Aircraft Painting

In arranging pieces to be textured, keep in mind that sometimes two parts can share the same locations:
The engine only has the forward facing side textured. The control panel only has the aft facing side textured. As such, there are no issues with putting them in the same location.

- Ivan.
 
I've just managed to get some air time on the local library PC, sometimes the local college overwhelms the system that the council use here. This is absolutly fasinating and will take many moons of study to get right but I shall persevere.
Thanks Ivan and thank Joe as well for me, he's doing a cracking job.
 
Hi Womble55,
Hope what I am demonstrating is making sense to you. I have actually gotten quite a lot further on the Demon than is shown in the screenshots. I just didn't see the point of showing the same basic technique 10 more times when there is nothing new to demonstrate. There will be more. I just haven't gotten to those parts yet.

- Ivan.
 
As stated in the previous message, I have actually gotten much further since the last installment. Tonight there will only be observations of a problem in texturing and not the solution (which is actually a fairly obvious one).

When texturing wings, I typically like to have the texture files be a fairly close mirror of each other. Often the insignia and camouflage will be different, but for the most part the panel lines and general outline should be identical or nearly so.

Such is the case with the CW-21B Demon. I first adjusted the wing part overlay onto the Left Wing Texture file (I like to start with the Left Side when possible) and just mirrored the texture file to the Right Wing's file.

As you can see from the normal CFS external Top view (WingTop1.jpg), the visible colour bands are not the same between the left and right sides. The specifications in AF99 are mirrors as are the textures, but the smoothing algorithm skews the textures almost one full pixel to the right.

As you can see from the exit screen (WingTop2.jpg) the textures are lined up pretty well when the smoothing algorithm isn't being used.

(To be continued....)
 
Folks here probably have already figured out what the next part is, but here goes anyway:

As you can see from the area that Joe is pointing out in AF5Paint in the first photograph, the coloured bars actually start at red bar that is aligned with the edge of the texture. In the earlier screenshots, we can't see that first red bar on either side. That tells us that the texture at the wing root has room to use. At worst, we will encroach into the single pixel red BORDER around the wing texture.

The dimensions from AF5Paint are directly used for the Right Wing as can be seen from the second photograph.
 
The left wing texture was moved outward a little to try to align the same colour bar with the edge of the wing component. The scale used for the wing is 15.00 feet over 256 pixels or 0.05859375 feet per pixel. In general, I have found that an offset somewhere between 1/2 pixel and 1 pixel usually work.

If textures for both sides are being moved, each should be moved about 1/2 pixel's distance. If only one side is being moved, it generally has to be moved 1 full pixel. Round down for fractions. In this case, I would have expected the left wing offset would be 0.05 feet, but that didn't seem to work. 0.06 feet was what I eventually used as you can see in the screenshot.
 
This is the final result from the slight offset put into the texturing of the left wing. I believe that the offset is slightly too much and that the green bar is slightly wider on the left wing, but 0.05 definitely was not enough. With pixels that cover so much area, often it is a matter of best rather than perfect alignment.

- Ivan.
 
while replies and comments may be sparse,
we are paying attention.
many thanks for taking the time
to pass along this information.
 
Hi Folks,
The trick here is trying to pick topics that are not entirely obvious. Glad it's doing some folks some good.

BTW, the Joes all want to get into the show, but my rule is that only custom Joes are allowed screen time. Pity the lighting needs to be dark so the faces really don't show up.

Thanks.
- Ivan.
 
I don't think that individual introductions
would be out of order.
give all the Joes their due.
 
Ivan said:
Hi Womble55,

Get yourself a copy of AF5Paint. After you try it out and still have questions, we can discuss it further. The software is free and resolves most of the issues you are likely to rin into.

- Ivan.
Click to expand...

Is this all I need to get started on repainting aircraft or I need more.... I have been keeping track of it but I dont know if I need more program...

Sorry if i tend to be off topic... :bump:
 
Hi Chacha,
Actually what we are discussing here is not REpainting aircraft. THAT is much easier though if you run into a problem where the designer goofed, you really can't fix things very well. A case in point is a Black P-38 Lightning I have seen. As long as you keep the general paint job black, you won't notice that the wings or fuselage share textures with the tires. If you paint the aircraft Green, you get green tires.

The topic of discussion is actually the INITIAL painting of an aircraft. THis is for the designer of the aircraft who needs to lay out the textures onto a bare model.

AF5Paint is pretty much useless for a repaint. To do a repaint, you will generally need a utility like CFShow or something else to translate a R8 format file into a BMP or something similar and perhaps MSPaint or GIMP or whatever to actually edit the file.

Repainting has been covered in many places, but I believe initial painting hasn't which is why I am doing it here.

- Ivan.
 
Continue on Man,

Interesting to know how these planes are done from by the designers lay out to painting and to our enjoyment and fun!

Thanks, Ivan :applause:
 
The next item to be textured are the Flaps. Note that when deployed, the flaps are neither horizontal nor vertical. They are somewhere in between with a typical angle around 45 to 60 degrees.

The idea here is to map the texture file in such a way that a symbol drawn on it will appear the same on the surface of the part in a way that doesn't require a stretch to look right.

The same idea applies to control panels: The texture should be layed out in a manner that if someone draws a round gauge, it LOOKS round on the panel in the aircraft.

In this picture, Joe (I really call him Jurgen the Hitler Jugend) is showing the coordinates of the Left Flap. Unfortunately in this case, the flap isn't quite planar. The inside and outside edges don't align with each other. I didn't notice this in building the model, but don't be surprised if I fix it later. At this point, we will just continue since the concept of what we are doing isn't affected.

Remember from the Pythagorean theorem that the hypotenuse (chord of the flap) can be calculated by H^2 = X^2 + Y^2.

The number we need is the ratio of X / H because we are using a top/bottom texture. If we were texturing fore/aft, we would need the ratio Y / H.

Because the angles are not quite the same, we get 0.549 and 0.619 for the two values of Y / H.
 
The flaps will share the same scaling as the wing and the same texture file.

The wing texture file is 15.00 feet over 256 pixels which means each half of the texture covers 7.50 x 15.00 feet. If we multiply by the X / H ratios, we get values of 4.116 or 4.643 x 15.00 feet.

Since we can only use one value, an average of the two 4.379 or 4.4 is good enough.

The first picture shows the flap projected onto the texture file at 7.5 feet.
The second picture shows the flap projected onto the texture file at 4.4 feet.

Note the location of the yellow pixel near where Joe is pointing.
 
On the next picture, Joe is showing the values used for the starboard flap.

On the following picture, Joe is showing the values used for the port flap. Note that the offset used here is the same as the one reached via experimentation with the wing: 0.06 feet. Since the scales are the same, this should work as well here as it did on the wing.
 
These last two pictures show the final result. Note in the first picture that the flap inner edges look very similar but there is a VERY slight light green line on the left flap. That is because as mentioned before, the offsets are close, but don't quite line up. I believe it is as good as we can get though.

On the second picture, various circles are painted on the forward / lower surface of the flap. They look slightly long up/down fore/aft, but I believe that if they were on the outboard part of the flaps they would look slightly short since the scaling ratios are not quite the same.

Good Evening.
- Ivan.
 
Hello All,
This presentation is done slightly out of sequence. The activities were done quite some time ago, but not recorded.

With the untextured aircraft, the different pieces were quite visible because of their contrasting colours and also because the pieces as components were "Sharp" instead of "Smooth".

When a gray texture is applied, all of a sudden the aircraft looks like a two-dimensional gray blob. Two areas where this is very visible are the canopy frame and the underside of the cowl.

In the first picture, Joe points out the canopy frame. In the second picture, Joe points out the underside of the cowl.
 
One way to address this issue is to exaggerate the edges of the parts a bit and create panel lines. A single pixel here is overly large, so as Joe is pointing out, the border of a part is shaded so that a fraction of a pixel shows past the part.

In the first picture, Joe points out the shaded area that is put on the edge of the texture of the Cowl Bottom.

In the second picture, Joe points out the shaded area that is on the fuselage under the Canopy Frame. Note that the canopy part is used as an overlay to figure out how much of the shaded area is visible.

In the third picture, Joe points out where the Canopy Frame covers the aft part of the fuselage. Note that the Canopy Frame is used again as an overlay part.
 
You must log in or register to reply here.
Back
Top