Not signed in (Sign In)

Vanilla 1.1.4 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome Guest!
Want to take part in these discussions? If you have an account, sign in now.
If you don't have an account, apply for one now.
    • CommentAuthorKGodvalley
    • CommentTimeNov 27th 2017 edited
    So... i got around to trying wilbur, specifically wanting to learn / know that I am capable of using it together with FT3.

    I encountered a problem, which is my main questions, but as it will be referring to images, I will ask my other ones first.

    1. What is the purpose of using Wilbur together with FT3 today?

    All the good guides i found on wilbur-ft working together seem to be from an era where stuff like Incise flow was not a part of FT. Is that right? I guess there are more erosion tools, but i havent gotten to play with those yet, hitting a stopping point on importing/exporting (see below). It also seems to me to be easier to get large flat areas using those erosion functions, but I could be wrong.

    2. How should you handle map info in wilbur? (i realize this is technically not a profantasy question, but seeing is that is where i come from, I'll give it a shot anyway).

    I somehow want to set it to the map dimensions divided by 2 (with minuses on bottom and left). Setting it to +180/180 and -90/90 also seems slightly intuitive. What are "good" values or rules of thumb for getting it, and what difference does it make in the context of the thread, namely the FT->Wilbur->FT cycle.

    3. I assume Zoom Full Size in wilbur should correspond to Zoom extent in FT. Is this right? When pressing zoom full size with the example map I am using I only get a fraction of the full map view.

    4. ... and here is my main problem: When exporting a world from FT (the whole world), I start by Zoom extent. This shows the entire world but with some black edges along the sides. Then I save as MDR using 4096 as my width (thus getting 2038 as height), and hit save. Then I import into wilbur by opening MDR. As seen in picture 1 I get a light blue "frame", which is same light blue as the shallow ocean in the import, corresponding to the black edge of the FT view. Doing some basin fills and incise flows will make this frame green , but that is not the point here. If I then save this and open as new-binary in FT, this blue (or green if i did some edits, but blue in pictures below) frame is still there, although it might be scewed to one side depending on the north pole position.

    The frame when back in FT is a part of the world with an altitude of 0, btw.

    I *think* the problem happens when I first save as MDR in FT, because if I zoom to a portion of the world in FT and then save that (this getting no black edges), there is no frame in the imported MDR in Wilbur. Of course, correct me on that if I am wrong.

    So.. what am I doing wrong? Is it in FT? In Wilbur? In my head?

    Thanks in advance.
      wilbur import2.jpg
    • CommentAuthorKGodvalley
    • CommentTimeNov 27th 2017 edited
    back into FT...
      ft import2.jpg
    • CommentAuthorjslayton
    • CommentTimeNov 27th 2017
    1. Wilbur and FT are very different tools, despite sharing a common heritage and many similar features. FT works on a spherical world and has features that vary based on your zoom level; Wilbur is an image editor with some extra parts tacked on to help it interpret its images in ways that favorable for being describe as "terrain". Why would you want to go back and forth between them? Wilbur has some features that FT doesn't and having detail information separate from interpolated editing information as FT does is inconvenient for certain types of processing. You can eliminate the detail things and interpolation using FT's "Raw Height Field" feature, but doing so isn't really useful in virtually all cases.

    2. I don't understand what you're asking here. The idea of going from FT to Wilbur is to take a snapshot of the area that you want to see in Wilbur and then open that snapshot in Wilbur. Getting data back into FT is then a matter of opening that modified snapshot back in FT. The MDR file format is convenient because it is an image format that supports the full native data resolution of Wilbur and both pieces of software support that format.

    3. Zoom full size in Wilbur and zoom extents in FT are very different. Zoom Full Size sets the zoom in Wilbur so that one pixel in the data image is equal to one pixel on the screen. If FT, Zoom Extents zooms the world so that the full extents of the world are visible (to the extent practical, of course; projections like Mercator are infinite in extent and cannot be shown all at once). One thing you'll notice in Wilbur that you'll never see in FT is scroll bars on the main window. Images in Wilbur always have a fixed size and the system can use that information to enable the common UI feature of scroll bars. FT's drawing area is an infinite space in which there may be one or more images of a globe; FT doesn't have scroll bars because it doesn't have a solid guarantee about how big its images will be, especially if interrupted projections are used.

    4. When you're saving your image out of FT to a file for Wilbur to ingest, you want to minimize that collar of unused data as much as possible. For your convenience, FT has the "Special MDR" image type in its Save dialog. This type saves the world as an equirectangular projection MDR image that covers the full -180/+180 and -90/+90 range (the full globe). This export is guaranteed to have no collar.

    Note that you can select projections other than equirectangular when saving a regular MDR file, but not when saving a special MDR file. You might want to use something like a polar projection to do some processing of a polar continent in FT. Admittedly, it would be tough to get that data back into FT, but you might want to do that anyway.
    • CommentAuthorKGodvalley
    • CommentTimeNov 27th 2017
    A) Thanks a lot :) That cleared things up, and I will definitely give the special mdr a go, think that will solve my import-export problems at the very least.

    B) However, it also opened up a lot of new questions.

    1) Could you explain more thoroughly, or perhaps I really mean "newb/stupid-friendly", your views expressed in 1). Now to clarify one thing first, I am new to both mapping and any image processing software apart from "MS Paint", which I actively used for a year or so when I was 8ish.
    a) what is interpolating?
    b) is that wilbur or FT?
    c) Do I understand you correctly that you should first make a world in FT to look somewhat decent and have mountains where you want them, ocean where you want that, and continents shapes the way you like, then go to wilbur to make the stuff look good by adding erosion and such?
    d) ... what else than erosion (and other than flow) is wilbur good at?
    e) I saw from this tutorialish thing ( the power of Wilbur first time, and it presented me with the FT-Wilbur-FT-Wilbur..-process. Is it not the "common" way to do it? I was planning ending in FT3 to go over to CC3+.

    Also: Maybe I should have been more clear on the purposes of my working with Wilbur. For the time being I am just trying to learn, but the (current) end goal is to map out a fantasy world I am building. I have a rough idea on continent sizes and positions, and a need for some specific terrain features in specific places to fit the story. The "end product" is a set of CC3+ maps, though as a secondary objective I want to have a spherical world that can be shown to people. My theorized process is as follows:

    1) Generate worlds in FT3 until I have continent shapes that somewhat matches my idea
    2) Work on continent shapes and placement of mountain ranges (and some other features) to adhere to both plate tectonic stuff as well as my need for specific features. These "feature needs" involve placement of features relative to other features distance- and angle-wise on a sphere. These features are far apart, and therefore distortions play a major role. That is also intentional, as these distortions will make in-story searching a bit more difficult:
    3) Clean up the world. Plan was using smoothing of flat areas to reduce raggedness (roughness in FT3?!?), doing fill-basin->incise-flow-loops. This is where I was believing wilbur would come in handy. I believe it to be some more errosion possiblilties there (or am I mistaken in that this is one of the strengths of wilbur?). Alternatively I could just stick to FT3 here.
    4) Save "finalized version. Use this finalized version to import maps for CC3.
    5) Either use maps directly or use the trace-method to trace over terrain. If trace then have FT3-map handy to remember where mountains should go.
    6) Make CC3+magic happen.
    7a) Use made maps as handouts in play sessions, hiding and showing symbols according to the price of the map, what they have discovered, etc
    7b) Use for my own novel in the same world, to remember where stuff is, measure distances when I need them, consider difficulties the terrain would present to give my story some flavor, etc

    Could alternatively use a branch at 4 making it
    4a) Create detailed terrain maps in Wilbur.
    .. this path not joining back with CC3+

    Now with that in mind, the purpose of going FT3 as opposed to just doing the whole thing in CC3 were originally:
    a) i believe it to be faster to create a world there
    b) i believe using the fractal function to generate terrain is better than my own attempts at randomness
    c) when getting one continent I want it is more natural, and I at the same time get ideas for "the rest of the world" from the other continents there (this has already happened)
    d) ensure that the aforementioned features align correctly on a sphere
    e) and perhaps most importantly, the ability to use different projections, most notably spherical projections, when exporting portions of the world close to the poles (to reduce distortions).

    *** Are these good reasons?

    In addition, the need to make the entire world and not just focus on a small part (which is what I initially at least need) give some more reasons for FT3 (I think - though this has little to do with Wilbur). These are
    a) To at all times keep an eye on long-lat placement of locations
    b) know that distances make sense. Knowing for instance that my main region is roughly the size of Europe I can envision its outline, create known features etc in world view so I know that I dont go "off scale" when creating local maps
    c) to plan out travel... like "here they can take a boat", "if they ever go here they must invent teleportation", etc.
    d) Slight OCD (meaning non-diagnosed but at times extremely difficult to handle), so non-completeness bothers me quite a lot.

    Now, as I said, currently I'm just trying to learn techniques that I feel are going to be necessary. One of the things i struggle with is getting the terrain to look natural, most notably reduce the "raggedness" in low areas (I am working under the assumption that this is roughness), but just lowering or setting that is not doing it. Therefore I searched for some erosion guides, and got some hits on Wilbur.. and the above-given link caught my eye.

    Now that was a long post, hopefully not too long.
    Thanks again for the answer :) Hoping the length of this post does not dissuade you from commenting further :p
    • CommentAuthorjslayton
    • CommentTimeNov 27th 2017
    a) When we talk about an image being composed of "pixels", it is really composed of data values with space between them. What happens with the areas between points? The computer produces new values according to some set of rules. The most common rules used to produce that new data are: nearest value, interpolating, and approximating. The nearest value (often listed as "nearest neighbor") finds the value of a nearby data point and uses that value, which tends to produce a surface composed of little flat pieces. Interpolating uses some nearest set of data points and produces a surface that passes through those points (linear interpolation is a common form that mixes the nearest 4 values on a grid to get a faceted surface). Approximating also uses a nearest set of values, but data produced is not constrained to pass through the points of the data set (this form can be used to produce smooth results even when noise is present in the data). FT uses a form of interpolation called a Catmull-Rom spline, which uses smooth curves through the data points, but has the undesirable property that it doesn't correctly reproduce hard edges: those get smeared out. This sort of interpolation is fine for most real-world data on the scales that FT uses, but might not be what you desire for certain types of data sets like moon-sized battle stations where the trenches need hard edges.

    b) Wilbur always has its editing resolution the same as its displayed resolution. FT has an editing resolution typically very much lower than its displayed resolution, meaning that it needs some way to produce those in-between values as described above.

    c) You can use Wilbur to adjust the effective editing resolution in FT. With FT3, Wilbur might not be a helpful tool for what you want to do. I added many of the features into FT3 that people were using in Wilbur, so the desire to use Wilbur is less for many people than it used to be. The major advantage (AND disadvantage) of using Wilbur is that it will do its work at the same resolution as its final result, meaning that the extra details that FT adds won't affect the final output.

    d) Wilbur has more shader options than FT, it makes prettier-colored rivers, and it has more data processing operations than FT. Some things that are easy in Wilbur are hard in FT, and vice versa. When FT is using some of the algorithms present in Wilbur (like river finding), it resolves the surface to a constant resolution, does the processing, and then tries to put things back into a form that's useful in FT. The conversion isn't perfect, though, which causes weird noise artifacts to appear in FT as seen at (my old tutorial for FT).

    e) The Israh tutorial is a nice one. If you look at the Wilbur processing, the intent is mainly to increase the effective resolution of the data using the precipiton erosion feature that FT lacks.

    As you discussed, the primary benefit of using FT as your world-builder is that you can get consistency of results across multiple maps, including changes in map projection. Those benefits don't happen when using Wilbur, because Wilbur doesn't provide much in the way of features for dealing with projections or consistency across multiple maps.

    One of the nice, but lesser-used features of FT is View>>View Window. It lets you produce a library of views of your world (including changes in projection) that you can then batch-generate images of CC3 files from. This means that if you need to modify your world for whatever reason, you can regenerate the basic level of your existing maps to accommodate those changes. I think that it's a better feature than the export multiple maps feature because it lets you control the extent of each map element.

    I don't mind long posts.
    • CommentAuthorKGodvalley
    • CommentTimeNov 27th 2017
    Thanks again. Now I must say... Profantasy does have some nice cutomer serive. I didn't know at the time of my previous post that you are the programmer, but knowing it makes the fact that you both read my long post and wrote a long one in return that much more astonishing. As well as making it less disheartening that you just pulled that interpolating-clarification out of your sleeve like that when I didn't have a clue what it was :p

    My key take-away on the Wilbur-or-not-question is that your overall opinion is that I wont need it. I'll still go through some learning rounds with it though, so get more a feel of what it *can* do. I believe it will be some time before my skill is up where I want it, but if/when that happens and I'm happy with the result (or if needing some more specifics tips to get there) i'll post some work here. Maybe even write a tutorial of my own.

    Thanks again :)
    So I am having a very similar problem, but when I tried exporting the special mdr file, I lost my elevation resolution, leaving me nothing to erode.
    • CommentAuthorLoopysue
    • CommentTimeNov 17th 2019
    Before you export the world, in FT3 run Tools->Actions->Normalise Data. This gets rid of any Not A Number data points (NAN)

    When you open the MDR in Wilbur run Window->Histogram. If all you can see in the histogram is a very thin blue line with loads of white space on one or both sides you have a couple of data points that are massively outside the normal -30,000 to 30,000 ft altitude range.

    That being the case, run Filter->Mathematical->Span and set the min and max values to your expected min and max altitudes, taking care to remember that these will be meters in Wilbur, where they were probably feet in FT3. For quick reference - 30,000ft is about 9,144m.

    That should put things right for you.