https://store.steampowered.com/app/2795660/Yume_Kakigori/?l=japanese

Serious Strategy Games tend to spread their scope across the three standard theatres of war: air, land, and sea. Zero-K follows suit, despite being markedly more gamey and fun-loving. Nowhere is this clearer than in the interesting application of realism to planes and gunships. The real world has some influence over aircraft behaviour, and very loosely informs their support role, but it is just a starting point. Our priority is for air to be fun and useful throughout the game, as too often it can end up being a bit one-dimensional, or just be relegated to gambits.

First things first, how should air units behave? We were guided by rough intuition, so if it looks about right, then it is right. Real planes exist, but tend to be too high and fast to see, so the average experience of flight probably comes from birds and insects, with a bit of influence from helicopters and drones. These things are fast, so air in Zero-K is fast. Common experience also motivates two types of aircraft movement which, as with ground, we use to sort aircraft into their two factories: Gunships and Planes.



Gunships are akin to helicopters and drones; they fly around, some are quite nimble, and they can all hover in place. This is the standard RTS treatment, and it is sometimes criticised for being unrealistic or for making air and ground combat play out in much the same way. That said, ground combat is solid, and gunship-style air units across many games distinguish themselves by being faster than land units, at least on average. Gunships are also great at attacking from any direction, and avoiding obstacles when trying to retreat.

Planes fly like real aeroplanes and come in three flavours: bombers, fighters, and scouts. They only stop when they land, and have wings, ailerons, and rudders. There is an underlying flight model with the aforementioned parameters, and more, which we use to dial in the realistically uncontrollable feel of planes. Complete Annihilation tweaked these parameters to make planes more controllable than in its predecessors, yet plane-on-plane combat is still quite hard to control, to the point where we avoid expecting people to click on individual fighters, as they are just too fast.

Fast units are troublesome, since they inherently dominate the early game. Speed is just an amazing attribute, especially when there is plenty of free space to move around. Aircraft have it worst, being by far the fastest units, so a lot of air design is focused on how to make them unusually bad early on. Without skewing air away from the early game, we would see it early, then never again.



Planes were nerfed early by requiring bombers to reload after each run. Prior to the change bombers could fly circles over the enemy base, continually bombing it, provided there was no anti-air (AA) in place to shoot them down. This would only realistically happen early in the game, most likely after a surprise attack. Reloading also drains energy, which makes it even more costly early compared to later on.

There is no clean solution to the early game for Gunships, but they are not quite as fast as Planes, so the problem is lessened. Part of their solution is their terrible stats: a gunship will often cost twice as much to move twice as fast, but perform the same, or worse, than land units. Many of them are also awful against raiders, and a bit too expensive to comfortably rush, which adds to their early vulnerability. The light gunships are still in an awkward spot though, and it is made more awkward by AA.

Anti-air is a necessary, yet slightly unfortunate, part of air design. Flying units benefit greatly from Physically simulated projectiles since it lets them avoid most of the damage from a lot of weapons. Even beam weapons, which hit instantly and precisely, either deal constant low damage, or have a sustained burst that only hits fast units for a fraction of their damage. Plus, beams have spherical range, so are bad at shooting up. So Zero-K has units that can only shoot at aircraft, called AA or dedicated AA. This is the largest exception to the idea that units can shoot in any direction they like, but it is a tradeoff that we made to free up the design space for aircraft.



In addition to dedicated AA, Zero-K has flex-AA. This is not a well-defined set of units, rather it is the idea that many units are decent against aircraft, especially if they fly directly overhead. Flex-AA is particularly good against gunships, since they are slower than planes, to the point of making dedicated AA redundant in some cases. In fact, we buffed gunships against dedicated AA on a few occasions. Back in 2010 we reduced plane health and anti-air damage by 25%, which is equivalent to making planes weaker to flex-AA, and gunships stronger against dedicated AA. This was repeated in 2014 with a reduction of 1/11th, since Raven had 1100 health at the time. There is a limit to this approach though, because at some point the most effective flex-AA starts destroying planes.

Dedicated AA poses a potential problem: it is monospammable if you are only facing air. We solve this by declaring the problem to not exist. Planes and Gunship are deliberately designed as support factories, they are not meant to hold a front on their own, and should be unploppable in 1v1. Of course they technically can be plopped, as to block a plop based on player count would be too arbitrary, so all we can do is discourage them. Gunships were a particular problem for 1v1, since forcing out a bit of AA then quickly switching to land confers a massive advantage. In addition to making their units terrible early, we also had to make their constructor too slow and expensive for early expansion.

Air factories are not like other factories. While land factories are a somewhat self-contained and flexible set of units, air factories only make up one side of a counter triangle. Air beats ground armies, which beats AA, which in turn beats air. So if the enemy makes a lot of AA, then the air player might need to switch to ground to kill the AA. This turns air into a force that escalates the game, as it adds an extra layer of interaction, similar to the relationship between nukes and antinukes. This makes air a bit less of a faction, and more of a tech choice, at least in 1v1.



Even as a secondary factory, built later on to escalate the game, Gunships have had a problem with viability. This is understandable when you realise that Planes offer the fastest units, the best scouts, and can execute devastating surprise bombing runs. Gunships are just a bit faster than ground units, can attack from anywhere, and have terrible stats. The scouting issue in particular motivated the addition of Sparrow: a light scout that can be produced by morphing a radar tower. This gives all factories access to scouting, so adding another land factory is more viable as well, which is a neat way to see cross-factory synergies.

In the end, air is not the primary domain of Zero-K, but it is constantly trying to break out and become the whole game, rather than just be a fleshed-out corner of a counter triangle. This is a lot to put on only twenty or so units, and as a result they are some of the most tweaked and reworked in the game, with Raven out in the lead. Specifics may come in future posts, as they would make this one three times as long.

Index of Cold Takes

Fixed an issue where input was not recognized with certain controllers.

Fixed missing translations in some languages.

• Fixed several typos

• Fixed reports from the “future” showing up (yeah, that was weird!)

• Fixed text misalignment on some files/folders

• Added info about how to open the menu

• Improved report system description and added clearer rules in the guidebook for better understanding

• Plus a few smaller tweaks

(Mac version is now up to date with the others)

- Temporarily disabled undoing on Fork Solitaire and Time Travel Solitaire until those are fixed

- Added a margin to all solitaires when playing fullscreen to fix visual bugs

- Clarified rules of Solartaire (empty planet slots can be used from any orbit)

- Fixed Lock-and-Key solitaire not letting you put a card back after swapping it back and forth

- Fixed Council of Secrets not having a setting without Roman numerals

- Fixed (hopefully actually this time) Royal Flush Solitaire deleting your streak when restarting after a win

- Tried to address multiple bugs related to cards going behind other cards incorrectly

- Fixed an undo issue in 52-card Solitaire

- Fixed being unable to place a transmuted card back into its slot (in cases where it hadn't been placed elsewhere first)

- Added the ability to recombine torn cards in Tear Solitaire if the torn half hasn't been placed elsewhere yet

- Fixed Murder Solitaire claiming it can be played with a standard deck

Early Access has begun, and I am exciting to bring this educational software to the Steam community. Changes made in this update:

• Expanded the text on several verbose questions
  
• Expanded the anti-repetition filter groups
  
• Tweaked the laurel follower to better position around toggle boxes

During Early Access I plan to update at least once a week, but I felt this one was important enough to be implemented without further delay.

Message from the Team

Even though it's been over a year, the Chaotic Hearts Team has not given up on our first project and has been working hard to implement what we can, despite life's hardships and the challenges of a shrinking team. With that said, we are almost finished with several future items to be implemented in our next update. Please stay vigilant and patient as we continue to do what we can when we can. Until then, destroy enemies and level up while exploring everything our newest version has in store for you as you let Chaos Rain down upon your enemies!!

Change Log

• Enemies will now spawn continuously through all Maps.

• Safe Areas added for towns

• SW England now has repeatable Caves and Temples

• All maps are completely designed and explorable

Works in Progress

• Adding repeatable spawners to all other maps for the caves and temples

• Changing enemies to align with the correct caves and temples

• Enemies will grow according to player stats and levels

• Quests and story finalization

• Finalizing upgradable weapons and armor

• Adding Bosses and Mini-Bosses to Caves and Temples

• Adding mini bosses to the continuous spawn pool

Known Issues

• We are aware that some save files are being transferred over from test runs, and we are working on a way for you to delete old saves as well. If you are new to the game, please go into the pause menu and restart the level after pressing continue in the main menu.

We are sorry for such a long delay, but health and life have made it difficult for our small team to progress at past speeds

- KaozRain(Freddie) and the rest of the Chaotic Hearts Team

The main UI got again a major overhaul: many buttons have been grouped under menu-buttons to make the UI more clear. Variations are now always picked using the variation browser, which now shows the influence of a newly selected variation in realtime.

Also, there was a nasty bug, which causes the image to turn black when editing certain fractals. This has been fixed,

Additionally, there are 50 new variations, some of them are brand-new and were supplied by Brad Stefanov. Thanks a lot, Brad!

There were really a ton of changes.

Complete changelog:
VERSION 0.108.1107 (13.07.2025):
- variation-browser: after selecting a new variation it is immediately applied to the
current fractal (temporary), which makes it really fun to brwose through all variations
- main-editor: variations node: removed the variation-type and variation-listbox
(as now one can use the variation brwoser to brwose though variations, so there is no
need for this two listboxes anymore. It makes the UI more clear, too.)
- SWAN-59: fixed: in some cases (fractal not contracting) the image faded to black while editing
and kept black even if the flame was brought back to a state where it should render properly
(e.g. when moving xforms)
- reimplemented hiding of xforms (was broken after implementing propagation of hidden
samples (which are hidden by variations like crops))
- hiding of xforms now also works with textures (image_rect-variation)
- new /v1/variations-REST-endpoint to list all currently supported variations (which
will be used later by JWildfire)
- main-editor: several ui-refactorings, introducing more spaces between controls (SWAN-78)
- gradient-library: introducing more spaces between controls (SWAN-78)
- render-image-window: introducing more spaces between controls (SWAN-78)
- render-animation-window: introducing more spaces between controls (SWAN-78)
- custom-code-editor-window: introducing more spaces between controls (SWAN-78)
- mutagen-window: introducing more spaces between controls (SWAN-78)
- gradient-editor-window: introducing more spaces between controls (SWAN-78)
- load-/save-flame-window: introducing more spaces between controls (SWAN-78)
- main-editor: decraesed the font-size of the labels of the flame-thumbnails, also
changed to alinment to bottom
- new "de_stijl"-variation, provided by Brad Stefanov
- new "drunken_tiles"-variation, provided by Brad Stefanov
- new "flora"-variation, provided by Brad Stefanov
- new "meeple"-variation, provided by Brad Stefanov
- new "multi_ifs"-variation, provided by Brad Stefanov
- new "natural_foam"-variation, provided by Brad Stefanov
- new "shape_warp"-variation, provided by Brad Stefanov
- new "square_rand"-variation, provided by Brad Stefanov
- new "tesselated_t"-variation, provided by Brad Stefanov
- new "cut_triantess"-variation

VERSION 0.107.1082 (10.07.2025):
- main editor: automatically open the variations browser when adding new variation
- main editor: removed the scroll-container at the main buttons panel, which sometimes
masked a button-click (which required two clicks at the button)
- main editor: edit xforms: "reset to defaults" now respects the edit plane (xy, yz, zx)
- main editor: edit xforms: new "flip horizontal" button
- main editor: edit xforms: new "flip vertical" button
- new "cut_sincos"-variation
- new "cut_kaleido"-variation
- new "cut_magfield"-variation
- new "cut_metaballs"-variation
- new "cut_hextruchetflow"-variation
- new "cut_randomtile"-variation
- new "cut_rgrid"-variation
- new "cut_shapes"-variation
- new "circlerand"-variation
- new "circletrans1"-variation
- new "cut_snowflake"-variation
- new "cut_spiral"-variation
- new "cut_spiralcb"-variation
- new "cut_spots"-variation
- new "cut_sqcir"-variation
- new "cut_sqsplits"-variation
- new "cut_swarp"-variation
- new "cut_tileillusion"-variation
- new "cut_pattern"-variation
- new "cut_glypho"-variation

VERSION 0.106.1057 (09.07.2025):
- main editor: grouped the several add- and edit-functions under two menu
buttons to make the UI more clear
- main editor: grouped the main buttons on top under new menu
buttons ass well
- main editor: changed the position of the thumbnail-view, which is
now below the main buttons
- load-flame-dialog: show file-path and file-name of the selected flame
- main-editor: when loading a flame remember the file path, so that when saving
the same flame no file-save-dialog shows up
- renderer: fixed: propagation of hidden samples did not work correctly
- new "c_symmetry"-variation
- new "checkerboard_wf"-variation
- new "pre_dcztransl"-variation
- new "cut_fingerprint"-variation
- new "joukowski"-variation
- new "cut_alientext"-variation
- new "cut_booleans"-variation
- new "cut_apollonian"-variation
- new "cut_bricks"-variation
- new "cut_btree"-variation
- new "cut_celtic"-variation
- new "cut_chains"-variation
- new "cut_circdes"-variation
- new "cut_fractal"-variation
- new "circlelinear"-variation
- new "cut_fun"-variation
- new "cut_hexdots"-variation
- new "cut_mandala"-variation
- new "cut_jigsaw"-variation
- new "cut_kleinian"-variation

Added some finer details. Grass and flowers and such. Also various adjustments to the island course. Tee pad heights, par changes etc.

Bug fixes and UI improvements.