Drawcall minimizer video

I have made a little demonstration video that shows how you can use the drawcall minimzer. In this video I will be converting a KMZ file to a MDL file. In that process I am using the mass texture editor and the drawcall minimizer to make sure the textures end up as efficient as possible. In this case I could reduce the amount of drawcalls from 45 to 6. Enjoy the video and if you still have questions about the functionality please ask them on the forum (in the absense of a manual).

ModelConverterX usage survey

To get a better understanding of how you are using the ModelConverterX tool I have setup a survey. So if you are using ModelConverterX, I would appreciate it if you spend some time to answer those questions. Knowing which functionality, importers and exporters matter most to you, helps me to make the tool even more useful in the future.

For example I have some ideas to change the GUI so that information is presented more clear. But to know if this would really help, I would like to have a better understand if you are normally running ModelConverterX full screen or not.

So thanks already for taking some time to fill in this survey. The survey is on the same server as the ModelConverterX development release and might take a few seconds to load.

Drawcall minimizer

Today I have been working on the drawcall minimizer and it is going quite well. For the diffuse textures I can now reduce the amount of textures by combining them onto one texture sheet, after which the texture mapping is updated automatically on the model. The biggest problem with this approach is that other textures (like night textures) are not updated yet at the moment. And also I would like to add the functionality to put the colours into the texture sheet as well to minimize the amount of drawcalls further.

But if I can’t fix those issues quickly, I will release the current functionality soon so that you can also use it. While working on it, I have also speeded up the loading of texture files  a bit, this should be mainly noticeable in the material editor.

Jetlag almost gone

We are back home for a few days now, after enjoying a two week vacation in California. We had a lot of fun there and now that the jetlag is almost gone, it is time to spend some time on my FS tools again.

Initially I plan to focus on ModelConverterX a bit more. I want to finish the function to minimize drawcalls and will try to fix some bugs in the COLLADA reader. Since my posts about using Sketchup it seems many people give it a try, and they find many new bugs as well.

After that I want to focus more on gPoly again. I got some great new ideas for this tool and I am looking forword to continue coding on it. My first focus will be on the functionality to actually export the polygons you draw to FS, since without that the tool does not have that much usage.

San Francisco here we come!

Just a quite note, at the moment we are packing our bags and tomorrow we are flying to San Francisco for a two week vacation. So that means no computer, no FSDeveloper, no email and no tool updates. I am looking forward to this break and hope to come back with many new ideas for tools like ModelConverterX and gPoly as well. So see you all in two weeks again.

Minimize drawcall functionality

One of the challenges when converting old API macros or COLLADA objects is often how to make them  framerate friendly. It is not uncommon for such objects to use many small texture files. If you for example load an object from the Google 3D warehouse, don’t be surprised to find it uses 40 different texture files.

For some time it was on my ModelConverterX wishlist already to add some functionality that will merge all these textures into one texture sheet and automatically adjust the mapping on your object as well. Today I started coding on this.

I am certainly not finished yet, but the first results are very encouraging already. The image on the right shows my test application, where I loaded a bunch of textures from a KMZ file. All these pieces have been combined into one big texture sheet.

What I still have to do is make the code to update the texture mapping on the object. And of course to make a user interface to access this drawcall minimizer functionality. But that should not be the hardest part.

So hopefully later this weekend I can have the first beta version ready. But now I am first going to enjoy the snow outside and have a walk.

openVFR

If you are looking to add more details to your FSX, have a look at the interesting project openVFR. It uses OpenStreetMap data and it really makes the scenery look a lot more realistic. With a bit of luck you can find your own street. I think this kind of scenery is a great alternative to photo scenery, especially when you are short of harddisk space. So have a look…

Mass Texture Editor

To assist with the conversion of textures while converting objects for us in FS, I have added a new tool to ModelconverterX. It is called the Mass Texture Editor tool and it will help you to rename, resize and convert to another format all the textures of an object at once. So this will be useful for example when you are converting a COLLADA file, since these often use JPG textures in sizes that are not supported by FS.

I have made a little preview video to show how the new tool works. I hope you find this functionality useful, if you encounter bugs or have other suggestions please post them in the forum.

Some of the things I have in my mind to add next are functionality to minimize the amount of drawcalls and a wizard that will guide you easily through the import, texture edit and export process of an object.

Free?

A few weeks ago I read the book Free by Chris Anderson. In this book it is described how many companies (mostly on the web) nowadays can make money using free products. If I try to summarize it in my own words, the author is saying that since the cost of distribution of digital content is approaching zero, the price of such products seems to do the same. And therefore people have to look for other alternatives to earn their money.

An example given is that many artists earn less from selling their records, but earn more from their concerts and merchandising. Or the so called freemium model where there is a free and a paid version, where the paid version has more functionality and from the earnings of that the free version is kept free. Or another example is free things paid by advertisement.

I am not sure if I completely agree with everything written in the book, but it is a nice mind experiment to see how this could apply to the world of Flight Simulation. So let’s give it a try…

Since addon products for FS are in general also digital content, according to this book their price should also develop in the direction of zero. But I am not sure if the alternatives mentioned in the book would work that well for FS addons. Using advertisement of something like that to generate the revenues is not something that would work I guess. Would it work to have a free and a paid version of addons? Maybe, but which features should be in the paid version to convince users to actually pay for it? Would it work to have a basic version of a scenery and then if you wanted things like animated gates you would have to pay? Or get the simple version of an aircraft for free and pay for the virtual cockpit? I am not so sure about this and would such a schema actually attract more customers than addons get nowadays?

Of course we have the difference between freeware and payware in the world of FS addons for a long time already. Where for payware you can in general expect a better quality and better support. Although there are also freeware projects that deliver top quality of course. But I can’t imagine more products being offered for free, simply because the amount of time needed to create them needs to be compensated somehow.

So while I am dreaming about how things could work, let me consider another model. Would it work if the Flight Simulator itself was offered for free and the revenues would be generated by the users who want extra functionality? So everybody can get the basic version, with some aircraft and a basic scenery of the world for free. But if you want better aircraft or scenery you can buy these. Or maybe you have to pay a monthly fee to use the ATC services? The advantage of a free basic version would be that you can reach a bigger group of end users and that might grow the total FS community. But I am not sure if additional paid content could pay for the development of such a base version. Especially since traditionally there are many small companies making such addons.

So I guess maybe the way the FS world works today is not that bad, where you have a product that has been designed to allow a community to extend it. And both freeware and payware developers do so. I guess the future will tell if a successor for MSFS will use a similar approach or not. As long as it is an open product that can be extended most of the community will be happy with it I guess…