F-103 Excalibur (Wing Commander)

[Resurrected Hobbyist]

So this is mainly a re-post of a thread I've had going for a bit over on WC News.com. I had gotten side tracked from working on it since... you know life and all that tends to come to a head. But working on the Marauder Bomber in a live stream has proven to be more motivational for me. So I'll be returning to this project in the not so distant future. So what project am I talking about? Well strap and get ready to launch!

Wing Commander 3: Heart of the Tiger is one of my favorite games of all times, and I absolutely loved piloting the Excalibur fighter in the game. So this is my effort to scratch-build a large scale (about 16" long) model of it. This is going to be a long term project. So this is going to be my project thread on here as I work through the project.

I did a paper-craft version of the F-109 Vampire recently, and that really rekindled an old interest I've had to have a model of the F-103 Excalibur.

Here is a gallery of images of the finished Paper-craft Vampire (will change link to forum hosted gallery once gallery is approved):
F-109 Vampire


I'll warn everyone now, this is going to take a while to complete. It will also be my first totally scratch built and self-designed model. The real challenge... I have no actual 3D design experience or systems. So I'm doing this all as I figure it out. Who wants to go on an adventure?

First thing I did was load up Wing Commander: Secret Opts and swap the in-game Excalibur with the Panther Fighter, that way I could control the fighter while I took screen caps of the fighter. So I took head-on shots of the craft from each of the axis angles. I then assembled those images into the collage you see here in Photoshop. I then used the full sized print out to gather measurements of the craft. here is the first batch of measurements.


This is the full size print out I assembled of the top-down shot of the Excalibur. Nothing particularly fancy here. Took the top down image, up-scaled it until it was the size I wanted (about 16" in length), and then printed it out. Cut the parts out and glued them together to get one sheet. From this I'm able to take measurements, which I record in the collage in photoshop.



I did use Pericles' paper-craft designs for the Vampire. The... I don't want to say the word but it's the only one I can think of, the 'problem' with Pericles' plans is they are quit literally the in-game model transferred to paper, which while it makes a game-accurate model, I'm shooting for something that is more universe accurate. The best example of what I mean on the Vampire can be seen in the nose:

If you look at the art work for the Vampire:


You can see it has a pair of twin-linked cannon barrel barrels (I want to say those would be the partial guns if memory serves based on placement).

But compare that to the nose of the in-game model:


You can see how the two barrels and assembly have been condensed down into what... well whatever that is. It's just something that didn't get translated from the artwork to the model or vice versa. Also the in-game vampire have a much leaner body thus appearing longer then in the art work.

For the Excalibur I want to go for display value. So I'll be interpreting and adding additional details, but I want to stay closer to the Wing Commander 3 version of the craft, but if I come to a point where I have a source of data from in-game compared to a source of data from say universe, unless I have an explicit reason to do other wise I'm going to take the source from in-universe as more accurate.

As I progress through the project I'm going to be posting a lot of information regarding how I go about things, how I got sizes, how I build parts, ect. I'm going to do this for a variety of reasons, and not all of them personal. As I said, I have no 3D design or architectural experience what so ever, I toyed with Miya for like 20 minutes once about 20 years ago, and have never had enough of an interest to go back into it. I've built.... the dark gods only know how many kit-models and I've converted hundreds more over the last 30 years, so I have a mind for "unfolding" things from their 3D construct to a 2d part. Still, there is a difference between knowing how to unfold an object and knowing how to build it.

So I will be posting pretty detailed progress updates on the project as I go both for a personal record so I can come back to it and see what I did, and to give people with more knowledge then I an opportunity to look at my process and provide feed back. And finally I'll post the detailed info as a basis for anyone else who wants to give it a go. I seek to inspire after all.

It was at about this point in my original thread on wcnews.com that a fellow Wing Commander fanatic chimed in. Whiplash has built several Balsa/Base wood models based on the Wing Commander series and had gone through a similar process designing the specs for his models as to what I was attempting to do. The big thing he pointed out was due to the limitations of the technology at the time the in-game models are very poor sources of accurate measurement information and he urged me to other assets to get my measurement information.

so I took Whiplashs suggestion, well the first part of it anyway since this is most certainly not going to be a quick process, and shifted gears from using the in-game models for size and measurements, and instead using alternate imagery, such as the Warbirds file info on the Excalibur. I also grabbed a metric crap ton of frames from the Kilrathi Saga intro and several in game videos from Wing Commander 3 showing the Excalibur. For the moment they are just sitting my library and due to the video shots they will be of little use for getting measurements. But as Whiplash suggested, they should be a good option for detail checks and the like. I've loaded the assorted collection of frames to my photobucket account, which you can view here. I've also gone back through my library of pictures (I've been collecting images from the internet ever since I discovered the "Save image as" command when I was about 12 years old) and collected several images I have of various artistic renderings of the Excalibur. Like the Screen Stills, I'm not sure how much measurement value these will have, but again they provide details that aren't as visible in the smaller images or the in-game shots. These images can be viewed here. if anyone identifies where the respective image is from, please drop me a line so I can add the info to the images in my photobucket. I do want to give credit to the artists who made the images.http://s14.photobucket.com/user/mav...03 Excalibur/Artistic Reference?sort=3&page=1

Beyond all of that I did start crunching some numbers for the model again based on the Warbird images. I've decided I'm going to scale back a bit for the project and rather then do a 16" long model, I'm going to do a 12" long model instead. I've done lots of small scratch build jobs before, but never a full blow from the ground up total scratch built model before. And I'm thinking about doing some lighting effects in the model as well. So the 12" version will be easier to manager and I think closer to my skill set at present. I want to challenge myself, not waste my energy and materials.

So, I took the warbirds image and pulled it into Photoshop where I upped the size by a factor of 6 so I have some room to work and make notes. The image is small enough that I will be forced to 'interpret' a lot of details, but that's what the reference library is for. With the image in Photoshop I started working out some numbers. I had to go way back to Algebra and Geometry to remember how to work with scales and conversions for this and I'm not totally sure my math is accurate. So if anyone who actually knows what they are feel see a mistake, please point it out.

Here is what I have:


Whiplash then provided some more information about scale calculations and such. A lot of it went over my head as to the practical use of it due to the size of the image itself. The printed image is 3/4" wide after all. But it did spark a thought in my mind.
I'll explain:

 

[Resurrected Hobbyist]

[Had to split post due to length]

See in digital graphics you have 2 types of graphical engines each with their benefits and minuses to design. The first is known Raster graphics, or sometimes referred to has bit-map pictures. This format uses individual colors dots to render an image. The more dots you have in a given space the sharper the image can be. A line is simply a series of dots of comparable color. This format has the advantage of being capable of storing far more graphical information and being the foundation most directly based in traditional art work so many of the concepts from that medium translate to this format. The second type of graphic is known as Vector graphics. These use mathematics to form shapes rather then recording each individual position of every point contained in the image. This has the advantage of being vastly easier for a computer to deal with but doesn't translate from math to art very well.

Why am I bringing this up? Well, the two graphic types have another consideration between them, how well they scale up or down. See a bit-map can only be scaled by either adding or removing details. How the program add details will effect how well the image scales up, but generally you can only scale a bit-map image up about 50-60% before you start seeing a noticeable loss in sharpness as additional quasi-random extra material is added.

As an example, look at the corner on the right side of the Excalibur image where the forward section of the fuselage meets the side of the wing and side mounts. On the 1 1/2" sized image it's moderately decent.

But enlarge it about 400% (so that the over all image would be about 4.5" long for reference) and suddenly it stops looking anywhere near as sharp.

Vector Graphics however, because they are mathematically based can scale infinity. The program simply remaps the location and calculates what's between the points as needed.



So why did I just explain all of that? Well Whip gave me the idea to start with a series of vector graphics, based on the Warbirds images, then upscale those vectors to the size I need them. Since the vectors expand infinitely, there's no lose of detail in the process. I started with just tracing the major sections visible on the Warbirds image. I did some digging and found my hard copy of the Wing Commander 3 stuff and scanned the warbirds images at an obscenely high resolution of 2700 dpi then pulled that scan into photoshop (as an aside, it's always a challenge working with 5"x2" image that's a staggering 400megs in size) and went to work with the pen tool to generate the vector graphics. the high resolution pulled out some details, but the size of the original image simply doesn't have a lot of detail to start with.

Remember how I said enlarging a bit-map image is done by adding material? Well scanning a printed image at such a resolution results in basically the same thing. End result? I could trace out the body, the engines, and the cockpit. The rest... eh I kind of had to take some artistic license with those.

Remember that library of images I assembled of the Excalibur? Well I used that a lot in this process. I primarily referenced the "armada" rendering of the Excalibur for filling in the details that were lost in the up-scaling of the scan.

So after a couple hours worth of work, I assembled the following top-down image of the Excalibur with

(website friendly version displayed. If you want to grab the 300 DPI version it's linked here)


I mentioned that these are first passes, why is that? Well... it's because they don't match up exactly with each other. It's too late for me to go into all details but as an example. The widest part of the front boom where the cockpit is, it is widest on the front view, narrowest on the top view, and just slightly narrower in the bottom view.

So what does this mean? Well, it means I have to go back, pair up the line art pieces with each other and decide which part of which version I want to take as the 'correct' version. While the web version of the image above sizes out to be about 4" wide, I want to build a 12" model from these designs. So if I scale everything up by a factor of 3, and I have one part from the bottom that is 1mm wider, then the same part from the top view, but the rear version is 1mm to the left, I'll wind up with a part that started out as a rectangular cube but when built comes out as a lop-sided rhombus. So guess what I get to do over the next few days....

 

[Resurrected Hobbyist]

Okay, so I know its been... well more then a month since I last updated. I work retail, retail service at that. So we're gearing up for the ever so fun Christmas season, OH and Corporate decided that the middle of October was the perfect time to roll out a new operational model. So yeah, that's been a lot of fun.


I don't have a whole lot to show this time. Because of the nature of this project, it's going to have several phases of intense work for little immediate result. This is one of those phases. Where I get to go through the line images I created, and begin extracting measurements. And measurements. and measurements. So far I have 25 measurements and I'm only... eh.... 80% finished. With the back view. I still have the front, side, top and bottom to do. And the top and bottom are going to be the big data-dumps! For reference, understand that the entire space the rear view takes up is 87.5mm high by 182mm wide while by comparison the top view occupies a space of 170mm wide by 305mm long. And the top and bottom views have a lot more detail then the rear. Still, gotta get started somewhere right? So what do I have to show today?

Well, this:


Lots of lines and measurements, still have lots of lines and measurements to go.

 

[Resurrected Hobbyist]

so a so-so update today. I haven't finished taking measurements yet, and I'm still working on that front. What I do have is a partial card board mock up of the forward fuselage. This is little more a simple mock up with the basic measurements I've got already. Not complete. This is meant to be an exploration of the construction of the finished model. As one example, I've already encountered one mistake. When I figuring out the parts based on the measurements from the line drawings, I thought the triangular section that would make up the side of the cockpit section was a right triangle. Well, it's not. So the original part I made based on the measurements didn't fit in the space.




And this was the point where I encountered an incident and was grateful I had been working in cardboard rather then plastic.

Well...... ssssssshhhhhhhhhhhhhhhhhhhhhiiiiiiiiiiiiiiiiiiiiiiiiiitttttttttt! I have to start over on the mock up. I was gluing the bottom panel of the body in place. What I didn't realize was some glue had leaked out of the body and onto my work space. So when it dried it dried to my work space. I went to pick up the body, and in the process tore it apart.

Well this is why I started with cardboard. So I could make these sorts of screw ups and not have to pay through the nose for it.

Still, this sucks.

 

[Resurrected Hobbyist]

and guess who's back with a new update just in time for the holiday!

As I said last time, I'd had an accident while working on the Excalibur card board mock up which resulted in a destroyed model. It was a bit of a bitter-sweet kind of screw up since on the one hand, I had to start over on the mock up. But on the other hand I specifically chose to work in cardboard for this part of the project specifically so that if I had a screw up, it wasn't going to cost me $20 in materials. So... I guess it is like planning for a problem and having the problem occur. Yeah you're prepared and ready for it, but you still had a problem. So.. yeah.

Anyway I have been working on rebuilding the mock up and been making good progress. Having built it once already I can zip through a lot of it while also correcting some mistakes I made the first time around

Here you can see I'm built the mark 2 model up to the level of the previous version at the time it was destroyed. The main body is actually stronger then the first version and has cleaner joins between the various parts.

So that's it? I got back to where I was last week and now I'm done with this post?

Yeah, not so much. I have started working on the next sections. Firstly there's the cannon mounting under-carriage, the part that holds the reaper cannons in Wing Commander 3.


You ever start to work on something and think it'll be super-easy and you can just zip right through it, but once you start working on it, you find out it's actually way more difficult then you thought it'd be? Well, that was this part in a nutshell. Cut 2 side parts so they line up with the under side of the body and the forward boom, what's so hard about that? Simple, I have no formal 3D design experience or training. So because of that I had to re-cut the sides a couple times after I found I had not measured properly.


For detail and variation I took a pencil and darkened the inner spaces of the under carriage where the reaper cannons would be placed.

And then, there is the power-plants... er missile bays.... er... whatever the boxy shapes on the sides of the main body are. This proved to be more challenging then it first appears.

The initial shape was easy enough to accomplish, measure and cut the upper and lower sections as they appears in the line art. I had to do some interpretation to make the side and inner-section. There's little enough I can say about the sides that would really help any aspiring crafters. I had to estimate my measurements, cut a piece and test it. Find out I was wrong, and try to correct it with a new version.

No the real challenge was the scoop intake covers. See, it's an angle that extends in 2 dimensions, backwards relative to the body, and outward relative of the body. It's something that isn't immediately apparent from the 2D line art. So I had to cut and fit the ram scoop cover about 6 times before I got a sizing that fits.


And this is where I encountered my latest little... challenge to this project: I lost the gen 2 mock up. I put it in a storage box with all the other parts and pieces I had cut, sealed it... and now the storage box is gone. So... yeah. Have to start over on the mock up... again!

 

[zathros]

When I design a model in Rhino 3D 5.0, I do not use a scale. Every surface is a part, and I mean every feature.. I then color them in. This allows for easy modifications, and if the parts are grouped, they unfold with the lettering, graphics where they should be. I can use "Parametric scaling" to scale the models to whatever I need too and don't lose image. quality. Of course, once you go to an image, I don't like .pdf's for models, you can make a huge image, and let the user scale it down (Canvas size) in XNView, one of my favorite photo editing programs, and quality is no longer an issue.

Wing Commander had some really cool ships in it.

 

[Resurrected Hobbyist]

When I design a model in Rhino 3D 5.0, I do not use a scale. Every surface is a part, and I mean every feature.. I then color them in. This allows for easy modifications, and if the parts are grouped, they unfold with the lettering, graphics where they should be. I can use "Parametric scaling" to scale the models to whatever I need too and don't lose image. quality. Of course, once you go to an image, I don't like .pdf's for models, you can make a huge image, and let the user scale it down (Canvas size) in XNView, one of my favorite photo editing programs, and quality is no longer an issue.

Wing Commander had some really cool ships in it.

... I have basically no idea what you just said, Zathros. I mean save for the "I don't like .pdfs for models" I understood that part. The rest of it... I'm assuming Rhino 3D is some sort of 3D Construction program. Never actually heard about, and have basically no training in 3D design of any form. And parametric scaling... um... I got nothing, dude. Like I know what scaling is. But the parametric part... I have no idea what that is.

 

[zathros]

Parametric scaling is grouping the wholoe model, taking a background image that you know has the dimension you want. You can the select the part of the model, slide it over, or make a circle or boxx that size, and do a #D scaling of the model to the part, every part converts then to that scale. When I model in Rhino, I make sure every single part unfolds before I move on to the next one.

Some of my CAD models. Every single one unfolds to a flat part, and I have made them at least once. Administrating this forum is my contribution, I don't publish my models, but will help anybody with theirs, and help then learn Rhino if they really mean it, I post my work for inspiration, and to sow that I do know what I am talking about:

 

[Resurrected Hobbyist]

Parametric scaling is grouping the wholoe model, taking a background image that you know has the dimension you want. You can the select the part of the model, slide it over, or make a circle or boxx that size, and do a #D scaling of the model to the part, every part converts then to that scale. When I model in Rhino, I make sure every single part unfolds before I move on to the next one.

Some of my CAD models. Every single one unfolds to a flat part, and I have made them at least once. Administrating this forum is my contribution, I don't publish my models, but will help anybody with theirs, and help then learn Rhino if they really mean it, I post my work for inspiration, and to sow that I do know what I am talking about:

Oh I have no doubt you know what you're talking about, I am somewhat unclear as to what you are bringing up the point of a 3D Design program though. Like I've said elsewhere, I presently work retail as my main job and we're heading into the christmas season. Add to that my own classes I am taking already and my other jobs. If you are offering to teach me Rhino it would have to wait till probably mid-march at the earliest. If you're bringing up Rhino for something else... um what are you bringing it up for?

 

[zathros]

Oh I have no doubt you know what you're talking about, I am somewhat unclear as to what you are bringing up the point of a 3D Design program though. Like I've said elsewhere, I presently work retail as my main job and we're heading into the christmas season. Add to that my own classes I am taking already and my other jobs. If you are offering to teach me Rhino it would have to wait till probably mid-march at the earliest. If you're bringing up Rhino for something else... um what are you bringing it up for?

Just as a way of making a model without having to cut anything. Rhino "Unfolding" abilities are quite amazing, and it is an intuitive program. It costs $1000 dollars, but if you have a kid in school (or you're in school), you can get it for $200 dollars, and it can be installed on 3 computers. You ay be in a position to get an incredible program rather inexpensively, relatively speaking. The tiles on the Space Shuttle were modeled using Rhino3D, for it's Non Uniform Rational B-Splines form of modeling, (N.U.R.B.S.) . This is the full commercial version too, not some scaled back program. Just throwing thoughts out there. I like the scratch built method, of course, it is only that that method makes models hard to reproduce, but is great for one offs, and if you are painting the model, filler and putty allow for incredible work.

Ref: https://en.wikipedia.org/wiki/Non-uniform_rational_B-spline



The line below could be manipulated from the top view also, and then Extruded into a surface, and promptly unfolded:


With this model design (A90 Orlyonok) I used this technique to it's fullest. I have a friend who designs models professionally in Europe, and he told me he could never get the front right, so he gave up on the model, he uses the same software. His exclamation that "I got it" was quite a validation. I gave him the file, and told him to do what he wished with it. He's one of few people I exchange build files with.

A90:

 

[Resurrected Hobbyist]

And guess what project I'm back working on?


Here is the version 3.0 mock up as it stands at the completion of my live stream tonight. Starting from 0 and ending with this after only 6 hours... I'll call that a win.


Using foam board for the main body helped to speed things up a lot.


Previously the mock ups had had sharp apexes at the sides of the cockpit. With foam board in there I could found them out a bit more like how the in-universe images appear. The points will be there after I build the cockpit simply because it's kind of difficult to round out cardboard. Regardless when I get to the styrene version, I'll be able to round out the corners.


I'm recording a lot of measurements directly on the mock up once again.

 

[zathros]

This is going to be a very strong model!