------------------ "When You're Up to Your Ass in Alligators, Today Is the First Day of the Rest of Your Life." -- Management slogan, Ridcully-style (Terry Pratchett, The Last Continent, Discworld) -------------------------------------------------------------------------------- Prakesh's Star Trek Site
posted
Looks good, but of course I have a few suggestions.
1. Try to keep the light source consistent. The shading on the primary hull suggests that the light is coming from straight ahead. However, the shading on the secondary hull suggest the light's coming from the top. I usually have the light coming from the top since that's how we humans are used to seeing things. If the light is coming from the top, the nacelle supports should be mostly in light, with maybe just the extreme outer edge in shadow rather than being half in shadow. For extra credit add a shadow cast by the nacelle supports on the secondary hull. 2. The top view seems a bit off. The secondary hull looks like the top view of a rectangle, particularly how it intersects with the primary hull. The highest point of the secondary hull, and therefore the point that should be furthest forward (to the right) in the top view, should be along the midline, not along the top (left) and bottom (right) edges. So on the top view the forward edge of the secondary hull should be a convex (outward) curve, not a concave (inward) curve. This curve should also be fixed on the side view. When I have had trouble visualizing how solids, especially curves, intersect, I actually take a ball and draw on it. You might also try making a little model in clay. The nacelles supports are also too close together on the top view. The rear view shows that the attachment points should be further apart (on the top view) and lower (on the side view). I think the position of the panel lines is off, which caused you to position the nacelle supports incorrectly. Finally, the outer gray ring on the top view seems a bit too thick. 3. The lettering secondary hull lettering on the top view should be foreshortened or squashed instead of just having the bottoms cut off. Also, if you're really good you might want to bend the letters on the primary hull side view to follow the curve of the hull.
------------------ When you're in the Sol system, come visit the Starfleet Museum
posted
Apart from the above criticism, it looks pretty good, especially if that's your first attempt at gradient filling. I usually try to make it appear as if the light is coming from above the viewer's right shoulder. The paint program I use has a "sunburst" gradient tool that does that pretty well. I put the bright spot at 30% from the top and 70% from the left. The linear fill tool I use isn't as versatile, and thus I'm still trying to figure out how to get the brightest part to fall in the right place.
Masao: Any tips on how to manage the intersection of different shapes of solids?
posted
What you need to do is identify the positions of landmarks on different views and connect them. It's hard to explain with words, so I'll try to work on a little diagram and post it later.
------------------ When you're in the Sol system, come visit the Starfleet Museum
posted
Aahh, here's some criticism I can actually do something with. Thanks!
About the curvings on the top view: OOPS, just a little error with envisioning the thing in 3D.
I started with the side view, and continued with the other views. I got a bit tired, so the latest pic (top) is likely to have the most errors. That also explains the lines on the top view. I couldn't (at the moment) think of an easy way to do it, so it's a little bit improvised.
And adding to that, CorelDRAW isn't that intuitive to use. Finding out how do to specific things wich you feel can go easier, takes quite a long time. Baloo, what vector-software do you use?
------------------ "When You're Up to Your Ass in Alligators, Today Is the First Day of the Rest of Your Life." -- Management slogan, Ridcully-style (Terry Pratchett, The Last Continent, Discworld) -------------------------------------------------------------------------------- Prakesh's Star Trek Site
posted
Vector makes use of mathematical shapes that are scalable. No matter how much you zoom, it will remain sharp.
Bitmaps use pixels, like MS-Paint. If you zoom in, you'll see large pixels! These kind of pics generally take more space than vectors, since it has to remember individual pixels, while a vector just remembers some kind of maths.
There's a third one, meta. I actually don't quite know what this is. IIRC, some kind of mix between vectors and bitmaps.
I think both PSP and Paint are bitmaps. I'm not sure about PSP. Inever used it.
------------------ "When You're Up to Your Ass in Alligators, Today Is the First Day of the Rest of Your Life." -- Management slogan, Ridcully-style (Terry Pratchett, The Last Continent, Discworld) -------------------------------------------------------------------------------- Prakesh's Star Trek Site
posted
In response to Baloo's request, I have some diagrams (3 at A4 size) and explanations about how I use cross sections to draw intersections between solids. I don't have a website. Would any one be interested in posting them for me?
------------------ When you're in the Sol system, come visit the Starfleet Museum
posted
You can mail 'em to me ([email protected]) and I will put them on my site.
------------------ "When You're Up to Your Ass in Alligators, Today Is the First Day of the Rest of Your Life." -- Management slogan, Ridcully-style (Terry Pratchett, The Last Continent, Discworld) -------------------------------------------------------------------------------- Prakesh's Star Trek Site
posted
I decided to redraw Prakesh's ship with vectors to show how I draw. I obviously didn't prepare a complete drawing; I only roughed in the basic shapes and a few of the details. I draw with Freehand (7.0 for Mac), mostly because I had heard that drawing curves was easy. Vector programs are extremely precise and have unlimited resolution. When preparing schematics, I often adjust points by 0.01 mm or less. I'm not sure how accurately you are able to draw in bitmap programs.
By the way, Prakesh asked me how to draw an outline around the ship. Freehand has a command called "union" which performs the Boolean "and" function. What happens is that all areas of overlap are removed from a selected group of objects to make a silhouette. To make an outline, you clone the ship, remove all fills (to make it clear), and hit union. This should create an clear silhouette of the ship. You then increase the line weight and move it to the top layer.
The best way to draw is to construct all views at the same time on a coordinate grid. When I reconstructed Prakesh's drawing, I constructed the side view with the center of the primary hull at an x,y grid location of 0,0 (Freehand allows you to put the center of the grid system any where you�d like. Grid dimensions are in millimeters.). Below this, I drew the top view with the center of its primary hull at point 0,-100. To the right I drew the front and rear views with the primary hulls centered on 100,0 and 100,-100, respectively. Be sure to place everything as accurately as possible (down to the nearest 0.1 mm or better) to prevent headaches later on. In particular, accurate initial places makes it much easier to mirror parts across midlines and to move parts between views.
The main problem with drawing Prakesh's ship was determining the nature of the intersection between the primary and secondary hulls on the top and side views. The method I use is fairly labor-intensive but is accurate and, I think, correct. It involves the use of cross sections to find surface landmarks in different views. Although I am using this method to determine intersections with the radially symmetric primary hull of Prakesh�s ship, I also used it to draw the complex intersectons of the nacelles and the primary hull on my Kestrel class schematic posted a few weeks ago.
The first point I establish is the furthest-forward point of the secondary hull, which is at the top along the midline. On the side view, this is where the top edge of the secondary hull intersects with the primary hull. If you draw a vertical line from this point to the midline of the top view, you can find the same point.
Now comes the fun part. To find the locations of other points, you will need to make parasagittal or longitudinal cross sections of the primary hull from the side view, meaning we make cuts that divide the ship into unequal right and left halves (most MSDs are midline longitudinal sections). As an example, let's make a cross section of the primary hull through the point where it intersects with left side of the secondary hull. First, on the top view draw a line along the left side of the secondary hull through the primary hull. At the two points where the line intersects the edges of the primary hull (point a), draw vertical lines upward to the side view. These lines determine the front and back edges (i.e., length) of the cross section on side view. To determine the height of the cross section, take a front view of the primary hull and rotate it 90 degrees. Place it so that its center is on longitudinal axis of the top view (its distance is irrelevant). Extend the line along the left side of the secondary hull through the 90-degree-rotated front view. Where the line crosses the front view will give the height of your cross section. Now, go up to the side view of the primary hull and clone it. Take the clone and shrink it to the dimensions you just determined, making sure that its center is still at 0,0. Draw a line along the midline of the side view of the secondary hull (which corresponds to the left side on the top view). Where it intersects with the cross section (point b) is where the secondary hull intersects with the primary hull on the side view. Now draw a vertical line from this point down to the top view: this (point c) is the same point and is where the secondary hull intersects with the primary hull on the side view. Now that the points are established, you can draw the curves of intersection on the top and side views.
Since establishing only one or two points means you have to estimate a lot of the curve, you can add more points by taking more cross sections. For this ship I added a second point halfway between the centerline and the left edge on the top view and a third point halfway between the midline and the top of the secondary hull on the side view. The positions of these additional points were determined with cross sections in the same way. The points were used to determine points both above and below the midline of the secondary hull on the side view, but only those above midline are used to find points on the top view (the other points below the midline would be used to construct a bottom view).
This method may seem difficult, but it's pretty easy to use and is the only I know how to construct these interactions without a 3-D drawing program. Give it a try!
------------------ When you're in the Sol system, come visit the Starfleet Museum
posted
I guess that was a tutorial. Actually, I am a school teacher (a university instructor). Bernd says he's going to write some tutorials about 2-D drawing for his site. If he wants we to help out, I'm available. But if anyone has any questions. Please ask.
------------------ When you're in the Sol system, come visit the Starfleet Museum
Printer-friendly view of this topic
