Mapping Shield Lands Part 6: Coloring to the Land

2021-03-02

Back to the techie side of mapping again for a first look at texturing, the art, and craft of coloring computer generated objects, in this case terrain. As always when it comes to CG terrain creation nothing is straight forward, so texturing will be done in stages. Texturing is in many ways the best part of my work; you get to see that the awkward looking mess you have been working on for a long time suddenly start coming to life. It is extremely rewarding, and I often get lost in details and dream of who and what lives there and the events taking place. Making sure that stories set in the area I am working on have a suitable location, is a part of my checklist all the way from early terrain generation phase to the end of texturing.

The primary purpose of texturing is to make the terrain look realistic, inspiring and give detailed information of what a certain area consists of, rock, sand, march grass or forested for example. For a map that is going to be used in a single way in a specific medium like printed on a poster, then you create a texture with the right resolution sharp enough to retain all the details in the final print. Printing needs can vary from small cards to wallpaper for a living room wall, all with different needs for resolution and detail.

RPG needs in the future will span a much wider need than prints, VTT’s, PDF’s, Geographical Information Systems and Game Engines are the ones I am aware of now, but more will come I am sure.  When you do a project of this magnitude a key part is to make the most to assure usability as possible and make the results customizable, prolonging their life. For example, forests can grow in places and be cut down in others, seasons come and go, and demons might invade, or magic calamities affect the lands.  In all these cases it is good to know what the terrain is made of and what grows there. That way you can turn all the vegetation in an area brown, black, or grey to indicate a blight, or change the color of deciduous trees to indicate fall but leave the pines and other evergreen trees the same color for your seasonal maps. This can speed up editing and improve the result a lot but providing the right data in the right format.

Area data is mostly stored using masks, a greyscale image with brighter pixels indicating a presence of something. This is great when you want precision, like for elevation using heightmaps. Another way of storing areal data is to use a Splat Map, it is a color image using each of its RGB channels to store data of some variable. Some image formats like PNG also comes with a transparency channel with means you can store four masks in the same file, improving storage efficiency and handling. For other use cases you do not need precision, instead you need to know which of several variables a particular area has, so you can use the color of that pixel to indicate a binary value. This way you can replace lots of masks with one image and save lots of storage and by using the appropriate colors also make them directly usable. Below is the set of colors I have used for my Shield Lands Splat maps.

By using various selection tools in World Machine, you can decide what type of terrain goes where. Remember I wrote about lakes and rivers in an earlier post, the result of that work was a mask that is now the source for which pixels will be water. Selection can be straight forward, like above or below a certain elevation, to extraordinarily complex like erosion flow zones phasing between 90 to 270 degrees and with and not steeper than 55 degrees for example. The variations are practically endless, but the key is to try and figure out where trees or grass grow turn that into rules and select for it. Below is an example of what a splat map in progress can look like.

And with added shading to make it more understandable.

It is still very much a work in progress, trees and bushes are still to be added and just like with the rivers an artistic step using Photoshop is needed to polish it up and remove some ugly bits that pop up here and there. To assist in the Photoshop editing the same masks that was used in the water editing can be used in this step as well. Since we are not messing with the elevation the terrain stays consistent and we can eliminate changes to the terrain that so often comes from changing the size and placement of the render area, which means you can create a smaller render area to quickly preview it in full resolution. When you work with elevation fractals the previews can be very misleading, by getting all the terrain building done in the first phase we now have a much more stable and predictable environment to work in getting the details right.

Splat Maps also have a key function as a coordinator between formats, for example to make sure that the large format texture matches the details generated in a game engine gets textures and models of trees and bushes in the right place and of the right type. Another example is in GIS you can use the Splat Map to create matching vector shapes and styles that are in the right place.

By using colors that are reasonably natural even the plat map can be understood and useful, but it is as a template for creating the final textures it shines. Looking at it even in this unfinished state I like it and combine it with the elevation, and we can start to see the final landscape emerge.

Next installment in this series will be creating the final textures using World Machine, Photoshop, and the Splat Map.

Thank you again for all your support!!

Mapping Shield Lands Part 5: Lay of the Land

2021-02-24

After a series of very technical posts, it is time to go back to the World of Greyhawk. This time as literally as possible and trying to figure out what the land looks like, surveying in cartography speak.  In real world mapping this is done by combining field measurement, aerial and satellite imaging and so on, but how do you survey a world that does not exist?

My approach to fantasy surveying is to use all the available sources in the form of texts and maps, real world inspiration and use logic to try and create a landscape that would fit the inhabitants and stories. Text is in my opinion often more reliable than illustrations and maps, maps and art is usually created by outsiders with limited knowledge of the setting, and tight time and monetary constraints. This factor often makes the text more accurate than accompanying material, a big caveat to this is that some writers does not seem to pay much attention to geography. Landscape is often not even mentioned, and if it is features are added either as flavor or because they are needed for the story. This makes it hard to deduct geography from RPG material, and there are lots of inconsistencies.  Small details are given lots of attention and vars areas are just glossed over. My job as a fantasy cartographer is in many ways to fill in all those huge gaps, to glue the few details mentioned, together.

For my campaign and as a first test area for a new generation of Greyhawk maps I am doing this for the Southern Shield lands, so let us investigate this process in more detail. Phase one is to look at the sources, in this case: The 1983 Greyhawk Boxed Set, From the Ashes Boxed Set, Living Greyhawk Gazetteer, each of them came with a setting map. Living Greyhawk modules and some fan created content gives a decent amount to go through.

To better understand the texts, I want to first go over the maps, starting with the oldest. In this case that is an old Greyhawk setting map that comes from Len Lakofka. I think it predates the Darlene map, and it is interesting since it is in large feature complete but still very rudimentary. The Shield Lands shows the layout we are familiar with from later maps with the exception that Admundfort is in the middle of Walworth Isle.

Next in line is the defining map of Greyhawk by Darlene. Its sweeping lines and shaded coastlines are beautiful and combined with the 30-mile hexes it gives us a basic understanding of the geography.

Next up is the Greyhawk map that came with Dungeon Magazine by Rob Lazzaretti. It is different in style from the Darlene map, but shows the same general geography. Many new settlements make their first appearance on this map, and Axeport, Herechel and Bright Sentry show up in our area of interest.

Living Greyhawk campaign was a source of a plethora of maps ranging from maps that are solid additions to the lore of the setting to maps that should in my humble opinion never have been made. I have found one map, apart from the Lazzaretti map, of the Shield Lands from the LG era. The map below adds several new settlements from the LG modules, but nothing else that we did not already know from the other maps.

Thankfully, Eric Anondson created a series of fantastic Greyhawk maps adding a lot of details from the LG era, it was and still is some of the best Greyhawk mapping and a huge inspiration for my Greyhawk mapping. I started my efforts before he did, but worked so slowly that he quickly covered more than I did, which gave me the chance to be inspired by his great work.

Erics map gives us a first look at how the elevation of the land and how it is drained and confirms the placement of settlements added in the LG modules. There is one river system that reaches a large part of the interior of the Shield Lands and flows into the Nyr Dyv west of Axeport. If the major port along the coast is not placed at the estuary is telling, if the river was easily accessible by boats it should have a settlement housing those who exploits the trade opportunities. Making the estuary of this river, that I named the Sanstin in my campaign, a shallow marshy delta stretching inland for several miles would make the Shield Landers looking for a better place to build a port.

Naming conventions for places can have many roots, and it this case I thought placing Axeport on a small peninsula that is vaguely axe-shaped could ad a bit more realism to the landscape.

I made Axeport a large town in my campaign which is why the guide overlay reflects that.

The Veng must be a huge river in the same league as the Mississippi or Volga rivers depositing large masses of sediment into the Nyr Dyv making this part of the lake shallow and muddy. Scragholm Isle has a core of rock that parted the river but is mostly made up of sediment that have solidified over time making it flat with marshy edges.

Critwall I reason is located where it is due to both good access to river and lake traffic and by placing it on a slightly elevated peninsula gives is both good flood protection and makes it easier to defend. The next major settlement along the coast eastward being Axeport some 70 miles to the east gives us a lot of coastline to detail.

Next step is to work out the climate, and here there are way more to go on. The Shield Lands are stated to be in the temperate zone, and the small overview map on p18 in the 1983 Glossography gives it a northern latitude of 40 degrees. This is equivalent to the Great Lakes area in North America or the northern coast of the Black Sea or the Caspian Sea. The latitude is only one component in the climate, how dry is it, how are the seasons, and where does the wind usually come from are other huge factors to way in. Oerth also have a larger axial tilt of 30 degrees which will enhance the seasons.

Oerik is a huge continent larger than Eurasia and the Flanaess are larger than Europe wo we can assume that the Shield Lands would have an inland climate moderated a bit by the Nyr Dyv’s mass of water but exaggerated by the axial tilt. The Flanaess has mountain ranges in the west and east, with the west being massive, this should leave the center of the Flanaess less humid with dry hot summers and not overly wet cold winters.

The texts hardly mention geography at all which and combined with the lack of map features gives us a good reason to presume that the area can best be described as plain, combines this with it being a bit dry and having an inland climate and grassy plains would be the dominating landscape. Sounds boring, so now it is time to add in detail and creativity!

We have a huge volcano and an out of this world sized rift a few hundred miles to the north east, so we have geological activity for sure. This means we have fault lines; old lava flows and other interesting features to add. Maybe not in the immediate area of interest, but things can and often should be worked in gradually, so we need to keep this in mind.

The geological history of the area is also especially important, has it been covered with ice, how old is it, has it been flooded often or recently, was it more humid in the past. Cannot find anything in the texts about this so we must improvise a bit. I will hold off on glacial erosion for now, faults and volcanic activity is enough to work with. My assumption will be that the area is generally sedimentary, meaning it is old lake or sea bottom that have risen to form the land. Geological forces in the past have created a rolling landscape with low hills here and there, and volcanic activity over the eons have sprinkled it with occasional stand out features, from old worn-down craters to lava flow areas forming craggy ridges and mesas.

Depressions in flat landscapes tend to fill up with water, and if the waterfall is rich and regular rivers form and erode the land further making some lakes disappear and creating new ones in other locations. All this means that there should be way more lakes that are on the almost lake free maps that we have seen so far, and in not so wet climate of the current Shield Lands there should be lots of old dry lakebeds now overgrown with grass or bogs that are either dried out or still wet.

Vegetation will be mostly bushes and trees near rivers and other wetter areas. The few hills and higher terrain could be home to more vegetation due to humidity from clouds being more common that down in the lowlands. Similarly, the river valleys and other low spots in the landscape could get a bit more fog and mist making them home to more vegetation. Edges of rivers and marshes would also often have a string of vegetation.  If trees are scarce areas close to settlement might have even less due to humans cutting down most of the few trees for building material. Firewood can often be replaced by peat to save timber for better use.

With these thoughts I am setting off to visualize the landscape of the Shield Lands, I hope I have inspired you to give fantasy lands some extra attention in your gaming!

Mapping Shield Lands Part 4: Lakes and Rivers

2021-02-17

Here comes a look at how to create realistic lakes and rivers detailed enough to match the high resolution of 5 ft per pixel that is the target for this series of maps. World Machine does an admirable job of naturally eroding the landscape to create a realistic large area drainage and then also add lakes as well. The problem is that it creates rivers and lakes everywhere that can possibly form, and they look a bit stilted at this high resolution, so a bit of manual artistic touch up is needed.

Below is first a top down views of the terrain with all the WM lakes and rivers in blue, and then the second image with my edited layer on top in lighter blue.

To make the export of the final mask easy I create a new layer, make it solid white and give it a mask that I can use black or white which makes all my adjustments non-destructive. A color overlay makes them easy to see and mean I can adjust visibility while working and keep mask functionally intact. A solid black layer underneath an turn off color overlay and the mask is ready to be exported.

I save it as a TIFF or PNG file which WM prefers for masks. Import it back into WM again using a File input device, and it is easy to place it if you have the same extent active as you used for the export, just click the "Place in Current View" button will get it right. And now we have the adjusted lakes and river system in World Machine. Note that this is only a test, so I have not done all the lakes and rivers, far from it. I need to spend another day or so getting the whole 16K area done. The first areas while take a lot of extra time learning a creating the right brush setting, but effectiveness will come with experience, so future areas will be done much faster. the type of terrain also plays a large role, some areas have a lot of lakes and rivers, and some might even be so water-rich that they fill out almost every possible area and little adjustment is needed.

Below is how it looks in WM using the newly created mask.

Next up after this step is to take this primitive elevation coloring and create realistic texturing and all the data needed for further enhancement in Photoshop, QGIS and Unreal.

Thank you all for your support, it is literary a lifeline for me!! 🙂

Mapping Shield Lands Part 3: Pieces of the Puzzle

2021-02-13

Time for the next installment in my series in how to map the World of Greyhawk, all over again. My last post was about how to use fractals to create the shape of the terrain, now its time to lift the perspective and look at the big picture. How to build the world, one piece at at time.

Even the best of computers cannot recreate a whole virtual world full of details and color in one go, it is like eating an elephant, it has to be done one piece at a time. Add the quirkiness of terrain generation tools to the mix and it requires you to be both careful and daring, and to plan how you do it. It has taken me years to figure out how to do it, trial an error and learning from others attempts. This says as much about my intelligence as it does about the difficulty in doing this, I don't advice anyone to do this unless they are VERY passionate and possess a large potion of stubbornness, if that is the case you should definitely do this. You will have something to occupy you for a very long time 🙂

When you have fiddled around with the fractals and have created a bit of terrain that looks perfect, as a world cartographer like me you get fond of it and want to save it and shift focus a bit and add more carefully shaped terrain next to it. This is where one of the many quirks of CG terrain creation tools come in to play, when you shift focus ever so slightly the results change. The same happens when you look at things close up or further away, and I find it VERY annoying and it took me along time to learn to live with it and figure out how to work around it. 

The way I have decided to deal with this problem is to first calculate how large of an area you can render at your desired resolution. The desired final resolution is 5ft, and the maximum render resolution is 16K (16,384 px). So a 16K map @5ft per pixel is 81,920 ft which is 15.51515151515152 miles. This is not rocket science and no real creatures are going to get hurt, so lets round it off to one decimal 15.5 miles. Hopefully it will be possible to work with bigger renders in the future, but right now this is the size we have to work with. there is an option of using tiles to cover larger areas in single render. It comes with its own set of problems so I'm going to avoid using it if possible, but it is an option that is good to keep in mind, and I'm also recommending to use a workflow that retain that option. This is another reason I decided to split terrain generation from the water creation and texturing.

Rendering of an 16K area take my now venerable 8 core Intel workstation around 6 hours or so for the height map alone. So if you don't have one of those new 32 or 64 core Threadripper monsters, I recommend to have a separate computer that can crunch in the background while you can work on other things. Overnight I often set both computers to render to speeding up things. The result of the render needs to be exported into some sort of file and stored for suture use. The only thing needed from this fist stage is the elevation data in the form of an height map. An heightmap is just an greyscale image with the higher areas being indicated by brighter and brighter grey. Any type if bitmap format that can handle greyscale images can be used, the key issue is the bit rate which is effectively the vertical resolution. Most images we take with our phones or see on the web are 8-bit, which is great for Minecraft like terrain but for detailed real world like mapping we need way more so its 16-bit minimum and preferable 32-bit. The files get significantly bigger with each increase but it is necessary to keep our options open. World Machine has a built in 32-bit native format that saves the date as a WM Library, the downside to this is that the data can only be used back in world Machine again, but at this stage that is not a big problem. The huge bonus is that the date is geo-located, so when it is imported back into a new WM project the it knows its place. This saves a lot of note taking and keeps the precision really high.

Importing the saved WM library file makes the next CG terrain quirk apparent, raised edges. When you render an area the center bits match each other almost perfectly but the edges stick up and need to be masked off manually by using a set of Layout Generators like this:

It is a bit of a fiddly nuisance at times, but with a bit of blur here and there added the result are usually more than acceptable. Now we have a ready rendered area to add to that are fixed so no nasty fractal surprises and to look at it in high resolution is much faster since all the hard work calculating fractals and erosion are already done. By using WM Library devices and keeping the exact same settings for the guide map, working on areas using different WM project files becomes possible without sacrificing quality. Instead of using a huge WM file with a large amount of terrain recipes in order to create everything from arctic mountains carved out from glacial ice to wind eroded deserts landscapes in single file, we can now use lots of file instead and export an import between them. The drawback is that you have to render at the full resolution at all times. I have tested use a master terrain rendered at lower resolution that covers the whole area first, and then scale it up to add detail but found it lacking in variation. Even at lower resolution you need to have a huge variation to create a believable landscape covering an area as large as the Flanaess, and a fantasy world should look even more impressive in places. This require fractal combinations in the hundreds, with as many different erosion variants and settings as well.

The first job is to be artistic and first try out different fractal combinations together with filters and different erosion settings, then you need to play around using Layout Generators to create the landscape you need. Then comes the tedious rendering, exporting and then import back the results and mask the ugly overlapping bits. Water is next, to map out the areas that can become lakes and rivers, and create the ocean surface or in this case the surface of the mighty Nyr Dyv. Fir this WM comes with a built in Create Water device, and as you can see in the screenshot below, it is not fully compatible with tiled builds. Lots of nifty and handy devices comes with this limitation which so try and use a non tiled approach if possible when you are using World Machine.

Next is a very time saving follow up device with a cryptic name of Composite Type, it is one of the hidden gems in WM. Another user pointed it out to me when I complained to have to fill in all the lakes manually. The Composite Type Device will do this automatically for you, and it will also give you a depth map, which is great for texturing which I will cover in a post coming soon. The Create water and its ability to generate a complete and believable natural looking lake and river system is a godsend and one of the major reasons I'm using WM, no other tool does this, this well yet. The catch, and there always is a catch in CG terrain work, is that you always get a maxed out set of lakes and rivers. In some areas that are rain drenched and relatively you genealogically speaking, this works great. Most areas though comes with a bit less rainfall and lakes have over time partially of fully filled up creating marshlands bogs or flat solid land instead of the lake that once filled the valley. Rivers carve out a deep valley but then are reduced to a lesser flow that meanders thought the old river depression. This requires new set of render, export, artistic intervention using Photoshop and then import back into World Machine again. It is tedious work but actually rather relaxing and fulfilling, it is hands on world building without waits for renders. It will also be the topic for a post in the series on how to map the World of Greyhawk coming very soon.

Mapping Shield Lands Part 2: Terrain Generation

2021-01-30

This was the post I was planning to post first today then the Ritersmarche area update got ready early and I posted that one first. First let me say welcome to all the new patreon members, thank you so much for making this possible!!! 🙂 The Greyhawk 2.0 project is experiences both setbacks and successes, it is one step back and two steps forward. I was hoping to use the new World Machine 4, but that seems to be a bit premature. One of the key features for me is to have overview, when you map large areas you need to have an idea of what this look like on a larger scale. This has always been where World Machine stood out against the competition. In the latest version emphasis have shifted more towards better view of the area in focus losing some of that overview. This is probably something that most artists and game developers like, but a problem for me. Thankfully the old version is still here, and I can have both installed side by side and use the best of both, and with a few updates WM4 might be as good as WM3 in this regard.

The image above shows how much of the world I've mapped at the first stage so far. From the Critwall peninsula and Scragholm Island in the foreground to the former Shield Land coast all the way past Axeport to Battledown in the east. The vertical coverage from - 1 km (0.6 miles) to 10 km (6.2 miles) with 32bit which means enough resolution to make it all smooth where that is needed and sharp when that is needed.

There are more water in this stage than it will be in the final version, especially lakes and rivers. The texturing at this stage show all the areas that CAN be lakes, rivers and other wetlands as blue. I will mask that down to a more reasonable level at the next stage, converting some of the blue areas to marchland or just flat land areas. This process will mostly be artistic and done manually in photoshop and then re imported into World Machine to feed the next step. This first step is mainly about terrain building, so lets take a closer look at how that is done.

Most computer generated terrain generated using fractals, and World Machine comes with three different fractals generators Basic Noise, Advanced Perlin and Voronoi. The first two are both perlin generators but with different ways to do the settings.

Here are an example of a Basic Noise output

and output from an Advanced Perlin generator

Perlin noise is great for creating all that under-laying fluctuations in landscapes, especially lowlands and the variations of large areas that are only visible against the horizon or from a high vantage point. These minute variations in landscape elevation forms the basis for lakes and rivers and how water flows across continents, and this is one of the main base data for texturing so even if it is hard to see the differences, the play a huge role later in the process.

Voronoi is a bit different, it created various forms of angular patterns in the forms of triangles or squares, sometimes very pure or broken up into intricate patterns. Voronoi can be used for a range of things from forming the base for large massive mountains to the pattern of farming fields around a village and even a random placements of houses in a city.

Fractals are rarely used in simple and pure forms, it is when you blend and tweak them the magic starts happening. You can blend data from tow generators using a Combiner, ad the data can be added, subtracted, multiplied or mixed in over ten different ways. Photoshop users recognize that this is just like layers of pixels can be blended in images, it is all the same type of manipulation of data.

Below is an example of how I have set it up in WM, this is a simple setup using a single Advanced Perlin generator. Fractals look repetitive over larger areas which in usually not the case in nature, so that need to be fixed. Large area variation is created by introducing distortion, which is created using a Distortion Generator. To make it more complicated the Distortion Generator needs to be given a pattern for the distortion and for that I've used two additional Advanced Perlin Generators one for the N-S (X) axis, and one for the E-W (Y) axis. WM works using gray-scale heightmaps so Z axis is not needed.

The generated fractal data is then fed into a Flow Restructure device, which makes sure all the valleys are flowing downhill creating realistic drainage. This device is a godsend but a late comer to WM, and if you look at all my maps so far they are not realistic in this regard. Thanks to this nifty device and its companion device Create Water it is now possible to create realistic lakes and rivers automatically. The data is then sent to an Erosion device and then a Clamp device to reduce it to form lowlands. Erosion is huge rabbit hole that needs to be looked at specially, so I'm going to skip over it today. 

The green device labeled TER1 at the bottom of the group is a Layout Generator that let you create vector masks telling WM exactly where and how much you want this terrain to cover. A Blur Device smooth out the edges of the vector masks to make them blend in a bit more, and then a combiner with the Max blend mode integrate the data into the rest of the terrain.     

More complicated versions of this uses two or more chains like this blended together in various ways.

You can learn my by watching my introduction to World Machine YouTube videos:    (56) Anna Meyer - YouTube or even better resources  can be found here: https://www.world-machine.com/resources.php?page=tutorials

Rittersmarche Area

2021-01-29

Here comes an small update that have been on my to do list a long time, make the Rittersmarche area to better match WGQ1 Patriots of Ulek.

Changed the Jewel river to flow a bit  more easterly further into the Suss. The town of Riddlings pass is now moved to the foothills west of Rittersmarche, and the foothills themselves have been moved west creating a larger grassy plain area between the Suss and the southern Lortmils.

The Suel Temple ruins had to be moved east to stay on the Suss side of the Jewel. I think that the new location near the Troll Creek and the Suss fens are even more appropriate.

Oakenburgh and Brenfluss now have the town symbols to reflect their true sizes.

All these changes will be on the 2021 map update.

A large thanks to Vincent Mattelaer for helping me research this area!

Mapping Shield Lands Part 1: Setup and workflow overview

2021-01-10

What do you do when you have accomplished something? You start all over again, this time with the goal of doing it better! So with twenty years of experience and technology progression I'm setting out to map the World of Greyhawk again, and this time in much better detail, from a mile to a few feet. This post is a first look into how I have decided to try and go about doing it.

Twenty years ago the main tool I used was Bryce 3 from DAZ Software, which was the coolest thing in the mid to late 1990's. This time I have decided to use World Machine 4 as the main tool for terrain creation, you can find out more about it here: http://www.world-machine.com

WM is uses a visual node based approach to creating instead of text based coding, which is what I need suffering from a mild dyslexia, but very good visual analytic perception. The key to understanding WM and similar tools is to see them as data manipulators. How that works in practice I tried to explain in a series of WM videos you can find here: Anna Meyer - YouTube.

The goal this time is to create a virtual model of the Flanaess and potentially the whole planet of Oerth at a resolution of 5ft. This model will include elevation, texture and auxiliary data like coastline maska, biomes and more. In parallel create "in game" handout maps as well, but more about them in future posts. For now I'm going to concentrate on setting up the workflow. World Machine will be the hub and main tool for terrain creation and texturing. I will use additional tools like Gea, Vue, Terragen, Blender and probably a few others to create certain features. This is going to be needed to ensure enough variety, which is one of the key aspects of believable virtual terrain creation. Nature look alike but never exactly the same, so in order to cover millions of square miles a large number of fractal combinations will be needed.

The first problem to tackle is the size. Virtual terrains are mostly used in computer games and as a backdrop for movies or as a ground for architectural virtualization, but this project is a on a vastly large scale. The Flanaess map covers 6000 miles West to East, and with 5 feet resolution that means 158,400,00 pixels across! We need to move to 128-bit computing to handle images of that size in one go, so it needs to be done in parts. In WM that can be done using tiling or re-import areas to work on adjacent parts.  I think both approaches will be needed since they both have pro's and cons, tiling makes consistency of rivers and the terrain over larges areas easier but you can only really see the result with the final render. WM can render 16K which means 16384 pixels in a single render. This is mostly due to memory limitations, which means you get a reasonably detailed view up to 125 miles. With tiling you can render up to 64K which makes means you can cover huge areas in one go. This will be handy when you have large areas when you don't need to closely match exiting maps, but want to have all the detail. With more RAM both can hopefully be increased and the accuracy of previews will hopefully increase as well in future versions of my tools.

My plan is yo get a new Threadripper workstation later this year with 256GB of ram a really fast SSD array and a beefy GPU. This will both speed up render times by a factor of ten, and make it possible to work on 32K or even 64K in one go. In a few years with even faster computing and GPU's number crunching it might be possible to render 128K in a few hours. I made the same bet on Moore's law 20 years ago, and I think it is safe to make the same bet again this time.  The Flanaess is withing reach in five or so years ad the planet within 10!

At first I tried to create everything, elevation, textures and the other data in a single pass. After years of struggling I now know that it is better to do it in stages. First a terrain base at roughly half the resolution to get coastlines hills, lakes and rivers in the right place and export that. Then merge the renders in a new file, detail the terrain with little things like boulders, cracks, wear and tear. Then render that again at the final resolution together with masks and textures. Since the underlying terrain is already done you can take quick sneak peaks at small areas and rest assured that what you see is going to look the same in the final render.

Keep things simple is the order of the day. I've learned that the hard way after years of experimenting.  Below is the principle of the first stage in world Machine.

I have turned the different part into macros so it looks deceptively simple, we will look into each of the components in future posts. The first part is Terrain Import, to handle areas and features that have been created using other tools.

The next stage is to blend in the different terrain "fractal recipes", Terrain 1, Terrain 2 and Terrain 3 in this example. Each of the macros can consists of a myriad combinations of fractals, erosion and filtering, and they are controlled by a Shapes Device. the Shape Device can either be inside the Terrain Macro or at the Combiner, both works, but can yield different results. They can be blended into the terrain using different combiner settings, like Add, Max, Subtract, Overlay or Screen, depending of how you want it to affect the final result. Next come the most complex bit, Lakes and Rivers. This will be a topic for several blog posts and videos, it is a deep topic!

The final bit is export and Guide Map overlay and the Overlay View device. To cover the Flanaess a dozen or more of these files with be needed, each with a dozen or so terrain types in them. This can create a virtually endless variety of natural looking terrain.

Another very important consideration is the vertical scale, how high are the tallest mountains, and since we are talking about Greyhawk how deep are the Rift Canyon. The Rift is a topic in itself, but I think it is too deep to cover in a the main terrain creation. The goal for the first stage of terrain creation is to cover all the landmass and the coastal waters. My reasoning is that some part of the land is below sea level so by including one you can include the other in the same setup, the coast below seal level is also close connected with the land and makes it a lot easier to cover how tidal and other changes to the ocean levels affect the coasts.  My belief is that 22 km (13.6 miles) is enough to cover everything from coastal seabed, low-lying terrain to the tallest peaks. Below is an example terrain 6 x 6 miles with a few little islands and a peak rising to a bit over 6 miles up.

This was a first look at how I plan to do this. Next it is time to dive in to each of the components and start looking at the details.

New Year - New Horizons 2020

2020-12-31

Another year is coming to an end and it is the usual time to reflect and try to look forward, and this year it feels way more significant that almost any other year. 2020 have been, despite all the bad things that have affected the world, a year with lots of new possibilities. When your work is visualizing fantasy worlds, living in a world where almost all of us have been forced to stay home is not the worst that can happen.

This have been my most productive year with many small projects and three big ones that have taken shape at my desk this year, two for Kobold press. Southlands redo for 5E and a not yet announced project, both of them very large that I started on back in 2019.  Going back to the Soutlands that I worked on 6 years ago was good kind of nostalgia, and I'm pleased how well the map has held up and now it is a bit better. The third big project was the Flanaess 576 CY map and its Atlas version. Originally planned to be a two or so month project, but it took over 6 months of hard work that turned out really well in the end.  This project is also a bit nostalgic, it is the last major version of the map that got me into fantasy cartography. For over twenty years I've worked on the same map using almost the same tools either creating it or updating it. That era is now coming to a close, future Greyhawk mapping will be using a new set of tools and techniques.

When it comes to tools this have been a year of ambivalence, great promise but few actually useful progress. Gaea was a tool that looked very promising this spring, but development have fallen behind and the features I need for my work has yet to materialize. Thankfully World Machine has made a comeback and is now ready to be my go to tool for the new generation of Greyhawk maps. I just now on New Years Eve downloaded the final build of WM 4 and it has all the features I need. Using tiled builds we can cover up to half a million square miles at a times at 5 ft resolution. That is a bit too much in one go and make sure you can terrain that is varied enough, so I will do it in part not that big but still many tens of thousands of square miles.

My idea is the work on each drainage basin at a time to assure that rivers, lakes and its shore lines are matching properly. Splitting it up along the major rivers and mountain ranges where possible and try and patch over the errors using GIS mending tools as well as import the broken areas into World Machine and adjust the ugly bits. Trying to keep the problematic areas to lake and river bottoms and remote mountain peaks will hopefully make them less prominent. Real world cartographers phase the same challenges, so there are good tools for this and mapping a fantasy word gives us two advantages, it is possible to go back into and tweak the output and no one can say its not matching the real world.

The biggest hurdle in the "Next Gen Greyhawk Mapping" is water. Having lakes and rivers draining the landscape in a natural and believable way procedurally, instead of having to do all that manually. It is now possible to place major lakes and rivers, and having World Machine filling in all the fine details. This will save almost half the time it takes me to map an area, getting a faster computer will cut that time another half. This makes mapping the whole world become possible and the Flanaess going from a "over a decade project" to a more feasible five year project. Below is a bit of test terrain I made in a few minutes, combining all the watery elements with two types of terrain and proper erosion.

This test is only low resolution but it proves that all of the details needed can be both controlled and procedurally generated fast and reliable. I'm especially pleased of the water texture. This is just a quick sample using two colors, with an array of colors and textures it is going to look fantastic in high resolution. The Darlene map had a detail level of 30 miles, my current map has 1 mile, my goal with next will be 5 feet. The maps I have created for my Shield Lands Campaign have been a test run and proven it is possible and WM4 and a much more powerful computer will make it doable.

There are so many aspects to a project of this magnitude, and how you decide you will go about might be something you have to live with until you're finished. I've learned that the hard way, so this time I'm way more careful, maybe even too careful and trying to figuring out the best way to go about this. I've seen the Middle Earth project stumble and grind to half after several years of great and steady progress. I remembered Peter Jackson in an interview a year of so before the first LotR movies stating that "Now was the time to do it". That is how I feel about starting all over to try and visualize the World of Greyhawk in a way that I don't think any fantasy world has ever been so far. Now finally I have the tools, the experience and most importantly you guys to help me make this dream of mine come true. Thanks to your support I can now treat this as a part time job, pay for the tools and set aside time to work on this.

It will be a project with many phases and some of them need to be done in parallel, so many thing depend on each other. In order to create accurate terrain I need to know where the Flanaess is located on Oerth, so we need to have at least a rough working model of the planet. Terrain is directly linked to climate so we need to have a model for that as well with all its variety natural and fantastic. Data standards for all the information need to be established, keeping withing the laws of computer storage limits and in formats that can be used by most without expensive tools and a decade of computer graphics experience.

As aside quest I've been learning QGIS and just started with Unreal, both will play big part of this project. My goal is to use them both to produce and present, QGIS for traditional maps in print and digital. Unreal for a immersive 3D experience. There will be an overlap between the two and to make the raw data available for others to play with is key in this. I want to se many different version of WoG in the future using the data from this project as a base. I call it data since it will be a huge collection of images, heightmaps, masks, GIS shape files, various databases and more.

I want to create an immersive, inspiring and useful set of maps, data and even tools down the line to help DM's and players have more fun in Greyhawk. It might be technical and look boring and "computery" at times, but when you see a landscape you recognize as being somewhere in Greyhawk the feeling is magical. I hope you can share that feeling with me, it is a magical time we live in when you can use both words and fractals to bring fantasy to life, for me they are both very important for my gaming.

Together we can do it!

Symbol Update - Settlements

2020-12-30

My Greyhawk Symbol Set have been around since the 1990's with some additions over the years, and they have for the most part done the job. But they could definitely do with an overhaul, some more than others. I created them using corel Draw back in the 1990's and my design skills were modest at best back then, and soon I will need SVG versions of them so it is a good time to take a thorough look and make them look a bit better.  

First out is the basic settlement icons. From Metropolis to Dorf, with and without fortifications. The main thing I have adjusted is the fortification part of the symbols to make them symmetrical and removed some editing artifacts the old ones had. Especially the Village icon looked terrible and is now in line with its bigger siblings.

The  old Thorp and Dorf symbols was some vague attempt of mine to make stylistic icons and they looked terrible. The new ones are attempt to get them in line with the rest and try and convey the meaning of a few small houses or a just the smallest of settlement. What do you guys think?

The relative size is a look at the size difference on the maps where they will all be present. Moving to GIS based maps mean that using some scales this will be the case, but on larger scales the smallest settlements will not present and on larger scale maps larger settlements will have a basic layout instead of a symbol, a first look at that is coming in a post soon.

My goal with the symbols have always been to try and keep the Greyhawk feel and looks from the Darlene map but expand and slightly simplify them to make them quick and easy to identify. 

Thank you!!

2020-12-24

I hope you are all staying safe, and have as good of a holiday as the times permit. I'll take some time off to test out my new VR headset, but otherwise it is mapping on the agenda.

It will be a mixture of a lot of different things: a refresh of the Greyhawk symbols to make them ready for GIS next year. Finish the first part of my Shield Lands map. More 150% Flanaess maps. New heraldry for my campaign. Create a World Machine Template for Lendore Isles and future Greyhawk maps. Learn more about Fantasy Grounds and Unreal.

Thank you again for your support!

1 17 18 19 20 21 50