Posts by MDIvey

    Thanks Jan.

    Well I've done the revised Aerodynamics for the Snipe and I'm very pleased with how its turned out. It really feels like a heavier but more powerful version of the Camel which the real thing was. It no longer climbs like a homesick angel, if fact it needs some up trim and rudder trim to flight straight and level, and to counter the torque of that huge engine and prop. Another notable improvement is the glide when the engine is cut. Before it was a really steep glide angle, but its now much more realistic.

    KR Matt

    Jet-Pack (IPACS)

    Hi Jan

    I've been gathering data up with a view to tuning the Snipes Aerodynamics. It seems likely that the airfoil used on the Camel and Snipe where the same... an RAF-16. I can find no data on that, but there is for the RAF-15:

    RAF 15 AIRFOIL (raf15-il)

    Not having a degree in aerodynamics to fall back on, I think I would struggle to interpret this for AF 4. I'm hoping I can figure out how to change the other bits of the aerodynamics ok, but I was wondering if based on this data there are any changes you would make to the Camel Airfoil code to be more accurate on the Snipe. I'm assuming it was you who did this Jan and no doubt did a very good job.

    I notice on the Camel wings the upper wing is in two parts, which is fine because it was straight. The Snipe top wing is in three section. The two outer portions have a dihedral to match the lower wings, but there is a straight, mid section that has a shorter chord length and I was going to try and have a go at that... Is there anything I should beware of in doing that?. I'm still in the process off reading, re-reading and digesting the stuff on the WIKI about this

    KR Matt

    Jet-Pack (IPACS)

    Jan I've taken the wing tips out of the main model TGI and made the two types of wing tips, option items, that can be added to repaints as required. I'll send you the updated files when I get a chance so it can be integrated with your master copy of the TMD, because it does require a few minor edits to get it to work. The Wing Tip collision hulls have been moved to the Wingtip Option system file, and the geometry names for the wingtips graphic objects need taking out of the geometry section of the TMD.

    KR Matt

    I'll be surprised if it is that simple... as you say they are usually on by default, but always worth a check. Otherwise you'll need Jan to chip in. The lighting is definitely an area that could do with some clear documentation by IPACS. I made a lot of mistakes and had to bother Jan a lot with questions when I was working on it the first time with the Buffalo. Anyway hope you have success... the Carbon Cub will be a gem.

    KR Matt

    It could be a problem with IPACS blender export, but maybe check you actually have the panel lights switched on. These are the panel and instrument light switches... include a variable brightness... Left is panel, right is instrument

    with panel light off and instrument on it looks like this...

    with panel light fully bright looks like this...

    KR Matt

    Jet-Pack (IPACS)

    I've been through all my models and changed them to remove those collision parameters, but I will wait to upload them to flight-sim.org until the new version goes main stream, even though it seems to work ok with the current version I have. I haven't done anything on the Duo Discus as that is in your hands at the moment so I'll leave you Jan to edit those bits if they apply.

    KR Matt

    Coding is getting more sophisticated, and I can now stop engine, but it isn't yet controlled by cockpit controls. Also custom engine sounds are now in place. Most of the repaints I did for the XP11 model have also been converted for the AFS4 model.

    KR Matt


    Thanks Jan

    Got it working.... finished code below. Something I hadn't realized about the collision entries, was that in the geometry line, it needs to be named as the object is in the TGI... in this case "tailskid" not "TailGear"

    KR Matt

    Physics Section:

    Geometry Section:

    Code
                <[hingedbodygraphics][TailGear][]
                    <[string8][GeometryList][ tailskid ]>
                    <[uint32][PositionID][Fuselage.R]>
                    <[uint32][OrientationID][Fuselage.Q]>
                    <[uint32][InputID][ServoRudder.Output]>
                    <[tmvector3d][Axis][ 0.0 0.0 -1.0 ]>
                    <[tmvector3d][Pivot][-4.178 0.000000 0.175]>
                    <[float64][Scaling][0.25]>
                >

    Hi Michael

    I never did manage to get rid of that odd square near EGCK. I'm thinking when I set up the new drive, I might try installing this afresh. Just to clarify... there are separate EGOV and EGCK files and both these are contained in the file Wales Airfields_Version2.zip.... Do I need both installed and in what order? Or are the separate files intended to replace the files in the combined package? The West Wales package is that a separate update to another west Wales package or a stand alone. Sorry I'm a bit confused.

    KR Matt

    Jet-Pack (IPACS)

    This is the Essence of a conversation I had in Private with Jan, that he thought should be posted here for the benefit of others attempting Aircraft development

    Hi Jan

    I really want to understand how the TMC contact positions and TMD Wheel Hulls and Physics Rigidbodies, affect the models position on the ground. Front wheels work great but rear end falls through the ground and I cant understand why? It looks to all intents and purposes as the Camel and DR1 so don't understand why what I did, did not work. Collision code is not currently present in the TMD. I tend to take that out for starters because it always throw out a lot of errors to start with (although I think I understand now why that is). Three Spheres appear correctly positioned in the TMC for the wheel contacts with ground.

    Note: The Snipe Tail skid pivots to provide a degree of steering unlike the Camel and DR1 which is fixed, but it doesn't move as much as the rudder (confined by a slot in the fuselage). That is why I've used a 0.25 scaling of the Rudder output for the Tailskid graphics animation

    KR Matt

    Visualization of Rigidbody set up... wheel rigidbodies are basically a box the size of the wheel so didn't bother to create those.

    TMC Contacts Code...

    Code
            <[vec32array][ContactSpheres][ ( 0.01334 0.7407 -1.29756 0.365931) (  0.01334 -0.7407 -1.29756 0.365931) ( -4.3579 0.0 -0.38231 0.0079985) ]>

    Tailskid TMD Physics Code...

    Tail Skid TMD Graphic Body Code....

    Code
                <[hingedbodygraphics][TailGear][]
                    <[string8][GeometryList][ tailskid ]>
                    <[uint32][PositionID][Fuselage.R]>
                    <[uint32][OrientationID][Fuselage.Q]>
                    <[uint32][InputID][ServoRudder.Output]>
                    <[tmvector3d][Axis][ 0.0 0.0 -1.0 ]>
                    <[tmvector3d][Pivot][-4.178 0.000000 0.175]>
                    <[float64][Scaling][0.25]>
                >

    Yes Please IZ0JUB , The IPACS default stuff for Florida is not terrible apart from the coastal Jaggy edges, but I've got a monster SSD to put the scenery on now (when I have the time to get it set up) and an upgrade would be nice for sure.

    KR Matt

    Have you looked at the DR400 textures supplied in the SDK? It was looking at those that helped me to work out what to do.

    That does not look right for a texture with _illumination suffix.... more like a _light one. Both the illumination and the light texture are involved with how the object looks in the dark when the lights are on. It shouldn't have any effect how it looks in daylight.

    To be specific about those two textures... the illumination texture is used to apply a diffuse lighting effect to the cockpit interior, and requires a point light to be set up for it in the TMD. The light textures is for lit up labels on gauges, buttons & dials etc, which are switched on by the panel light. I didn't bother with using the illumination, I just used the light textures, to light up all the dials and buttons etc for night flying... and that suited the Military style aircraft I was working on.

    More critical to how the object looks in daylight are the color, ambient, normal, reflection, specular and specular alpha textures:

    Color.... obvious this is your repaint texturing

    ambient... Ambient shading application. I really like how this make things look. Its a lot of extra work because everything needs mapping out single space and then rendering the AO map for each object... which can take quite a while... I think my Spitfire fuselage object took about a half hour. When the grey scale AO mapping is done it can blended over a white background and you have your ambient texture.

    Normal.... bump mapping of panel lines, Rivets fabric, wood and surface texture. Created from a greyscale (but RGB) texture by a photoshop plugin in my case (but other software options are available for that). I only use this for external objects. It does not work for me on internal surfaces or transparent items like canopy glass.

    specular and specular alpha... greyscale textures work together to produce shine. On Spit both where very dark to give matt finish. It seems to me darker grey is less shine, lighter grey more shine. These snippets from another model so you can actually see something...

    specular alpha

    reflection... This texture creates reflection when sunlit. Greyscale blended over a red back ground. Lighter, More Red and saturated more reflection. Darker, less red less reflection seems to be rule. Here is a snippet from Navy Buffalo where I wanted paintwork matt and metal shiny reflective..

    light... mostly a black background with details in red or whatever colour you want the labels lit up in

    Anyway hope this is of some help

    KR Matt

    Sopwith 7F1 Snipe. Another of my models I've been tinkering with to try and get it into AFS4. Its already had a life in XP11, for which I created it, but it would be good to have it in AFS4 as well. Not sure if it will ever get completed... we'll see

    KR Matt