Stranded III Dev. Blog - Comments
Offline
Also I find everything very beautifully designed!
Forum
Projects Stranded III Dev. Blog - CommentsAlso I find everything very beautifully designed!
In addition to shape, dispersion of influence would be another good setting/feature to add. It would control how the effects of the brush (painting, height adjusting, terrain changing) are spread and dispersed within the brush area. This could, for instance, decide whether raising the terrain in an area would form a spiky mountain, a gradual hill, or a plateau.
Also, the array of green dots that represent the brush could each have their brightness represent their relative influence, thus in the case of very centralized dispersion the dots in the very center would be much brighter than those outside, while a more gradual dispersion would have brightness fade out more gradually closer to the edges, and in the plateau setting all dots would be the same brightness, as the influence would be uniform.
Also, smoothness of brush dispersion could be a option as well, to be able to change from more of a grainy 'spray paint' range to a smooth 'fountain pen' range, and all the areas in-between.
The ability to choose the magnitude of influence would also be very useful, as it would be able to control the sensitivity of the brush.
Of course, some people may want to save their settings once they get them perfect, so the ability to save and load brush settings would be convenient.
Anyway, just a few ideas. I hope they help.
UPDATE: also, the inability to pick/destroy flowers in-game in Stranded II was an annoying feature that should not be repeated for Stranded III. Also, Stranded II Titanium has a feature in editor 'areas', a derivative of which would be quite useful in Stranded III. It gives the ability to place specific types of plants with the editor paint brush (ex. herbs, flowers, trees, stones, stones(ore), bushes, vines, etc) which could consist of a single type of entity or specific groups of entities. Other groups, like jungle floor or sea floor, could also be included. Units and items could also be included if you so desired, if for whatever reasons it is required. Also, the ability of custom making area types and saving them would improve flexibility the entire system, and could result in improvements over existing.
Water logic (the implementation of calculating river flow, lake size, and other aspects of how water acts and affects things around it) is something I personally have never seen in any game I have ever played, but something which would be very valuable if successfully implemented (ex. by using water volume estimates, flow estimates, and evaporation level, etc., obviously limited by practicality of course). It carries the possibility of improving many aspects of the game dramatically. Also, if underground water flow (ex. underwater rivers, the water table, soil humidity) is calculated, it could result in the ability to position plants and animals better, so monkeys for example do not thrive in the desert, and neither does moss. Also, the paths of fish could be calculated for rivers, and unlike Stranded II, you could make it impossible for fish to travel under land, and instead in lakes, rivers, and the ocean. This would also open up the possibility of saltwater and freshwater plants and animals, in oceans or lakes and rivers respectively. Implementing this may take a bit of time (browsing GitHub and game dev forums) but the results would easily be worth it.
Also, creating a modular building system for in-game building would be worth considering. Specifically, one that builds things piece-by-piece. It would be a lot more elegant than the Stranded II method of right clicking until carpal tunnel sets in. If the modular building method included the ability to design modular buildings in editor, it may make it especially easy to design buildings that can be made, and the buildings would take little storage. Of course there should be rules to when designing a building to prevent obvious structural instability, which would eliminate most horrible or unrealistic and impractical designs. Also, things could be created to extend on a previous building, and thus improve them, or to lower requirements of storage for buildings. Different methods of pooling resources could be used when building a building out of single construction objects (ex. logs, rocks) in order to prevent inefficiencies, especially by pooling all resources in single building structures, and consolidating it as one object. Also, you can check out Cast Away by
Hurri04 (the German version works better) and get ideas from that, but of course extend on that, as it is fairly simple.Edit: Extension mod also has a modular building method, which would be useful to check out for ideas. It isn't perfect though. The main problems not due to the system, but due to the fact that the models aren't correctly aligned.
edited 4×, last 17.06.22 05:10:39 am
OnlineHere we are 10 years later, and I'm still checking in now and then.
I know how hard it is to keep up with a project that doesnt seem relevant to you any longer, with work and more important things on your life. Its been some of us's dream to see this come to light. It was my first survival game and it sparked interest in me to go out camping.
Also, when its out, I'm still going to translate it.
Sabowulf, Hah! I was using an IRC client built into CS2D and told DC to create Stranded III. I don't remember what he answered, but right on the next day the "Stranded III" page on unrealsoftware.de appeared. Here's a chance that nobody would even believe me but I have only one friend who can confirm I'm not lying. 
DC. You could also add options to allow a certain amount of randomness in scale, color, and rotation respectively every time you put something down (so you don't have to change things afterwards, and because it would make things look less homogenous) hopefully without causing too much performance loss. Also, another idea that struck me is the possibility for objects to be co-dependent, for example, a house which has the door as a separate entity (because it saves space and makes so you can't open the door from behind the house) and when the house is moved or scaled, the door is as well. When the house is deleted, so is the door. Having the choice to do this one or both directions (ex. house deleted deletes door, but door doesn't delete the house, it just leaves a house without a door. Or have the door delete the house as well) would open up a lot of flexibility. This just makes sense, as it fixes lots of problems, saves space, and improves flexibility (for example, re-using the same door on another house, or even making houses made of multiple objects) and could even be extended on by using the same system for building in-game, to be able to make modular and customizable structures. This additionally opens up the possibility of having an editor where you can make things ("structure") out of objects, which is saved as all coordinates, scale, and rotation values for objects.Another possibility is making it possible to rotate, scale, or move the entire structure as one object, even if all of its different parts act different (i.e. doors open/close while walls do nothing).
In such an editor there would be options like: "create new structure", "edit structure", "save structure", "import structure", "add object/item", "place unit spawn".
A few options to have with the regular editor could be whether units animate in editor or not (in regular s2 they do, in s2 titanium they don't. Having the option to choose would be convenient), which animation is played (idle1,idle2,walk,die,etc.), the ability to unpause/pause without testing (basically, paused would be default, and unpaused in editor would make it work similar to creative mode in Minecraft) which would make everything either frozen in place, or having things act the same as they would in-game.
Anyway, that's all for now.
ModJuicer: Thanks for the feedback!Post from June:
Different brushes & more brush settings are a very interesting and nice idea. I'll think about it. Would certainly be a great addition.
Flowers will be actual objects you can interact with in Stranded III. Decorative ground foliage however (mostly grass and tiny plants) must be non-interactable however because it can be reduced/disabled in the graphic settings for better performance.
An "object placement brush" / "fill area with objects" is not implemented yet but planned. Internally this logic is required anyway for procedural/random map generation. My plan is to make it possible to use biomes which are defined for the map generation but also to define custom lists of objects which should be spawned (along with weighting etc).
I agree that the water stuff would be cool but Stranded III is not a simulation and such a detailed system would be too complex. Maybe some aspects of it will be in the game though - in a very simplified way.
A modular building system is definitely planned but not implemented yet.
Post from August:
Random rotations and scales are already implemented.
See this blog post for details https://stranded3.com/index.php?entry=97
I won't implement random colors because of the performance impact. Would still be possible to make it happen with Lua.
That object dependency is something that even Stranded II already had. E.g. fruits hanging on trees which will fall when the tree is destroyed or the tree house which will be destroyed when the tree is removed. A comparable, probably more sophisticated system will be required for Stranded III as well but is not implemented yet. Nice idea to be able to load/save collections of objects / structures though!
Unit animations: Interesting idea. I don't think that it's an important feature but certainly nice to have.
The buttons/feature you have mentioned are actually pretty close to what the editor looks like
thanks for the ideas! 
And now we get a some fancy file systems. This could have other functions too, but what those are depends on how it's made.
, external features (like hair, shoes, clothes, etc.), things like tusks that could be broken, and a vast assortment of other stuff. The idea just came to me, and it seems like a simple fix to a lot of problems, if it can be done. It could also allow you to attach colliders to specific parts of units to make collision seem much more natural, or having units consisting of more than one piece (for example, a ship). I wish you luck 
edited 3×, last 23.09.22 05:39:38 pm
Also, your Hold-Keys feature made me think of another possibility. Suppose that, to allow for even more potential actions, you could introduce the possibility of Key-Combos. For example, Q+C could be Quick Crouch, which would crouch for a short period of time before automatically getting back up. Also, it would be quite useful when using Lua scripts to add extra functions to the game.
Side note, you managed to implement all the features I was attempting (and failing) to implement on my mod. Quite impressive.
ModJuicer: Interesting idea... key combinations actually already exist and can also be displayed by the control info UI. Right now they are only used in editor though and they are also fully hard coded so they aren't accessible via Lua.Not sure if I'll change that. I think it's better to avoid complicated combinations for game play input.
Additionally, there was another potential feature I had forgot to mention in the previous post. If you could do Key-Combos, it wouldn't be hard at that point to add code words that trigger specific things when typed up in-game, which would be ideal for easter eggs and other fun little features.
I also had another idea to do with building. I like the way minecraft has a modular approach that allows you to make all kinds of things using a subset of smaller parts. Doing that in a game that it isn't dominated by voxels would be cool.
EDIT: I just got a new pc with the ability to run a lot more recent games (also, the previous user did not remove the archive of pirated games they had installed, so I was set to hame). It was fun at first, as most novel experiences are, but I quickly came to realize some annoying design choices within these games. There were too few options in many of the games. As much as a good immersive experience can be fun, if I can't do all the things I can do in real life, I would rather immerse myself in reality. The game menus were made to be simple and fast to use for that sake, which reduced functionality too much for my standards and sometimes just made it more confusing, counterintuitively.
Also, while I was playing one of my new games I made a realization: an important part of game design is making the player feel like they're in control. Many games fail to do this, sometimes in an effort to make the game dynamics more realistic. However, first off, this is a slippery slope for the developer as after the first time they will then justify reducing further control the player has in the name of novel features, and it tends to undermine the feeling for the player that they are in control of the game, rather than some authoritarian developer code. (not to discourage novel features, of course)
Also, even more important in games than books, in my experience, is the application of chekhov's gun (https://en.wikipedia.org/wiki/Chekhov%27s_gun?wprov=sfla1). If there is an item that cannot be used, that's bad design. The player will try constantly to figure out what it can be used for and end up irritated. This also applies, to some extent, to items that are only usable for very specific/uncommon uses. In the latter case, an easy fix is to have more ways in which it can be used.
I think there is often within game developers the urge to control as many aspects of the game as possible, which can come at the expense of the player control. It is important to realize which features cross the line between realistic or novel design and player autonomy.
This is wisdom gained from experience with other games, however, and Stranded III seems much more open. It is more of a cautionary tale of what can go wrong.
As far as player control is concerned, it would be cool to see a system for modular/dynamic buildings and objects. The ability for players to create custom, working, functional systems out of interacting objects, especially systems that make sense physically, would make the game much more popular with creative people - the same kind of people that make mods and help drive development.
I also still think having both 1st and 3rd person mode would be better. User preference is notoriously unpredictable, and overall more people will play the game. I have no real preference regarding 3rd vs 1st person, just a preference for having a choice between the two. 1st person should obviously be default.
What is your take on these ideas?
Also, have you checked out my concept art and the underlying ideas further than just a quick glance? I would like to know your thoughts and opinions on them in that thread, especially as I'm always coming up with ideas (most of which I forget before I can write them down) and it's nice to have some insight into which are good and which ones are not, and why, to improve my future ideas.
EDIT: For inspiration for modular/dynamic objects you could include, see simple machines, the history of machines, and machine elements (https://en.wikipedia.org/wiki/Simple_machine, https://en.wikipedia.org/wiki/Machine#History, https://en.wikipedia.org/wiki/Outline_of_machines#Machine_elements)
edited 7×, last 08.05.23 06:37:10 am
Or whether control from the map editor will be fixed. Its problem is that it didn't fit with the remapped keys, which makes it difficult to create maps,If the player's standard control keys do not work (WASD).
I had an idea a while back about being able to change how evenly the brush on the map is dispersed. Of course, while drawing I came up with a few more ideas. Note that every setting on this drawing has a thought-out purpose, including every terrain modifier button.
Here are the purposes of each:
Brush Shape: Self-Explanatory.
Zones of Influence: Decides how much effect each area of the brush has. Shows center, middle, and outer, but any number of zones could be used. In fact... Why not make an option that decides the amount of zones? Anyway, an example of this concept: If the outer zone had 100% and the rest had none, it would essentially have a ring-shaped influence. Besides the zones, there is a slider labeled "all" which would basically decide the overall sensitivity of the brush.
Zone Transition Abruptness: Determines how clearly separated the zones are. Using a square brush with a more stepped transition, for example, one could make some fancy pyramid-looking thing. Doing the same thing with a circle one could make something resembling a tiered cake. Using a completely smooth transition, it would just make a hill.
Noise: If noise is added, the brush will paint with random noise.
Grain Size: Controls the size of each random variation that constitutes the overall noise, from small to big.
Granularity: Controls how smooth or rough the grains are in the noise.
Color: Self-explanatory
Terrain Modifier: The list of different things the brush could be used for. Each of these 10 modifiers shown have a distinct purpose. I will describe them from 1 through 10:
(1. Flatten Terrain
2. Raise/lower Terrain
3. Smooth terrain, or round of terrain.
4. Place objects (will request information in the box labeled "Modifier Settings" to get the type of object, along with any other necessary details).
5. Effect Brush - Requests the specific type of effect or even several effects, if desired, be provided in the box labeled Modifier Settings. Once the effects are provided, you can use the brush to add these effects (fire, lightning, poison, etc.) to everything within the radius of the brush.
6. Script Brush - Works similar to the Effect Brush, but requires a specific script, which it will then be added to anything you use the brush on, so long as it lies within the radius of said brush.
7. Object modifier - Effects all objects within the radius. Can scale the objects, and maybe move these objects in X/Y/Z and Yaw/Pitch/Roll directions if you want to enable that in the Modifier Settings box.
8. Hmmm.. So much time has gone by I forgot what this one was for. Something to do with objects. Maybe it was meant to be a tool to use the brush to shade objects within it's vicinity using the color chosen in the color option. Maybe, Idk for sure.
9. Water color, literally - A tool to make different patches of water. Maybe green to signify algae, brown to signify muddy water, etc. Could also make some parts of the ocean harder to see through when underwater if you are able to implement that kind of stuff. Water coloring could, technically, also be used during island generation to make islands seem more realistic, interesting, or aesthetically appealing.
10. Delete/Erase - Can be set to delete any individual specific thing or broad category of things within the brush area. What it deletes may be based on the Modifier Settings enabled by the user.)
Modifier Settings: Shows custom settings for whichever modifier is enabled to allow functionality.
Feel free to use any of these concept ideas, improve on them, or just mess around with them for fun. They were inspired by the idea that "What if instead of a bunch of different brush types we just had a single brush that we could customize ourselves?". Maybe also the abilities to save custom options would be cool too. And its possible you may even come up with more features that fit alongside these.
What do you think? Would a bold step into uncharted territory constitute an unacceptable deviation or a significant breakthrough? I personally was motivated partially by how inefficient it is to just have a bunch of different, individually created brush types when you can just have sliding scales that determine the brush type. Also, it grants far more freedom when you can choose exactly what you want your brush to be like. The majority of people ignore the majority of brush types anyway, and would probably welcome the new method as a much needed paradigm shift.
That is my opinion, anyway. Good night. I'm overtired and cannot continue on in a state of sleep deprivation, so until next time..
Grekklek: Hopefully he finds some cool AI/ML techniques that can actually be implemented in Stranded 3 without blowing everything up. Grekklek: Natives and other NPCs are planned. They won't have "real" AI though. Just simple state machines like in Stranded II. They will be a bit more complex though which should also make them look a bit smarter.Not sure what you mean with the editor controls. Right now I don't plan to make those configurable.
@
StirlizZ-Fapicon: thanks 
@
ModJuicer: Thanks again for the great input and sorry that I don't manage to comment on all of this!Some of your ideas match what's already implemented:
Flatten/Raise/Lower/Smooth
Place/Delete Objects
(this still only uses radial brushes though. It can however fill with a specific object, a list of object with chances & rules, or even a defined fill logic for a biome. so you can basically fill any area of the map like a specific biome would be filled)I really like your ideas for effect/modifier/script brushes! Pretty cool idea! This made me think about another type of "script"-brush: Run a script on every object the brush touches - directly in editor. This would give you ultimate freedom. There could be presets for scripts which do different things. The script would receive a variable which is the ID of the object that was hit. It could also receive values about the brush intensity at that point, distance to the brush center, brush size etc.
Regarding zones of influence / noise:
That's a nice approach in theory but I don't think that there are many use cases for this in practise. Noise does make a lot of sense but the zones of influence don't really help much. I think it's super rare that you have a scenario where you want to paint rings or where the inner part of the brush has less influence than the outer one etc.
Actually I have a concept which is comparable to your zones: The brightness of the pixels in the brush. The brighter, the higher the influence. By reducing a "softness"-slider you can turn up the brightness of all pixels which aren't entirely black which then makes the brush sharper.
Noise doesn't exist right now as a general brush setting but for color painting there's a spray paint mode which uses noise.
DC: I think what Grekklek is talking about is the somewhat confusing way map navigation works on Stranded II. When I re-downloaded it (when I was 11, after remembering how much fun it was back when I was 4-5 years old) it took me several days to figure out that you had to right click and hold to change the angle of the camera in editor. I figured it out eventually, but if you think of a more intuitive way to do it in Stranded III it would probably be easier for newbies to use.About the zones: I used the ring example, as well as the pyramid and tiered cake example, to serve as a disambiguation as to how the zones would work. I agree, though, that it is unlikely in practice for people to use rings much, though they may be used a bit
The idea was conceived mainly for the benefits it introduces regarding the ability to form terrain properly, so the desired end result can be achieved. I have spent enough time on Stranded II (and I know Stranded III is a bit different, but hear me out) trying to shape terrain that ended up misshapen and weird looking, and it inspired me to propose an improvement. The zones settings could determine what kind of contours the brush shapes in the terrain, and it could be useful for precision terrain modification.
Another feature for it, and a feature that would make it more functional, would be the ability to change the radius of the different zones (not allowing overlap, of course, possibly by requiring a minimum distance allowed between them). This would mean more flexibility. It would also, generally, require less zones to get the desired effect (if one were to allow an option which adds more zones). This, of course, would exclude the very outer zone, which is going to be exactly as large as the brush itself.
Remember - the more control and functionality you give to people, the more they enjoy it. Plus, I personally thought the zone idea was one of the best parts of my idea, because of how unique it is as a whole, and how universally applicable it felt, like something that could catch on other places because of the simplicity, and yet the functionality offered by it. I especially hope it would catch on with those horrible photo editing suites that have way too many options. It basically turns things from a million options, none of which are guaranteed be exactly what you want, to a gradient which can be navigated intuitively to get the desired result.
Brightness and softness, while having some similar functions, just don't provide the same functionality that this system would. First - the settings aren't as intuitive or hands-on - they're just a few sliders. Second, they cannot provide custom contours and allow for minor parameter adjustments. Third, of course we all want to be able to make instant-volcanoes. That would be awesome.
About the noise, I came up with that idea because I wanted to be able to have any variation from the extreme of full-on graffiti-optimized high-splatter spray-paint to smooth felt-tipped markers. It makes more aspects of it controllable.
Anyway, I am almost sure option 8 was the ability to color objects. I wonder what would happen if there multiple options were allowed at once?..
Also, I had an idea. You could include brush options where you can whitelist, or blacklist, different things (based on the position of a whitelist/blacklist switch). In the case of coloring the terrain, you could make so it doesn't color over a specific color, or that it only colors over a specific color, to make things easier. In the case of deleting objects/units/items, or whatever entities are labeled in S3, you could also blacklist or whitelist specific things.
...
And, another idea.. What if there was not just a circular brush influence option, but also a sphere option? In fact, you could try out all kinds of 3d shapes along with the 2d ones.
Part of my idea about having an optional(!) 3d brush influence area is that you could have a set distance between the influence area and the camera position, so things can be affected far from the ground.
...
In fact, that could just be a global editor positioning option: Whether the target point is set a specified distance from the camera, or whether the target point is set at where the mouse point coincides with the terrain. Or, where the mouse point coincides with the water.
And, the distance the target point is from the camera could be changed by scrolling or some other mechanism. Maybe shift-scroll or something. And, it may work quite well to supplement the grid system in high places, for when you need to put something down that doesn't align with the grid...
..speaking of which, did you ever implement the Y grid option? It is a critical part of functionality of the grid system, and I don't remember seeing it.
...
Also, another idea I had (man, I'm just full of these) is to make so the behavior system is separated into two systems - an simple logical input-output system, and the system that makes the decisions based on the inputs and outputs. The reason being that it is easier then for other people to create their own AI plugin or script, and theoretically it could run on basically anything. Both systems should be able to be modified, of course. It would make it easier to adapt outside AI code into the behavior and so on.
...
I'll see you around. I must end this response before I get another idea, or I will be typing forever.
Thanks again for the ideas! Some interesting stuff there!
T-69: 
I chose the stylized turd because it's a well known icon/emoji that is easily recognized by everyone. I don't see problems with this as long as there is no other shit related state. You are of course right that "watery excrement" would be more accurate - but it would also be harder to recognize what it is.One fact I didn't mention in the dev blog is that all these icons will be colored in-game so I could (and will) choose a brown color to emphasize what it's all about. So if I need to make more shit states the color will certainly help.