipped in. After that I will start updating the user manual. I hope to finish those two activities in a day or two, so by then the new release should be there.

Time for an update

It has been a while since I wrote on this blog, other things kept me a bit too busy lately. So it is time for an update!

I am still working on the next release of ModelConverterX. I had hoped to have the version 1.0 release ready by now, but the last few bits seem to take slightly longer than planned. The only main feature I am still working on is the LOD Creator one. I want to add functionality that considers the texture mapping of the object as well. This however means I have to update my matrix inverse code and that is what I am working on now (I’ll save you the math details).

After the version 1.0 release I am planning to close down the beta testing group that is around now and make the latest development build available to everyone. So that users have the option to download the latest stable release (that is stable) or the latest development release which has all recent features (and maybe some bugs). But first I need to focus on finishing the current version for release.

Converting SCASM animations

This week I made some good progress on ModelConverterX again. The ability to convert animations from SCASM macros was on my wishlist for a long time already and I have now taken the first steps. Animations that are defined with a SCASM TransformCall command are supported. For these animations you specify for each axis which timer should drive the rotation around it.

Other types of SCASM animations are not yet supported, but I will try to add them as well once I come across good samples to do so. So after this version is release, please report any problem you encounter. Speaking about a new release, with the animations implemented, the next release is getting near. I still plan to make the LOD Creator functionality a bit more user friendly and then the next release can be done.

Is this the end?

The recent developments of Microsoft shutting down the ACES studio are making me a bit sad. Let me try to explain why.

When FSX was released it included some great improvements over FS2004. I am talking from a scenery designer perspective of course, so I was most happy about the improved terrain engine, that allows more resultion that 4.8 meter on the photo scenery, a better autogen system and the improvements to the modelling SDK, which means that more things can be done without tweaking the code. In my job I am also working with visual simulations and I can assure you that most of these systems can simulate the entire world with the ease that FSX can, so from that point of view it is a great product.

But from a developer point of view FSX is also a big step backwards. Personally I think developing for FS2002 and FS2004 was most fun. In these two versions you could really build anything you imagined. The scenery design techniques were very flexible. Want to make a completely working docking system, of course you can. Want to make an animation of a hangar door that is trigger by the approaching aircraft, of course you can. I have great memories of how we were searching for those limits in the community.

But in FSX that is all no longer. The scenery format is so restricted that I don’t even have to think about a conditional scenery object, triggered hangar door animations or seasonal textures on my object. And then I am not even considering a completely working docking system, which is much more complex. The progress on the points I mentioned above have removed all the flexiblity to make interesting and nice things for the developer.

All this time I was hoping that this step backwards with FSX, would be the price we had to pay to make three steps forward again in the next version. I believed that if we would use the product we might find some work arounds and in the same time we could make sure ACES knew about the flexibility we wanted, so that the next version would be better again.

Now ACES is closed, the development of a next version will at least take longer. And personally I am not sure if I can keep motivated to develop for FSX given all the restrictions it puts on the developer. I have always enjoyed most to find the edge of the product, make realistic and complex objects, like working docking systems. FS2004 was great for that and FSX makes that almost impossible. Maybe it is time to move back to the FS2004 techniques and start making those interesting and challenging objects again? I guess the future will tell…

Microsoft closes ACES studio

Yesterday while driving home I heard on the news that Microsoft would be cutting jobs. But at that time the thought never came into my mind that the consequences of that would be as big as we learned today. While looking at some simulation related news at work today, I suddenly came across the message that the Microsoft job reduction would have big influences on the ACES studio, which makes Flight Simulator, as welll.

After looking around the web for more information, things became more clear. Former ACES team members posted more information, like Steve Lacey on his blog or Phil Taylor on the AvSim forum. All the current information seems to indicate that all development on FSX, Train Simulator 2 and ESP have been stopped at this moment. I guess the coming days will make things more clear how things will evolve.

But how this affects the people that worked at the ACES studio is of prime importance at the moment of course. I hope they are all well and will find other employement soon. All I know is that they made a very clever simulation product, that is capable to simulate the entire world in great detail. Something they can be proud of.