Sponsored By

Postmortem: Loop Hero

Loop Hero developer Four Quarters shares the story of the game's creation from pitch to launch, and reflects on what went right and wrong throughout development.

August 26, 2021

9 Min Read
Loop Hero's text logo framed by a building in the distance and a sprite of the protagonist in the foreground.

Author: By: Four Quarters

Game: Loop Hero
Developer: Four Quarters
Publisher: Devolver Digital
Release Date: 03/04/2021
Platforms: PC, Mac, Linux
Number of Developers: 4 quarters, 4 more people who helped us
Length of Development: ~ 1.5 years
Development Tools: Game Maker Studio 2, MS Paint (from windows XP), Aseprite, FL Studio


The Loop Hero team assembled around 2013, piece by piece. Initially, Aleksandr "blinch" Goreslavets
suggested that he and Dmitry “Deceiver” Karimov work together (like “hey, you can draw, and I can make music and code games, we’re just made for each other”). It turned out that it was quite difficult to develop games with just two people, and a little later Aleksandr "Finlal" Vartazarian joined as an artist assistant, and then Dmitry “theRandom” Lagutov, as a programmer.

In 2014 we started participating in Ludum Dare jam, a three-day game-making competition, where the goal is to make a whole game from scratch just for the fun of it. And after six years, one released game, 14 Ludums, and several frozen prototypes, we came to LD45.

Four individuals pose for a picture together outside near a building.

The theme was “start with nothing”, and shortly before that, Deceiver talked about his idea for an idle game about a hero who walks along the path and chops monsters on his own, and all the player has to do is put on the hero's equipment and build up the road. We didn't have any more suitable ideas for the jam, and so the story of Loop Hero began.

An early vision of Loop Hero's interface

A handdrawn mockup of Loop Hero's interface with early versions of a game map, inventory, and card system.

1. Starting tavern. Where we restore HP in each round. The less HP we have - the more the hero rests there. It’s not bad, just the day timer is ticking up.

2. We placed a cemetery here, so now the skeletons will spawn once per n days (so we don’t want to spend time wandering around and get into long fights). Furthermore - skeletons are more powerful, but drop better loot.

3. Here we placed a landscape card not meant to be walked on - the mausoleum. It should be placed adjacent to the road but not on it. Mausoleum spawns skeletons that move to the nearby road tile (as you can see there are two adjacent tiles, and one of the skeletons went to the wasteland. It will be weaker there, but it’s still an option).

4. Default wasteland. A slime has randomly spawned here.

5. Our equipment.

6. Our HP/LVL bars.

7. Our stats. Mostly affected by items.

8. Inventory. Dropped items go here. We can equip any of those items. But if we run out of space, the oldest item will be destroyed when picking up a new one. This way the inventory will fill up, but not clutter. And the player has some time to think, but can’t waste time.

9. Cards are analogous to the inventory with the dropping/discarding mechanic.

What Went Right

The Idea

We sometimes thought about making a full commercial game out of our Ludum entry, but only two out of fifteen came out.

Not every jam entry is suitable for further development; it is better to leave some projects as small competitive games, otherwise the attempts to develop them further may get bogged down. Therefore, it's very important to correctly assess the result and understand its potential.

Loop Hero became a perfect chance for us to develop a competition project into a full-fledged game. Already having a basic foundation, we had a lot of room to improve and develop this concept.

The Demo

One of our best decisions was to start with a small extended beta version of the Ludum build, rather than rush into making a full game.

After the Ludum, we decided to modify the game we had, add things we did not have time for during the jam, polish the gameplay and post the beta on our social networks in order to get live feedback from the players. This way we were able to find out people's reactions at a very early stage, we received hundreds of responses, which subsequently greatly helped us to improve the game, balance, and various aspects such as UI and UX.

Based on all this, we were able to make a polished demo version, which we used when searching for a publisher.

The text logo for Loop Hero's demo, LooPatHero, and an early look at the map.

Devolver Digital

From the very beginning, we decided to find a publisher in order to focus entirely on development and just not think about marketing, translations, etc.

Of course, we still had to do these things, but in retrospect, the amount of work that we had to put in ourselves is incredibly small compared to what it would be like if we went to self-publish.

Devolver helped us in every way they could. And, well, they put up an excellent advertisement. We certainly couldn't make the game so successful ourselves.

A Sober Assessment of your Strength

Over the years of working in a small team and more than a dozen Ludum entries, we have learned how to evaluate our strengths. When you have only four people on your team, it becomes very important to focus on the most important aspects of the game, without wasting energy on everything at once.

Often, when planning a big game release, each developer wants to add a lot of elements and mechanics to their game, to do everything at once. And the development of such a game in most cases will collapse under the burden of ambition. This has happened to us too before. And so in Loop Hero's development, we tried to limit ourselves to a small number of elements, but polish each of them.

Chaotic Workflow

This point seems to contradict the previous one, but we are talking about the end of development, when all the elements are already in place, the game can be played from beginning to end, and all the bosses are done. Initially, we had a rather abstract idea of ​​what the final game would be like, but everyone had their own interpretation of that idea.

The plan was this: Four bosses, a camp, and a hero looping infinitely. That's it.

But when all that was ready, we started to have many mechanics that greatly changed the game, and which were not in the original plan at all. The most obvious example is the traits mechanic. The right trait at the right moment can determine how the run will go, and is a very important part of the game, but they only appeared two-thirds of the way through development. And the perks that drop from killing bosses happened at the very end of development.

What Went Wrong

Pre-release crunch

It's rare games come out without it, and we are no exception. From early January to March 4 (the release day), and for two weeks after that, we worked every day for many hours without weekends or breaks. We did not crunch ourselves to death, of course, but still worked a lot. The main reason was the desire to make the game as good as possible, add a small animation here, a little movement there, redraw this panel. Finlal once spent a whole day making a little shader of a light stripe going through the auto-pause gear after it went all the way up. It is barely noticeable, and very few people even know about this function, so this shader could have been postponed and he could have taken a day off, but the desire to make the game was too great.

The habit of controlling your working hours turns out to be a pretty healthy habit.

Release on Mac and Linux on the first day

In general, this is not bad, but it was a mistake in our case. We did not expect the game to be so popular, so we announced the release for Linux and Mac, which led to many problems - it turned out that games made in Game Maker are not compatible with the new M1 chip from Apple, which is why the owners of the latest computers needed to manually run the game with Rosetta, and we had no way to inform all the players at once, and because of that many people couldn't start the game normally. Plus, compilation on each of these systems takes from 40 minutes to two hours (as opposed to 10 minutes for Windows), which is why the release of hotfixes was very slow in the first couple of days.

We are certainly glad that Mac and Linux users could also play from day one, but the amount of stress in the first days increased significantly.

Very interesting but very confusing bandit mechanics

There are bandits in the game. They appear if you place too many villages, and have several mechanics that distinguish them from all other enemies.

For starters, they can steal things and directly delete items equipped by the hero, which other enemies cannot. They are also able to enter villages, which other enemies cannot. They also can leave the villages, if there are more than three of them, and go to the adjacent tile. And yes, other enemies cannot do that.

But this is not a problem in itself. The problem was that we added the bandits in the last week before the release, AFTER all the tests.

Of course, these mechanics became the source of a huge number of bugs that appeared for many people in the early days, and made us stay up late and fix it. If we had tested everything beforehand, this wouldn't have happened.

A sketch of Loop Hero's bandit enemy type


Perhaps due to the fact that our team has many years of experience working together, or maybe it's luck, we didn't stumble too much during the development (although that may still wait ahead!).

It's hard to draw any conclusion now. We have plans for Loop Hero, we want to add new content, updates and release the game to new platforms. The last 1.5 years have been crazy and difficult, but everything is just starting and new difficulties and challenges still await us.

Read more about:

game developer launch
Daily news, dev blogs, and stories from Game Developer straight to your inbox

You May Also Like