RedBeard2

Apologies for being absent so long. I spent most of last year dealing with aging parents and had little time to dedicate to anything else. I'm going to start spending more time on this again and hope to get some more stages out soon. Current state is animations are mostly done and the writeup is mostly in sync. I need to spend time figuring exactly where I was and what else I needed to say and cleanup. Other than that, I expect to get another draft out soon.

The - should not make a difference if they are in strings such as "Mq-9_Reaper" or filenames such as '/Input/Mq-9_reaper', but if you use them in LUA variable names, then the - is treated as a subtraction in an equation. Take a look at Mq-9_reaper.lua. If you have something that looks like the following: --mounting 3d model paths and texture paths mount_vfs_model_path (current_mod_path.."/Shapes") mount_vfs_liveries_path (current_mod_path.."/Liveries") mount_vfs_texture_path (current_mod_path.."/Textures/Avionics") Mq-9_reaper = { then you are writing an equation that say subtract 9_reaper from Mq. Remove the - from here and use the new variable name at the bottom as well in the add_aircraft function call. See the section called Renaming the Mod in my guide for details.

See my Beginners Guide to DCS World Aircraft Mods in the How To forums. I have several how to videos referenced there.

Also, Cya and Cymax are related. Do you get to Cymax before your Aldop departure point is reached? Mzalfa is tail pitch coefficient, Mzalfadt is wing pitch coefficient. Lastly, kjz is related to the elevator control power coefficient in some way but it is not obvious to me yet. See my aircraft mod guide for descriptions. There's also some threads around that discuss what happens where some of these values vary.

What kind of max AOA do you need and have you updated Aldop for that max AOA?

I'm currently trying to use a dummy on each end of the shock absorber and using LookAt constraints for the dummy and aligning the arg based controlled shock to the dummy to bake the keys. It only kind of works at the moment as any time you use need to check the flip check box, any alignment updates cause terrible side effects. I'm still looking for a better way.

Is the LookAt constraint usable in DCS World animations? I just tried it and couldn't get it to work, but Beczl said (a while back) that it did. I looked for the wmv file, but it appears to be gone. In particular, I have a landing gear strut with shock absorber and hydraulic actuator attached. If I use arg based rotations for the strut, I see the LookAt constraint working in 3ds Max, but if I try it in the model viewer, the orientation of the shock absorber parts are fixed and do not follow each other.

Here's the last part in the series. Let me know if you see ways to improve this.

Here's (part of) the next video. I had to break it into two pieces because it was getting long. This will explain the basics.

Heh, I gotta find some time for that last video since it appears that everyone is right. :-)

I keep the IK solvers and bones in case I need to make updates.

Part 2 on the HD IK Solver is now ready:

I've created the first of a 3 part series of videos on how to do this:

Success! I've finally managed to find a way to use IK Solvers to animate parts in DCS World. I'll be writing up the process in the guide and creating a video to demonstrate the steps. In short, I had to create a double bone structure. I use one set of bones for performing IK solutions. The second set of bones are controlled by arg based controllers and I use key frames to align the arg based bones to the IK bones as I animate. There are more workarounds that need to be applied than just that, but that's the core of the solution.

Hmm. Well the test didn't go as well as I expected and for multiple reasons. I had the IK chains working just fine. I could grab an animation controller dummy, rotate it, and the IK chain bones would animate the parts of the model as they should be. I then tried creating an arg based rotation controller for the animation controller dummy and animating it using autokey. I would move the time slider forward 10 frames rotate the animation controller dummy to the next position, and click key to create a key frame. So far so good. After I got out of auto key, I tried to play the animation back using the arg based controller. When I did parts started breaking loose and generally not going where they should though some parts did animate correctly. Just for grins, I tried exporting that and playing it back in the modelviewer. The IK animated parts were frozen. So, it appears that either I'm not baking correctly or there is no good way to drive an arg based controller which is used to drive IK. I tried checking more about baking animation and I saw that people were exporting to FBX and importing, but I'm still not sure how that gets driven by an arg based controller. I also saw references to IK chain support for key frames, but it wasn't clear how to use that or tie that to an arg based controller. If anyone has ideas, please let me know.

While I'm doing the last bit of testing, does anyone know if wire parameters work for animation in DCS or is that another not supported item that would require baking in the keys?

After lots and lots of experimentation and little documentation, I think I've figured out how I can do this using the HD IK Solver. At least it works in 3ds Max when I just have a bone chain. Now, I just need to apply it to the model, do the animation arg keys, and check if everything holds together outside of 3ds Max. Wish me luck. :-)

The HI IK Solver is not far from this, BUT you're limited to motion in a single plane. I can fake the single plane for the upper push rod by creating a coplanar point for the upper ball socket and connecting the root bone to it, however, I tried this with the lower control rod and also reversing the root and goal of the IK chain on the lower bone chain. It kind of works, but the geometry is such that the farther out of plane you are, the more of a problem there is. Creating a coplanar point works and it will follow the bell crank movement, but the geometry won't cause the wheel to fully rotate if that is done. I was hoping to avoid manually positioning everything as I still have a shock absorber, gear retract actuator, actuator floating assembly, gear doors, and gear door push rods all connected to the retracting strut. I may try to automate most of the joints using IK and just manually position the wheel mount.

Well, I've determined that the HI IK Solver will definitely not work for this. The HI IK Solver expects all joints to be coplanar and all joint movement will be in plane. The plane of movement that I need is defined by the 3 joint positions of the bell crank. This leaves the root joint of the upper push rod out of plane. I'm going to see if I can figure out the HD IK Resolver params and get that to work instead. Docs for it seem sparse though.

That's good info. Thanks. That still leaves the problem of getting the IK animation correct before 'burning' into the animation frames, which I guess I'll have to go read about (I'm assuming it's something more than just key frames at short intervals).

I added a link in the first message to a short video clip which illustrates the problem. I'm not sure why the embedded video doesn't appear to work, but the link will get you there. Once the problem is solved, I'll replace the video with a "tutorial" on how to get it all working correctly.

I'm trying to animate the complex main gear of the F-104 and I'm running into trouble figuring out how to set the bones and constraints. My current iteration looks like the following (left main strut): I currently have two HI IK Solver chains and two dummy controllers. The first IK solver chain links the upper pushrod mount to the bell crank ball socket and finally to the bell crank pivot point. The goal is the bell crank pivot point and it is linked to the strut dummy controller. The second chain is linked from the bell crank ball socket 2 to the wheel mount toe crank to the wheel mount pivot point with the goal at the pivot point linked to the strut dummy controller. I'd like the bell crank and the wheel mount to rotate about their pivot point. The bell crank and wheel mount are linked to the second bones in each chain. However, if I rotate the strut dummy controller, the bell crank rotates mostly correctly, but eventually twists and aligns with the upper push rod and the wheel mount rotates correctly initially, but then gets pulled out of place. I'm guessing that part of the problem is that the joints of the chain are not coplanar (or rather in real life the plane of the joints would rotate as the crank changes position) and that the HI IK solver expects the joints to be coplanar and the plane doesn't change. I've tried applying rotation constraints to a particular axis, but it doesn't seem to do much good. Here's a short video clip illustrating the problem ( ): Does anyone have any suggestions? Solution: I've managed to work out the solution to the problem and I am creating a 3 part video series to illustrate how. Here are the first videos of the series: Thanks, Red Beard

Thought I'd post an update. I had been working on animations and had nearly completed them just prior to releasing the SFM update. However, with the last few animations, it was clear that I was going to need a higher fidelity model and it didn't make sense to just do a part or two in higher fidelity, so I made a choice to throw the block model and animations out and pretty much start over. At this point, the model is getting close to completion, or at least far enough along to start some animation work. It's not going to be a high poly model of the current DCS standards we see, but it will be close enough to serve as both an illustrated example of the development process and something people wouldn't mind flying occasionally. At best, I'm guessing the poly count may get as high as 15k. Anyway, here's the model as it currently stands. Texture work won't be done until the animations, damage model, and LODs are done, so that's a ways off.

LOL! Thanks. I'll make the updates.

For anyone watching this thread, the SFM tuning chapter is now complete and may be found here: http://forums.eagle.ru/showthread.php?t=116901