posted 01-18- 01:29 PM               

A possibly related problem is that, on landing, once the plane slows below about 80 mph or so, it immediately slews around 180 degrees and stops! Now, that might be good for making sure I can land on the carrier, but it's not real realistic It does basically the same thing if you push up the power for takeoff and then pull it back to idle at about 50 mph. If you leave the power up you can keep it straight with the rudder for the takeoff, but if you try to just "taxi fast" it'll slew around and stop.

Any ideas?

posted 01-18- 05:25 PM               

------------------
Zoycite {GS}
GUNSLINGERS

visit The War Paint Factory

posted 01-18- 07:23 PM               

I had the same problem and I fixed it by taking all of the orientation out of the Wheels parent and grandparents, and putting the rotation in the mash instead of the Model matrix.

If you get B1.1.1 of OPS It should have a reset XForm button which will help get rid of the strut and piston rotations. If this doesn't make sense let me know what I've turned into gobbledegoop.

posted 01-18- 08:18 PM               

I do have B1.1.1, but have been using the non-beta for most work. I can play around with it some using the xform thing.

Is it possible to make the wheel's parent the strut instead of the airplane?

P.S. both wheels are turning visually...

------------------
--jedi--

posted 01-19- 04:54 PM               

To try and explain my techo babble above:

There are two ways to get the right rotation in parts on a aircraft:

1) Make it anyway you want in AC3D/Max etc and then rotate it in OPS to get it to look right.

2) Setup the rotation in AC3D/Max so that it doesn't need to be rotated in OPS

For things like wings and airfoil like objects, the angle of a part effects the way things fly, so the best option for this is (1), probably lose the "Make it anyway you want" statement though since the angle the wing is rotated and the visuals need to be considered, but the point is that things like wings need to have some rotation defined in OPS to get the aerodynamics right.

For static objects (aerodynamics wise) such as the wheel struts and pistons, it really doesn't matter how much or little rotation is included in the part in OPS. So IMO the best way to make these things is to do all of the rotation in AC3D/Max so that it imports into OPS looking good without the need for rotations. Normally this would only be a personal preference of mine, but becuase I have a bug with OPS that can't handle setting the DOF's when parent models are rotated too much, this is probably mandatory, for the short term at least, so that you can use OPS to do your DOF's.

As I said, I had almost exactly the same problem you're having, but doing as above solved it for me. The new align dof tools in B1.1.1 also have problems with the rotations, so once you sort this out you should be able to get precise alignment on axis and origin using these tools.

I hope thats more understandable.

posted 01-19- 06:50 PM               

Suppose I built a part in AC3D, and then rotated it in OPS. If I then import that part back into AC3D, will it have the rotation, or not? If I rotate that same part in AC3D, will it then look "rotated" in OPS, or will OPS remember its prior orientation?

Hehe actually I'll probably know the answer by the time you get back to me, but humor me anyway

posted 01-20- 05:59 AM               

When you rotate something in OPS you are adjusting the tranformation matrix (the transformation matrix defines position, rotation, and other transformations), not the actual vertices etc in the Mesh/LOD. As I understand it, AC3D adjusts the vertices when you rotate, at least thats the effect on the exported .LOD file / .ASC file.

posted 01-20- 06:45 AM               

Is there a way to align things manually in the text file, i.e. some trick with the vertices to make sure the wheel is actually pointing "straight" forward? The only vertex information I can find in OPS itself appears to be related to the centroid of the wheel and the size of the cylinder it makes.

I can do the Mustang-import thing to get rid of most of the wobble, but it's just irritating to be unable to get it right

I suppose I COULD import ALL my Corsair parts into the Mustang model, being particularly careful with the gear, but that's almost like starting from scratch...

What if I just delete the LODs for the gear completely, insert blank models, and then import known "good" LODs like those from the P-51?

------------------
--jedi--

posted 01-20- 06:47 AM               

1. Always have the object in AC3D aligned with the x,y, or z axis. DO get it facing the right direction though. Use the rotate command under the edit menu - type in degress (+/-90) and axis.

2. Rotate as needed in OPS. Most parts should not need rotation - wings need dihedral though. I also rotated my windshield back for my SE5a.

Doing this makes AC3D editing of parts posible. If parts like wings become rotated in AC3D it is IMPOSSIBLE to edit that part! Also, how would you get it to the correct angle? Also note that flaps/ailerons are both imported level with the wing. Then position them relative to the wing. Now rotate the wing into the correct dihedral and the flaps/ailerons will rotate with the parent.

If you look at the shipping SDOE planes you will see that almost every part is straight on one of the axis. The only exception is for strange diagonal parts like the inner wing struts and wheel struts on the Fokker DR1 - these objects don't have an easily discearnable face. These parts tend to be built on an axis parallel to the fuse object or something.

Why is this? Because 3D aps like AC3D rely on top-side-plan view editing, so you need to object to have flat edges along axis in order to select vertecies for editing.

This is one big reason not to build your whole plane in AC3D first and try to export each object after.

Ever try and break your wing into a top half and a bottom half? In AC3D this is impossible unless the wing is not rotated. This is why I had to rebuild my SE5a from scratch in order to texture the top and bottom of the wing with seperate textures.

How about the other 3D aps (3DSmax, etc.) - is this an issue for those programs? Any other insights on the modeling process? This is interesting...

------------------
-Sv =FC=

WWI in SDOE!

posted 01-20- 08:36 AM               

You touched a little bit on something that concerns me bigtime. When importing a wing into OPS, if I have to do ANY rotation to it I always end up fudging the flight model, and it's hard to fix it as well. So, what I do is export a wing, for example the left Me262 main wing. I then import it into Max as an Obj file, look at it's rotation, and build my new wing. I position my new wing's wing root at the SAME point in space (that same axis I should say instead), and then I delete the 262 wing and export the new wing obj. When importing it back into OPS, if it's not perfectly aligned and req's me to move it, I go back to my previous Obj file in Max, move it more according to what I just saw in OPS, and then export it again.

This is the ONLY way I've been able to keep the FM rock solid stable. We all saw what happened with the Go229, it flew horribly and even to this day, still is whacked pretty bad. So for my F-14 and F-15's the FMs are solid (expcept when the 14's wings are swept back... I need to align left and right dofs perfectly for perfect high speed flight). Right now left wing swept back is a little higher than right wing, so it's hard to keep level.

Anyhow, my entire point is that I wish there were an easier way to deal with importing wings, and not worrying about fudging FMs. But for now I don't see any other way. (That too explains why some of my objs don't have the proper center pivot points, because I didn't want to throw the fm off).

posted 01-20- 09:12 AM               

Are you talking dihedral rotation or incedence angle? Minor incedence tweeks can be made via an FM setting on the wing without changing rotation of the model. What rotation are you talking about?

posted 01-20- 11:59 AM               

You can automate this process by rotating the wings level in max, then picking align pivot point to world then rotate them back.

TS

posted 01-20- 03:31 PM               

Tailslide, I'm truly not following you. Could you explain more?

While in OPS importing a wing, unless I don't have it prealigned in Max I always fudge the FM when positioning it in OPS. Oh, I just realized what I've been meaning to say before. It'd be very cool if you could import an object (a wing in this case) on one side of the model in OPS and have the other wing upon import match the exact position of the first wing, except on the opposite side! This would also be great for elevators/hz stabs as well.

Without telling me how to do it, just explain how you guys import into OPS to keep the FM good and working well. That'll probably help me.

The Tomcat does have a difficult wing position with the anhedral and the fact that they are swept a little as well. If I can position a wing without moving the inertia boxes, well, I didn't know that. Also I've heard about sizing inertia boxes with the mouse, but I've yet successfully do this. How does sizing work?

posted 01-21- 12:24 AM               

I know that OPS is very weak when it comes to accuracy of positioning and rotation, and it would take a large amount of work to add all of those little nicities. This is why I am concentrating on scripts etc that not only import meshs from MAX, but also position and rotation. (sorry AC3D guys)

The things you speak of can easily be done acheived in MAX, and once I get the import right they will be pulled into OPS and hence OpenPlane.

I had hoped to release what I have so far, which actually does this using a OBJ and a MAXScript, but the problem is that it requires certain setup in MAX for it to work, so I wanted to write detailed instructions to stop the certain flood of questions.

Problem is I'm really busy at the moment so what I might do is release it as a "Downloader beware" thing for the moment.

posted 01-21- 12:20 PM               

-rotate your wings level to the horizon in max
-select the wings
-in the heirarchy tab pick "affect pivot only"
-pick "align to world" to align the pivot axis to the world
-rotate the wings back to whatever angle they were in the first place when attached to the plane.

Now when you export into OPStudio the wings will come in so they are level with the top and bottom of the yellow bounding boxes. Rotate them so they line up with the plane.

Doing this causes the flight model to model the dihedral of the wings.

posted 01-21- 02:31 PM               

Problem solved, once you guys explained the nature of it. Had to delete the LODs for my landing gear in their entirety, then copied the "good" main struts from the P-51. That "dragged" all their "children" (pistons, wheels) with them, properly aligned, into the Corsair. As long as you don't rotate the wheel DOFs separately from the struts, they'll stay properly aligned, no matter what you do to the struts themselves in OPS.

Then just a little nipping and tucking to make the Mustang struts look like Corsair struts, and I'm back to where I was a week ago only with straight wheels

She still pulls into a 180-degree skid and stops anytime you try to go above about 20 mph or below about 60. Makes the middle of the takeoff roll pretty challenging, but landings are REAL easy

I THINK I may have a body vertex from the "hidden" body parts contacting the ground and keeping the tailwheel from making "good" ground contact. I can see the tailwheel turning at first, but with any speed at all, it seems to stop, which I suspect means it's not doing anything to help me go straight, at a speed where the rudder doesn't work well yet. I've noticed if I rev the engine, I can use prop blast to flick the tail around, but once I back the power off, the fuselage has a mind of its own.

Right now I have the options set to couple the tailwheel to the rudder for steering. Any opinions on which way works best?

posted 01-21- 05:24 PM               

-rotate the wings back to whatever angle they were in the first place when attached to the plane.