Because this is the first ship I have designed I needed to learn a few things. What follows are some experiences.
To me, learning gMax has been difficult. First attempts to use it discouraged me and I did not touch it again for at least a year. Because so many great aircraft designers were using it, I decided to give it another try. It was like those trying to stop smoking, I tried to learn gMax and quit several times. The processing speed advantages in FS became a real factor and finally I purchased "The gMax Handbook" by Clayton Crooks II which supplied enough hints to really get me involved in gMax. The gMax tutorials and those provided by MS were also very helpful. Once a base of knowledge of gMax was learned I found the gMax help system to be very good and supplied very much more information than anything else I have used. Hopefully MS will find a way to provide gMax to the FS community in the future. Its apparent demise is discouraging.
The Belle was designed with gMax. The ship had to be designed as an aircraft with the ship oriented with the front of the ship pointing in the positive y direction and then exported as an aircraft. The length and width dimension of the Belle were obtained by measuring satellite/aerial photos from the National Geographic Survey/terraserver web site. Since I live in Louisville I made several digital photos of the Belle for textures and design details. The design has been very much simplified and excludes features such as a few flags, support cables, the calliope and many other small details.
I still do not know most of the innards of the air, aircraft.cfg and effects files but was able to learn enough to modify those by Holger Sandmann to float the ship correctly, and to activate and position the effects to suit the Belle. Some effects are activated within the aircraft.cfg file using the [LIGHTS] section and the [EFFECTS] section. While trying to use the [LIGHTS] section to display some of the effects, the result was that the effects appeared as a spectacular blazing conflagration at night, but of little practical use. Texture files used by these effects include fx_1.bmp, fx_wake_1.bmp, and fx_wake_2.bmp. Also used are the effect files named fx_navred.fx, fx_navgre.fx, fx_navwhi.fx and fx_beacon,fx. These files are believed to have been included with FS2004.
I learned to use the clone capabilities in gMax to copy paddle wheel components and to reposition them and reorient them to save a great deal of time. Cloned objects were textured before cloning them to save having to apply textures separately to the clones.
Because the Belle design became rather complex and as additional parts were being added it was convenient to build those parts away from the main Belle objects so as to isolate the new part and to see it in all views clearly. Some of these parts were later cloned while the original part was left some distance away from the main Belle object. When the isolated part was some distance away from the Belle and exported into FS the Belle would not appear in FS. There seems to be some limit as to how far these isolated objects may be from the main in order to successfully appear in FS.
The animation of the wheel, cam and drive shafts were interesting learning experiences. First the axle of the wheel was created and experimented with using the powerful animation features of gMax to get the wheel to rotate at a speed that seemed to agree with the set speed of a taxiing vehicle as set by FS2004. The Link features were used to propagate the motion of the axel to the paddles and cam. A LOOK-AT contraint feature was used to simulate the reciprocating drive shafts and appeared great within gMax but failed to operate properly when exported to FS. The drive shaft motion was finally simulted by calculating positon and angle orientations of the shafts and reorienting the shafts by those amounts through the keyframes and in synchronization with the wheel frames. Four keyframes were used for the wheel and eight for the drive shafts.
An attempt was made to simulate a flag which had a wave motion, where the flag was distorted in a wave shape and which traveled across the flag with time. It was successfully made in gMax but did not export to FS. The final flag simulation had the initial wave shape but only was able to flap back and forth.
The bell atop the Belle was created starting with a multi section, in the z axis, cylinder. The mesh was first adjusted in height and a scale factor used to shape the bell at each level.
One must assign new names to objects that are to appear in motion in FS. Objects in motion in gMax will not be in motion in FS unless the objects are renamed with prefixes of tick18. The Belle required three objects to be renamed, tick18_wheel tick18_shaft and tick18_flag. Only three objects required the tick18 prefix because the other motion objects, such as the cam and paddles of the wheel, inherited their motions through Linking.
