Scale and--

[Alex]

That is correct - the first number is always a "1". As to the 2nd number - the examples I'm going to refer to here are very simple (and not entirely accurate) - but are used to explain the difference in the last number of our scale:

Lets imagine you are flying up in a hot air balloon and you take a photo of a forest from 200 feet directly above it. The amount of detail in that photo shows individual branches, leaves, and perhaps even the ground beneath. That scale is 1:200 -and would be called a LOCAL map.

Now, take that same photo from 5000 feet up. While you can see that same forest, you can't make out individual branches, leaves etc. You can see MORE of the forest, but small details tend to fade away. That scale would be 1:5000 and would be called a REGIONAL map.

Go 30,000 feet up and take that same photo again. Now that same forest is a small patch, and details of it specifically blend in with the surrounding countryside. The scale would be 1:30,000 and would be called a CONTINENT map.

Does that help?

Yes, that certainly explains why the smaller the number, the more detail it has. Philámayaye Silverleaf.

IF your doing an entire continent - say the size of the United States - then the scale could be 1:250,000 - if however your doing a map that features a Castle and the surrounding village - then 1:5000 might be sufficient. But the short and fast answer I lean towards is it depends on the type of map your making. Once you know that, then you can decide what scale you need it to be.

What are the different types (these are brief examples)

Continental (a map of an entire continent)
Regional (a map of a village on the above continent, and its surrounding farmlands)
Local (a single blacksmith shop in the village mentioned above)

I see. Well for me I'm doing a world map (would the continental scale be good?), but would like to have a scale for the full version and a scale for cut versions of let's say different continents. If that makes sense...?

Does the size of the map (canvas size) matter -- actually, I mean does it mess with the scale -- when it comes to scale?

If your referring to this tutorial:
http://www.cartographersguild.com/sh...%96-a-tutorial

Thats a solid question and I'd check the thread in question for a possible answer - maybe even asking the question in it- or Private Message Saderan himself and ask. I don't use Photoshop, so I'm less than useless when it comes to handing out advice on it. I'm sure others will chime in that can answer your specific questions soon.

Hope that helps,

Yes, that is exactly what I was referring to! Alright, I shall try to ask there.

Alex~

[Hai-Etlik]

I see. Well for me I'm doing a world map (would the continental scale be good?), but would like to have a scale for the full version and a scale for cut versions of let's say different continents. If that makes sense...?

Does the size of the map (canvas size) matter -- actually, I mean does it mess with the scale -- when it comes to scale?

Yes, obviously a larger physical map can cover the same extent ("Extent" is the area that a map covers) at a larger scale, or a larger extent at the same scale.


You should be aware that if you want to fit an entire globe into a map, you are going to run into problems. The larger the extent, the more distortion is involved in flattening the globe into a map. (What kind of distortion will depend on the way you flatten it. The ways of flattening a map are called "projections") With an unbroken map, any full world map is going to massively distort scale, and at least one of shape or area. If you want consistent scale, you will need a map with breaks in it (like when you peel an orange and lay it flat) The more breaks, the less distortion. The Interrupted Goode Homolosine and Dymaxion projections are well known examples of interrupted maps (http://en.wikipedia.org/wiki/Goode_h...ine_projection http://en.wikipedia.org/wiki/Dymaxion_map)

Also, the projections suitable for a full globe aren't at all suitable for a continent. This means you can't just crop a section out of a world map, scale it up, and add detail. The result will be distorted. There is software to convert between projections. NASA's G.Projector is one of the simpler ones, and is free and multiplatform, but it has the downside that any input into it must be in a particular projection (Equidistant Cylindrical) Other software can convert from most projections to any other projection, but it tends to be more complex to use. (I use QuantumGIS or GDAL myself)

Also, projection conversion can produce ugly results for "styled" maps or maps with labels. It's best to project raw data and then style it as needed for the particular projection.

Of course you can ignore all that and draw something that looks pretty and resembles a map. Sometimes something that looks like a map works just fine. However, if you want to get it right, so that you can do fun things like put your map on a 3d sphere you can rotate on screen, or if you just have a persnickety streak and want to do things correctly for the sake of doing them correctly, then it gets complicated, and it's the kind of thing you need to worry about at the beginning, not something you can slap on at the end as a quick fix.

[Alex]

Firstly; thank you very much for the reply Hai-Etlik. I appreciate it very much.

Secondly & Lastly; Oh yes, I know this isn't something you can just get overnight, or something you can slap on for a quick fix. I know it takes time to understand scale etc etc, and I am very willingly to learn about it all. I am a very committed person, so when I sit down to do something, I really sit to do it and learn it. I must admit I do have difficulties understanding things sometimes (a lot of the time I cannot help reading things backwards or wrong -- I am dyslexic), but I really want to learn. I want my worlds to be fresh, realistic and as stated, made with the soul and heart.

Of course you can ignore all that and draw something that looks pretty and resembles a map. Sometimes something that looks like a map works just fine. However, if you want to get it right, so that you can do fun things like put your map on a 3d sphere you can rotate on screen, or if you just have a persnickety streak and want to do things correctly for the sake of doing them correctly, then it gets complicated, and it's the kind of thing you need to worry about at the beginning, not something you can slap on at the end as a quick fix.

So yes, I do want to do those fun things and would very much like to do it correct *since it is too late for the map of my novel...*. May I ask for suggestions on where to start for someone who has little to no knowledge of it? Perhaps books to buy/rent *that doesn't read as if its assuming you are a professor or something lol*, or easier sites I can check out? I have tried wikipedia, but seldom do I understand it...

Thank you & good day,

Alex~

[Hai-Etlik]

So yes, I do want to do those fun things and would very much like to do it correct *since it is too late for the map of my novel...*. May I ask for suggestions on where to start for someone who has little to no knowledge of it? Perhaps books to buy/rent *that doesn't read as if its assuming you are a professor or something lol*, or easier sites I can check out? I have tried wikipedia, but seldom do I understand it...

Well I learned most of it in university (I've studied Computer Science, Mathematics, and Geography) and by reading Wikipedia, Wolfram, and other websites. The math on most of the Wikipedia pages for instance really isn't that hard. Undergrad level at worst (My own math knowledge is pretty much all undergrad level) and plenty of it is high school level.

So, if you want to approach it via math, you might want to try working on your geometry (Particularly Spherical and Analytic Geometry) and trigonometry, say at Khan Academy.

Otherwise, it's a matter of trying to remember a bunch of odd things that don't seem all that related. It's possible to deal with without a mathematical understanding, but it's harder to have a clear idea of what's going on. Since I didn't learn that way, I can't give a lot of advice.

I'll try to walk you through a few basic projections and what their good and bad points are.

The most basic form of projection is probably what's called "Gnomonic Azimuthal" It's the basis of the Dymaxion projection I mentioned above, but it's rarely used directly. Basically, you can think of it this way: you take your flat map, touch it to the globe at one point, stick a light at the centre of the globe, and "project" the features on the surface of the globe onto the map, then trace. Obviously it has the problem that you can't fit the entire globe on the map. You can't even fit an entire hemisphere on a map as the light going through the globe 90° from the central point is perpendicular to the map and will never reach it. Also, the further from the point of contact you get, the more distorted the map gets.

Another simple one is called "Equidistant Cylindrical". It can be varied a bit, but the simplest form is called "Platee Carre" and works like this, you take the longitude, and that's left and right on the map, and you take the latitude, and that's north-south on the map. The map is essentially wrapped around the equator in a cylinder, and then the light used to project it is 'curved' to make north-south distances work to a fixed scale (All other distances are distorted). You can vary it by making the cylinder smaller so it cuts through the Earth, and you can tilt it so it doesn't wrap around the Earth's axis but these are much less common. This projection gives the higher latitudes an ugly "stretched out" look. If you use it, you have to draw that distortion into the map, otherwise when you convert to another projection, or put it on a sphere, you will see the opposite distortion (The poles will look "pinched").

Besides wrapping the map in a cylinder, or laying it flat at a point, you can also wrap it around in a cone. This is good for smallish continents or similar sized extents in middle latitudes. Canada and Europe are often mapped with conic projections. If you see "Conic", "Cylindrical", or "Azimuthal", they mean "Wrapped in a cylinder", "Wrapped in a cone", and "Laid flat" respectively. The also have standard shapes, Cylindrical maps are Rectangles, Conic maps are Fan shaped, and Azimuthal maps are Circular, though you can crop a part of the map to any shape you want.

Probably the simplest projection that's fairly widely used for serious maps, is Stereographic Azimuthal, which is often used for maps of the poles. It's much like the Gnomonic Azimuthal above, except that the light is on the far side of the globe from the map (The "antipode") rather than the centre. This can fit a hemisphere onto a finite map, but not a full globe. It also has a very useful property called "conformality". A conformal map preserves angles. If two roads meet at 45° on the surface of the earth, then the roads on a conformal map will also meet at 45°. In a more vague sense you can think of conformality as preserving "shape". It does this by distorting area. The further you get from the centre, the bigger things are drawn. (Gnomonic does this too, but it does it too much and so isn't conformal)

There is also a conformal cylindrical projection, and it is called Mercator, after a cartographer who developed it. When wrapped around the axis (What's called the "Normal Aspect") it also preserves compass directions and is the ONLY projection which does so. This is why it was used for early marine charts when the only navigational instrument available was a magnetic compass. It's also used by zoomable web maps like Google Maps where it's useful to be able to zoom in and get something that's mostly the right shape. It makes things bigger as you get further from the equator though, and the poles are infinitely far away. If you've seen the traditional rectangular wall map of the world that shows Greenland as being as big as Africa, that's Mercator.

There's also a Conformal Conic ("Lambert's Conformal Conic" or "LCC") which I won't go into. There are equal area and equidistant projections for all three shapes, with Equal Area squashing shapes more to keep areas equal, and equidistant sort of sitting in between.

There are also projections that are neither Cylindrical, Conic, nor Azimuthal. These often have elliptical or oblong shapes but they can vary a lot and most are purely for mapping full globes while trying to balance distortion of area, distance, and shape. Most modern full world maps use one of these, with Winkel Tripel and Robinson being popular.

If you don't mind a labour intensive solution that gives imprecise results, you could try drawing your world map in a suitable projection, with a graticule (A grid of latitude and longitude lines) and then set up a blank map covering the area of a region you want a map of with a graticule for a suitable projection (say a conic for a mid latitude continent) and then use the two grids to transfer the shapes on the map across. All you need are suitable graticule templates. I've drawin up templates for Normal Mercator and Equatorial Stereographic Azimuthal here http://www.cartographersguild.com/sh...ector-Template and I'm working on a program to draw Equidistant Conic graticules for any particular latitude range for larger scale maps of continents. For very large scale maps, it gets a lot simpler as you can mostly pretend it's flat.