From: <luc...@li...> - 2015-03-13 22:41:48
|
Dear CRRCsim developers, I've just pushed quite a large number of updates concerning: - a new powered aerobatic model (Sebart Angel S30E), - fixes and updates to propeller modeling, - a new F3A "game" mode, - a new "smart camera" camera mode, - and a few nice improvements to scene rendering (corrected lighting and shadow, user defined sun position, lense flare effect, fog effect). The following two videos show some of the improvements: https://vimeo.com/121510640 https://vimeo.com/121598832 It all started when I wanted to create a model of my new aerobatic airplane, the Sebart Angel S30E, and practice aerobatics. I felt that the visual feedback was a bit of a weak point and started some effort to try improve it. At the end I spent a lot more time working on CRRCSim development rather then on improving my piloting ability, which was the original aim... Those of you who wish to get more info about the updates may read the following notes. One further comment is about issuing a new release of CRRCSim: the last is now three years old and is missing a lot of significant upgrades. It would be nice if anyone of the project administrator will have the time to do it. Any chance ? Luca ----------- F3A GAME MODE ============= This is not really a game but rather a mode where a user defined grid is drawn in the sky to help checking airplane's trajectory. You can also select to display both the trajectory in 3d space (gray line) and the trajectory projected on the grid plane (red line). When flying negative g's the trajectories are dashed. You can further activate two instruments showing respectively airplane's bank angle and airplane's CG trajectory angle to the horizontal plane. SMART CAMERA MODE ================= With "smart" camera mode I tried to get a better feedback about airplane's position in the sky. The idea is the following: - define a rectangular region of the sky, in terms of range of azimuth and elevation. - when the airplane is in the center of the region it will appear in the center of the window. - when the airplane moves toward the top of the region it will also move toward the top of the window (minus some margin) and similarly when it moves toward the bottom of the region. - when the airplane moves towards the left margin of the region it will also move toward the left of the window (minus some margin) and similarly to the right. - when the airplane moves beyond the edge of the region the window "moves" following the airplane, which thus remains close to the window's edge. Thus the position of the airplane in the window gives a feedback of the airplane's position in the sky, which is not the case when the airplane stays in window's center, as in standard mode. The definition of center and extent of the sky region is done in one of two ways: - if a game mode is active (F3F or F3A) then CRRCSim (or better the game handler) knows exactely which is the region of the sky where the airplane should fly (the area between the two basis in F3F mode, the area of the grid in F3A mode). - if no game mode is selected then the region (center and extent) is constantly estimated and updated based on the history of the last few minutes of flight, thus a precise region is identified if the ariplane is mostly flying back and forth in one area of the sky. The fraction of the windows (minus some minimum margin) which is mapped to the sky region can be selected in the "video" dialog: a value of 0 switch smart mode off, while a value of 1 allow to move the airplane around most of the windows. Give it a try and decide yourself if it is worth using, in place of the standard mode. LIGHT & SHADOW IMPROVEMENT ========================== I noticed that the lighting of the airplane model (in fact of any object in the scene) was not consistent with the definition of directional light source (the sun). Additionally the model's shadow on the ground (in volume shadow mode) was not projected properly so that the overall scene was looking unreal. This was caused by a few bugs which have been fixed, so that now light and shadow are consistent and, as a result, contribute to a more realistic visual feedback. In the process I've also added the following features: - the position of the sun can be defined in the scenery description file. For a given scenery, each sky option can have its own settings. >>> This means that all Joel Lienard's nice photorealistic sceneries can now be configured to match the sun position appearing in the pictures. - the colour and the intensity of the sun light (and of ambient, i.e. non-directional light, as well) can also be defined, again to better match the sky box pictures. - opengl fog effect has been added, to (optionally) simulate haze effect. The colour and density of the haze can be defined for each scenery's sky. Of course the fog is not superposed to skybox pictures, just to objects (including airplane model) appearing in the scene. - the nice PLIB's "lens flare effect" has been added, to render the effect of the sun through the lenses when the sun appears in the window. This gives another nice touch of realism when flying. The effect is on by default but can be turned off if the sky has dense clouds and it would thus be inappropriate (e.g. in "Dome of thunder" sky option for Cape Cod scenery). The lense flare effect is available for both sky box and sky dome. - in the standard "little sky box" the sun appeared to be in a somewhat strange NW azimuth; I renamed the pictures so that now the sun is at a more reasonable 210° azimuth (SW). - all standar sceneries have been updated with appropriate sun position and a slight haze effect. See, e.g. the "simple" scenery with any of the sky option. PROPELLER MODELLING =================== Fixed bugs concerning the effect of propeller's position and downthrust, and in the process added the following as well: - also rightthrust can now be specified. - added a simplified model of the P-factor effect. - added option to set rotation direction and set clock wise direction (seen from the cockpit) as the default one (as usual on rc-model). This affects torque induced roll. - added an empirical static thrust correction factor (taken from Gabriel Staples work, http://electricrcaircraftguy.blogspot.it/2014/04/propeller-static-dynamic-thrust-equation-background.html) to try better match real-life performance. - since it seemed that the empirical factor was somewhat too strong in reducing the thrust, I also added optional user defined "tuning" factor in the propeller's definition to tune: static thrust, flying speed, absorbed power and P-factor. A factor to tune torque effect was already present. I quickly checked that helicopter and multicopter were still working properly, and updated the power documentation. AIRPLANE MODEL ============== Airplane model definition can now include, in addition to a launch default, a mixer default as well. Mixer default can be automatically loaded and applied similarly to the launch default. This is very handy when changing from one model to the other, since each model may require its own mixer's settings. At present only a few airplane models include a mixer default (e.g. Wasabi, Angel and Generic_F3F). Finally I've added a model of the small electric aerobatic aircraft Sebart Angel S30 E, plus the relevant engine, propeller and battery definitions. I've tried to match the real behaviour as much as possible but some tuninig is still required. Anyway, it is a nice new model to fly with. That's all, Happy flying. |