Trident Redux

[technoid]

No problem. I'm a bit surprised that there isn't an automated routine to do that since in my mind it's sorta like rendering it. It seems like there ought to be a way to use the baked on colors to let the computer do at least a rough draft of it. Since I've never done it, I guess I still don't understand exactly what you need to do. But since I've seen several of you complain about doing it, I do understand that its gotta be a pain. And I've still got the betas to play with.

I know what you mean. I need to change the cartridges in my reverse osmosis water system & change the batteries in its tester, but I've been putting it off for a similar reason. The water still tastes great & doesn't gunk up the coffee maker or my CPAP machine's humidifier, so I've been putting it off.

It just depends on what you want. If you want other people to be able to change the colors you can't use baked on colors to paint them. And I was requested to be able to paint the motor so I'm mapping it too. So in short, if you want others to be able to paint something you can't use baked on colors for it because they're built into the 3d model and can't be changed. After a while mapping isn't that hard it's just boring and no fun. The fun is creating the model and mapping is just work. And like they say, all work and no play. So I've been playing one of my games so I'm having fun.

 

[Bill Stuntz]

Hey, I'm not trying to teach my grandmother to suck eggs... and I'm not complaining. You deserve to have fun! All work & no play... and all that.

 

[technoid]

Hey, I'm not trying to teach my grandmother to suck eggs... and I'm not complaining. You deserve to have fun! All work & no play... and all that.

No biggie I didn't think you were complaining I just explained why baked on colors won't help with the mapping. But yeah it would be nice if they created a built-in tool that would map the parts for you. And I guess while I'm here I might as well say I just finished mapping the wings. But since you don't really know the process it's kind of a four stage thing.

1. You map all the individual parts but at that stage each part is the full size of the grid you use to create TGA file for the color scheme.

2. After you map all the individual parts you have to go back and resize each of the sub parts to fit the main part. By that I mean the ailerons are the full width of the wing because every part you map is created the same width. So the wing and hstab are the same width. So after all the individual parts are mapped you need to go back and resize each of the sub parts to fit properly to the main part. (like the wing, ailerons, and any other part you mapped for the main part). An example of that is the ends of each aileron and the wing cutouts where the ailerons fit. All of those sub-parts must be resized to fit the main part.

3. Then after all the parts are mapped (1), and the sub parts are resized to fit the main part (2), then you have to resize and position all those parts to fit the grid where the CS TGA file is created from.

4. Then you have to do what's called 1x1 mapping which is fine tuning the location of the parts so when you add a stripe that extends from the wing to the aileron the stripe aligns properly as it crosses from the wing to the aileron.

So there's lots of stuff to do during the mapping process. Which is why most people say they don't like it.

 

[Bill Stuntz]

I understood that it's complicated. And I suppose that's why I'm surprised that there isn't an automated way to do it. The whole resize thing surprises me. It seems like it should "just work" since they already fit in the original design. It seems like the software makes it much more difficult than it seems (to me) like it needs to be. It sounds like the people who designed the software have never needed to actually DO it. Just like engineers never have to actually repair anything they designed. That's one of my pet peeves. IMHO, programmers/engineers should be required to actually use/repair anything they'll design before they're allowed to participate in the process.

As an example, back in the dark ages I worked with a 6ft H * 6ft W * 3ft D pulmonary function machine that needed to be calibrated each day before use. The adjustment pots were in the back, the displays in the front. It was IMPOSSIBLE for a single person to calibrate the machine without taking at least 20 trips around the machine. Tweak a pot, go see if the number's right. No? Do it again... for each parameter. The pots were 3ft deep in the machine, right up against the front panel. Turn the pots around, provide access holes in the panel to tweak from the front, & the calibration could have been done in about 5 minutes instead of 1/2 hour. PITA!!!!!!! If the engineer had been forced to do it at least once, it would have cut many miles off my uncomfortable Cordovan uniform shoes over the 3+ years I worked in that lab.

 

[technoid]

I guess I'll go ahead and post something. There's really not much to say.. duh.. I mapped something. Which is why I didn't think I'd do any updates while I'm mapping but it gives me something to do when I stop. Like I said work is slow because I have to force myself to work on it but it's getting there. I have the fuselage, wings, and hstab mapped but there's still lots to do. I start getting excited to be able to work on the release color scheme but that's still a long way off.

 

[space boy]

I'm surprised that there isn't an automated way to do it.

That was my initial feeling too. When I watched a video or two about it (UV unwrapping), I began to understand why there's so much manual effort involved.

I'm sure you're familiar with different globe projections...

(I'm still really new to UV wrapping & mapping, so I might have the details skewed.)
Some 3D Software DOES have the ability to automatically unwrap your 3D model onto a 2D surface. This works really well for simple objects like cubes or spheres.

But as you can see in the globe example above, going from the 3D sphere to a 2D image could be done in at least 4 different ways. There are plenty of other globe projections, as I'm sure you know. Some are best for more accurate distance, some for more accurate depiction of area, and some for the intuitive representation of direction.

What if the 3D object to unwrap isn't a regular solid? What if it's a cow, or a... human, or an... AIRPLANE!

At the least, each piece needs to be unwrapped separately, and the way the software unwraps it may not give an intuitive result, because there are so many unique ways that a 3D object can be covered by a 2D surface.

I'm typing this out for my own understanding as well as anybody else's. If you more-experienced unwrappers have input, PLEASE give it.

 

[Bill Stuntz]

That was my initial feeling too. When I watched a video or two about it (UV unwrapping), I began to understand why there's so much manual effort involved.

Like I said, I don't really understand, so that really clarifies things for me. Wouldn't top & bottom plane projections work? It would miss the engine, the wheel pants, and any internal components that are obscured by other bits & pieces, but wouldn't it be a good start? Or am I missing something?

I start getting excited to be able to work on the release color scheme but that's still a long way off.

I'm curious - surprise, surprise! I'm SURE that comes as a surprise.
Will the release CS be similar to the profile Trident or do you have something else in mind? I'm sure you know how important appearance points are in C/L Aerobatics judging. Are you planning something REALLY fancy? I suspect my CS is too simple to get many appearance points - especially since it's just simple shapes done with plain Monokote & a little black dope.

 

[space boy]

Wouldn't top & bottom plane projections work?

I can tell you already know how limited that would be, but it would be a really basic start, wouldn't it?
blind leading the blind here.

 

[technoid]

Like I said, I don't really understand, so that really clarifies things for me. Wouldn't top & bottom plane projections work? It would miss the engine, the wheel pants, and any internal components that are obscured by other bits & pieces, but wouldn't it be a good start? Or am I missing something?

I'm curious - surprise, surprise! I'm SURE that comes as a surprise.
Will the release CS be similar to the profile Trident or do you have something else in mind? I'm sure you know how important appearance points are in C/L Aerobatics judging. Are you planning something REALLY fancy? I suspect my CS is too simple to get many appearance points - especially since it's just simple shapes done with plain Monokote & a little black dope.

I have an idea of what I'm going to do and it's very close to the profile version so it probably won't win any prizes. But most of the time I like pretty simple color schemes.

I just finished mapping the vstab, spinner, and canopy so the main part of the plane is mapped. Now I need to do the landing gear and engine.

I was looking at the plane with no landing gear and to me it looks great, but retracts aren't reasonable for this plane. Too bad.

 

[technoid]

I can tell you already know how limited that would be, but it would be a really basic start, wouldn't it?
blind leading the blind here.

One thing I'd really like in mapping is if you could "tag" a part, like the fuselage, and then when you mapped each part (wings, hstab, vstab) it would keep the relative size of those parts instead of mapping all of them to the maximum width of the grid. That way when you finish all the individual parts you wouldn't have to go back and set their relative sizes. But I'm used to it after all this time so I just use what I know how to do. Which like you said, isn't much compared to what the tool can really do. I just learned the basics of how to do it and just use that.

 

[Bill Stuntz]

I was looking at the plane with no landing gear and to me it looks great, but retracts aren't reasonable for this plane. Too bad.

Yeah, I thought about that, too. Too bad the wheel pants are so important to the design. They really add a lot to the total effect by mimicing the rudders. And AFAIK, retracts didn't even exist in C/L Aerobatics back then. I'm not sure they're even done now because extra lines add more drag on the inboard wing than the decreased drag from "missing" LG might. I suspect that they'd be counterproductive. Maybe electric retracts would work, but I'm not sure that would be allowed in the rules. And the added weight would certainly affect performance.

 

[Bill Stuntz]

I can tell you already know how limited that would be, but it would be a really basic start, wouldn't it?
blind leading the blind here.

I hadn't thought about the planar projections until you pointed out the different possibilities & problems. Conical seems to be what happens when rendering, so that's what I think I was envisioning since that's already being done in the sim. Different problems.

 

[technoid]

Yeah, I thought about that, too. Too bad the wheel pants are so important to the design. They really add a lot to the total effect by mimicing the rudders. And AFAIK, retracts didn't even exist in C/L Aerobatics back then. I'm not sure they're even done now because extra lines add more drag on the inboard wing than the decreased drag from "missing" LG might. I suspect that they'd be counterproductive. Maybe electric retracts would work, but I'm not sure that would be allowed in the rules. And the added weight would certainly affect performance.

To me this is an RC plane so the CL issues don't come into play anymore. But the wheel pants are definitely important so that does matter.

 

[Bill Stuntz]

To me this is an RC plane so the CL issues don't come into play anymore. But the wheel pants are definitely important so that does matter.

I agree completely. But I thought that stuff was worth mentioning even though it doesn't directly apply to the RC conversion, because it was certainly a factor in Larry's original design. He couldn't do retracts in creating a competitive design, which effectively forced the wheel pants to be so important in the design. It seems to me like wheel pants are usually an after-thought that's only important aerodynamically rather than really emphasizing the appearance/style of the plane. IMHO, most wheel pants are boring and pretty much the same... which these are NOT.

 

[space boy]

I hadn't thought about the planar projections until you pointed out the different possibilities & problems. Conical seems to be what happens when rendering, so that's what I think I was envisioning since that's already being done in the sim. Different problems.

Oh RealFlight definitely doesn't use a conical projection. It uses the UV data from the model to map a 2D image onto the 3D model. UV refers to the equivalent of X&Y coordinates. Each vertex in the model is assigned a UV coordinate to use from the 2D image.

Here is a really good example of how automatic unwrapping can be unintuitive, if not useless, and why doing it manually often makes more sense, even though it's tedious:

from this page.

 

[Bill Stuntz]

WOW. I haven't been considering how important different (isometric?) views would be, either. I've been thinking in terms of the obvious (cardinal direction?) views that would completely ignore/conceal the edges of the slats at the sides/centers of that box that are perpendicular to the inner panels. It's become obvious that my conception of the problem has been grossly oversimplified. You guys teach me new stuff every day! I suppose that if I'd ever actually done it I'd have recognized the problem. My mind games often work, but don't go deep enough.

 

[technoid]

All my mapping is very simple it's done from either the x, y, or z perspective so it's flat. But it's interesting to see how they mapped the crate I just haven't learned how to do that. I have unwrapped a thing or two like the outside of an electric motor for the label going around it, and I think a tire once.

 

[technoid]

I just finished mapping all the plane parts now I need to map the engine. But now it's break time.

 

[technoid]

I swear my brain has taken a permanent vacation. I was going to start mapping the engine today but noticed one of the fins on the cylinder was missing. So I went back and looked at some of the old versions and it was there so it got deleted at some point. (duh) Then while I was fixing that I noticed I hadn't made a collision mesh for the muffler. (another duh) So I'll fix that after the mapping is finished. I normally hide the engine while I'm working on the plane so I never noticed I missed the muffler collision mesh. SO.. things are moving along but I decided to play my game since I've already worked enough for this morning finding and fixing the missing fin and collision mesh. At the end of a project it seems it goes on forever and won't end. HA!

 

[technoid]

I just finished mapping the engine so the first pass of mapping is finished. Now I need to resize all the separate parts and get them ready to put on the grid. But now it's time to go grocery shopping. But it's getting there.