Itch don't allow as is to copy the devlog with many gif-demonstrations. And also I am not able to translate 227 pages of text into english. Therefore, this post is a heavily abridged retelling of P.E.G.A.S.'s journey from the R.F.C. prototype to the long-built Royal Flying Corps. Here is a brief disclosure of what was planned to be done and what was done in the end.

R.F.C.

(October 2018 - December 2018)

The original concept of the game was to drive a Retro Flyable Car (hence the working title R.F.C.) through a retro-futuristic metropolis overrun by chaos to catch up with and run over Discord. For the first half of the game you have to dodge viscous clouds of cotton candy; from the rain of cupcakes, which throw off the direction of flight; from levitating pieces of earth, which are possible to fly around only through the interiors of skyscrapers; and from other environments. The second half is about trying to ram the boss with yourself.  In the process, it turns out that the spirit of chaos made a mess to restore its magic, so that it can extend the life of a very old Fluttershy for a few more moments.

On paper, everything looks easy to implement, but then, in the fall of 2018, in 3 months of game dev courses teamwork with three developer friends, only a couple of assemblies were implemented and such a raw techno-demo that it didn't survive to this day.... Ironically, we got so caught up in the crunching that we didn't even notice when the deadline happened.

I think there are two reasons for this fiasco:

1. We've separated the roles of programmer, designer, and game designer too rigidly. If all four of us were to puzzle over the implementation of flight physics in UE4, we might have come up with something. If I personally stepped out of my comfort zone and got into programming, maybe our programmer wouldn't passively read the documentation.
2. UE4 proper. It is too demanding on both hardware and brains. Even if more than one laptop could pull it, there would still be problems with deploying the repository to GitHub. You'd either have to roll an LFS addon or deploy to Azure.

For all of 2019, I just fleshed out concept docs and modeling. At the very end of the year, I worked with Unity as part of one of my coursework.

P.E.G.A.S.

(April 2020 - September 2020)

In the spring of 2020, I entered the Sixth Herd Gaming Contest. The deadline was twice as long, but the team was three times smaller (my buddies-developers were already employed and could only consult). On experience after game-dev courses, I already understood that even for six months I will not be able to cope with the physics, and modeling retrofuturistic metropolis, and rendering of cutscenes. So the concept was simplified to a pegasus delivering an important package to the ancient times of Discord's rampage.

After a month of development, the prototype was ready. Not because it finally got the physics out of Retry, but because with each subsequent change, it only got worse.

Somewhere in the process, the idea of replacing the abstract pegasus with the princess sisters who go on their first adventurous misadventure to overthrow an epic monster dawned on me. Shift the focus from maneuvering to containment of magic, you can cast fireballs to clear the way, but if you cast too much/frequently, the princess sister will go mad and turn into Nightmare Moon/Daybreaker, sowing speech into the boss. Bet on localized co-op, when the sister is played by a friend sitting next to her. Well, and the whole game to submit as a diary of princess-sisters, where each menu is designed as a page, and each level - a chapter of their lives. In the beginning, they overthrow Discord, establishing a diarchy. At the end, they overthrow each other, establishing a monarchy.

There was a synopsis: "The ponies live in a black and white world ruled by a white Princess Celestia, who gets all the glory so that the diarchy is worshipped, and a black Princess Luna, who takes on all possible sins so that the diarchy is feared; suddenly Discord appears and paints everything in the seven colors of the rainbow, where the moon is not black like darkness, but blue like night.". However, everything somehow got bogged down on the prescription of the specific game action.

I spent the next 5 months unsuccessfully poking into the code and into the text. The physics became less and less playable. The plot was getting more and more lost. The situation was saved by the traditional extension of HGC. Having rolled back the physics to the latest stable version and forgetting about the story, I started to work on the other aspects of the game, such as levels, menus, victory.... Actually P.E.G.A.S. Public Demo was made during these extra 2 weeks.

P.E.G.A.S. took the SHGC podium. 3rd place in both tabun and itch voting.

Royal Flying Corps

Alpha

(October 2020 - December 2020)

Being inspired by the success on SHGC, I managed to push through the updating of P.E.G.A.S. as a university course project. The deadline was again from October to December. Again there was a team of 3 people, though from classmates-developers. There was already a passable build. The concept was still the same (except that the final name Royal Flying Corps was coined, whose acronym refers to the original working name R.F.C.) Specific plot events in the level still hadn't been spelled out. We continued the strategy of the last two weeks of SHGC and focused on doing basic tasks.

ADDED:
> Local co-op, where the camera is centered between players and scaled in proportion to the distance between them. If one player crashes, the camera locks onto the surviving player. For now, just the ability to fly next to a friend
> Shooting fireballs. But there's no one and nothing to shoot those fireballs at.
> Takeoff, but not a landing.
> Ability to change the control layout. Gamepad support
> Chunked background loading. Clouds have been converted from static to dynamic
> Precipitation in the form of cupcakes. Thanks to chunk loading, they can fall throughout the entire map
> The faster the flight, the farther away the camera gets, more gusts (streaks?) of wind and louder the humming of this very wind
> Sounds of wings flapping and beats
> The theme music is individualized for each of the princess sisters

CHANGED:
> Flight physics. The pony now has a lift force, which is calculated based on speed and angle of attack. Now the direction of flight reacts to the pony's tilt. Now the pony does not fly nose up, but as it should. You don't have to tap constantly anymore, you just have to tap. However, at low speeds, for some reason, the pony's nose to the top, and the flight is too inert as on such a narrow and skinny map...
> The gap in the clouds that could be used to fly around the entire map has been patched. In its place, a shaft was dug. However, it turned out to be too narrow, and the new flight is too inertial. Flying through the main obstacles is much easier. The accidentally created gap in the clouds had a much better level-design than a specially made shaft
> The entire interface has been redrawn. Now it is adaptive. Sensitivity and volume settings have been moved to a separate menu-page

A lot of work was done, but the planned was not fully realized, so the resulting build felt much more raw demo than the original P.EG.A.S. Public Demo. Local co-op did not help, only prevented me from concentrating on his princess-sister. Fireballs also didn't fulfill their game-design intent and didn't clear the way. The level with narrow corridors and the need to brake sharply did not fit the realities of the updated physics....

On the long distance, the method of “just do tasks” was not very effective. Many different things were done, but they were different and didn't form a cohesive whole. Qualitatively the game did not develop much. We should not completely forget about writing the design-doc.

And yet the course project was safely defended.

Royal Flying Corps

Beta

(9 January 2021 - 23 January 2021)

While I was still working on a term project, the Chrismas HGC 2020 was announced, lasting from mid-november to the end of december. I had an idea for the final chapter of the princess sisters' diary:

"The 90th year of the reign of the sister princesses of Equestria. Radiant Hope, heir to the crystal throne, arrives in Canterlot. However, not for the expected start of her apprenticeship under the Princesses Sisters, but to bring news of a certain Emperor Sombra who has overthrown Princess Amora and appropriated the Crystal Heart. The Emperor's intentions to remove the curtain that protects the entire continent from the northern frost from those who hide in it.

Princess Celestia saw this as an easy diplomatic victory, Princess Luna saw it as just another monster. Both princesses saw it as a small victorious war. Both sisters underestimated Emperor Sombra, that he could just disappear with his Empire, removing the gateway between the Arctic Wastes and Equestria.

A blizzard accompanied the sister princesses the entire way back. When one princess fell, the other would pick her up, for they were interdependent competitors, for it was their opposition that created harmony between the individual and the collective good. For without the all-loving Princess Nightmare Moon would be declared a monster, and without bugaboo no one would listen to Princess Sunshine. At last they reach Canterlot, the first surviving city they encounter, the last hope of the entire kingdom. Canterlot must stand!

The city's supplies were not designed for so many hungry mouths. Canals and gutters clogged with shit. A plague broke out. Local and county strife coalesced into one global confrontation between the sedentary - who thought it was suicidal to migrate south, and the nomads who thought it was madness to stay in such conditions. This confrontation reached the princess sisters as well.

Celestia, as the eldest sister, remembered the fall of the United Kingdom, and was bitter and frustrated, but was of the opinion that the ponies should leave Canterlot now, so that tomorrow there would be someone to build a new one. Luna remembered only the horrors of the migration and simply could not leave the City of the Sun, which she had built with her own hooves.

Still, in between enforcing the city's order, Luna sent expeditionary corps out to explore. Celestia, in her spare time from reallocating resources, was doing Amore's job of raising Hope to be a princess. She didn't take her instructions very well, but Celestia was also remembering what a princess should be like.

Luna thought it was an unacceptable waste of resources that instead of rebuilding the Crystal Heart, Celestia was celebrating with Hope. 

The princess-sisters began a heated debate over Hope's usefulness, which resulted in the vidnigos breaking through the temporary dome powered by Celestia and stealing a test sample of the Warming Hearth from the guild."

This is where the gameplay of the final level begins. The princess sisters are chasing the windigos. They catch up, run over, get the stolen goods and try return Warming Hearth back to castle. One of the following options can be ruled out along the way:

• Both sisters use their magic within reasonable limits. Warming Hearth is returned, the sturgeons win, and Canterlot survives;
• One of the princess sisters casts fireballs too violently and turns into a boss. She is banished. Depending on which sister survived, the ponies either stay in Canterlot or migrate to warmer lands;
• Both princess sisters have gone on a rampage. They are both banished to limbo.

ADDED:
> Windigo fleeing with a stolen Warming Hearth
> Drop, pick up, pulling and delivery the Warming Hearth
> An invisible dome behind which cold death awaits
> Sister princess sprite mirroring when reversal

CHANGED:
> Now to win, need to return heart
> Localized cupcake rain converted to snowfall for the entire map
> Disabled the original map, which no longer fit the current flight physics

FIXED:
> The chunking system did not work on the return flight

It was much easier to control in the open. Fireballs finally had a place to go. Even the local co-op camera now somehow helped to catch up with Windigo, because in co-op the camera zooms farther, and therefore you can see farther.  The new build felt less crude, but it felt empty. Still, the level turned out to be even too simple and fast-paced. Critically lacked scenery-orientation. Accompanying vindigo that would attack the player. The ability to choose chapters/modes/levels...

And yet, I consider this build worthy of publication. It still realizes the basic gameplay of the final chapter. All the same, it has already been played qualitatively different than previous builds.

Alas, I was only able to start working on the Christmas update in January, when the deadline was already safely past. Moreover, right after CHGC 2020, the Big HGC started with a whole year long and very solid prizes. And Royal Flying Corps loses to the winner of SHGC already at the concept level. Combined with the stark loneliness after brigade work, all of this hit my morale hard.

After realizing the basic gameplay I found myself in a stupor. I had no idea how to implement vindigo behavior more complex beyond “just fly right”. I wasn't sure I could model the whole Canterlot, and I was pretty sure I couldn't draw it. In general, there were doubts that even after a year I would be able to do everything I had planned for Royal Flying Corps.

So I switched to a more persepctive long-build, whose prototype I wanted to realize since the time of game dev courses and to which all the other chapters of the devlog are devoted. On February 3, 2021 the active development of Royal Flying Corps was frozen.

In 2022, the original Retro Flyable Car is modeled
as part of an already different retrofuturistic pony game

On the eve of CHGC 2022, one of the participant shared an idea for a game where you have to fly, dodge obstacles, and physics gifts into chimneys. I thought it was a great expansion for the final chapter. 

While the princess sisters are delivering gifts to comfort the ponies, some vindigos try to disrupt the delivery of the gifts, and some windigos besiege the castle and try to steal the Warming Hearth. After successful delivery of the gift, the house will be lit up with lights, Christmas ambient will play and there will be no more vindigos sleeping above it. When delivering presents to all houses in the neighborhood, ponies will start playing on the streets and if the princess sister falls down near them, they will help her to recover and get up.

This also gives new opportunities for cooperation. With the friend will be twice as many princesses, lives and lasers. You can mop up two neighborhoods at the same time. You can resurrect each other. You could cover for each other. You could divide up the duties, like one driving the sled and the other shooting.  Or while one shoots presents down the chimney, the other shoots a Rogolaser at the vindigo.  While one delivers presents, the other stays in the castle and guards the Warming Hearth.

I supplemented the design-doc, but it didn't go any further, because I was still busy finishing the devlog at the time.

(April 2023 - May 2023)

In March 2023, the Cooperative HGC was started. In April, Royal Flying Corps development was unfrozen and conducted in cooperation with the winner of SHGC, CHGC 2020, BHGC, CHGC 2022.... Just the dominator of all HGC. Well, one of the developer friends finally managed to free himself for a full-fledged co-op development.

Meticulously mapped out the tasks in Jira. Promptly modeled gifts, sleds and the layout of the house. In the current realizations, it turned out to be not very convenient/pleasant to shoot at the chimney. So without coming up with a better realization of schmaling gifts, focused on network co-op.

The moral of the primary R.F.C. came out sideways here. The line between frontend and backend was so thin that I got carried away with the button functionality and went straight for the network connectivity functionality. It got confusing. It slowed the overall pace. It led to 101 conflicts when merging branches. Nevertheless, the minimum required network functionality was implemented. However, on playtests, it turned out that with real ping, network co-op was unplayable.  We have not been able to determine whether the problem is on the side of NetCode root libraries, on the side of a particular RFC SS implementation, or on the side of the virial local network.

ADDED:
> Network co-op. Mostly menu synchronization. Gameplay synchronization functionally works, but it's not playable in a real runabout. Too much lag, too many bugs and too few cases have been debugged
> Landing Mechanics. Sometimes it even works correctly. And if you hit it really hard, you can see a snow BOOM
> Models of sleighs, gifts and house

CHANGED:
> Visuals have been converted to 2.5D with character sprites and volumetric environments
> Redesign of menu. Now the menu looks good on both narrow-format and wide-format. Settings menu is united with the menu pause, death and victory. Added network co-op menu

FIXED:
> Stabilized the operation of chunks that during the player's turn could sometimes teleport to the wrong direction

BUGS:
> Local co-op fell off
> Terrible synchronization of players' positions in network co-op
> Сan't connect to an already running game
> Сan't disconnect from a running game
> Сan't swap a character in a running game
> Client doesn't always pick up the heart
> Death menu synchronization does not work
> Reset resets the level and only one player
> When leaving network co-op in the main menu, the root script falls off and nothing works
> In network co-op, fireballs lag far behind the client, a little more than half a screen
> Localization doesn't work. One half of the interface is in english and the other half in russian

We've spent 3 months, halfway through, and RFC SS is nowhere near completion. Rather the opposite. Then dominator proposed the concept of a new game, with half the number of questions and three times shorter description; a game that minimally needs not a whole neighborhood, but only one room. During the remaining half of the term we managed to make a completely new game. However, all of a sudden it didn't win. It took 2nd place out of 3. It was the first HGC that the dominator didn't win. To say our morale was broken is to say nothing. Fortunately, or unfortunately, the new game lost not because it has problem, but because the competitor was rare genius. Not even the release RFC SS would have been able to beat such a competitor.

RFC SS failed as a Co-op HGC game. However, as a game in general? I'm still fascinated and excited by the concept, where with each gift the gray city was gradually painted with the colors of garlands, the hum of the snowstorm was replaced by lamplight melodies, and the dead streets were filled with playful subjects that do not let the vindigo to sleep in their neighborhood, and when they see the princess unconscious, rush to bring her to her senses... You just need to find time to implement all of this.

And yet the Co-op HGC build has broken too many things that worked fine before. From the interface to the local co-op. It makes P.E.G.A.S. Public Demo look like a release build. So I hesitate to publish it.

***

It's been 2 years now, but I still can't find time for Royal Flying Corps. It is not big and ambitious enough to fight the mastodons U.N.I.C.O.R.N. and Prewar Phoenix in free time. But it's not small and primitive enough to go for a new HGC with its upgrades.

And yet Royal Flying Corps isn't dead. It is just waiting for me to finally finish Find Your Mare Simulator 2025 v1.2, Prewar Phoenix, the first stage of U.N.I.C.O.R.N. and decide on the fate of Puzzle (so are they still merged into the U.N.I.C.O.R.N. as a learning experience or continue to develop in the original direction of playerization of My Life And Game?)

With this post I wanted to show that the game periodically wakes up from the coma. That since the P.E.G.A.S. Public Demo there have been a few multiple builds made, it's just not that they're unequivocally better.