Bomberlands - development history, annual report and tips for beginners

If you completed your first game, then something is wrong with you.

This phrase, read in one of the posts, was typed into my memory 2 or 3 years ago. And the author was right. Starting to make his first prototype, most developers throw him halfway and take on the next for various reasons.

Despite the fact that this is not the first game I finished, this is exactly the one with which I started. A year has passed since I wrote here about my prototype . In short, I wanted to reanimate the old tube Minebombers. Transfer them to 3D and add network interaction. After that article, I was very lucky :). I received not only a few offers of help, but also real companions :).
Now looking back a year ago, you understand how many mistakes were perfect and time was wasted.
This is what I want to share. History, mistakes, experience.
Immediately I apologize for some confusion.

About copyright

After receiving such a peculiar blessing, we began to develop.

Where to begin? Dizdok!

Setting and heroes

With this, everything must begin. After the reviews of the prototype, it was clear that a simple transfer in 3D is not enough. You need to create your setting, add history and RPG elements and link them to the gameplay, forcing you to level through level, using a large arsenal of all kinds of bombs. Since we wanted to keep the original without a bit of humor and add humor, then minions would become ideal heroes. Unfortunately, the brand was busy, so we decided to create our own. Since we were inspired by the “Despicable Me” cartoons, then everything should look very organic. The background confrontation between “superheroes” and “supervillains” results in the destruction of the underground base, on which one of our minions lives, after which it is responsible for the fate of their fellows.
Since the base is underground, they took moles as a basis, endowing them with a good amount of mime. So the krotans appeared :)

A lot of krotans .

Subject campaign and enemies

Since the confrontation between "superheroes" and "supervillains" was the background, and Professor Kopatel after the attack disappeared in an unknown direction, it was necessary to come up with a plot campaign and enemies. Fortunately, it was not difficult. The destroyed underground laboratory allowed organically settling it with all the main types of enemies: robots, slugs, stone golems and even zombies :)

Description of the types of opponents.

Basic gameplay and GUI

Another prototype revealed a problem with control. Too many weapons (about 20 types), and only 2 buttons to switch them. Yes, and where does a small Crotan even have so many weapons in your pocket? Although the characters we have and cartoony, but the environment of Hi-Tech. Therefore, it was decided to release crotan to levels only in a special pumped exoskeleton, the main feature of which was the presence of a weapon printer, so that it could quickly create any bombs to choose from the available materials, rather than carry them with them. Also, 6 types of materials (metal, plastic, electronics, explosives, fuel and uranium) from which one could make bombs were put into the game. The bombs themselves were divided into 6 groups according to certain criteria. And those that stood out much made class. Yes, this is where the class tree came in useful.


A brief description of the setting and plot can be formulated as follows:

The battle between good and evil ended in a crushing defeat of the latter. The combined forces of superheroes dealt a crushing blow to the den of the supervillain. Its underground base has been severely damaged.
You are one of the surviving, loyal minions. You must find and save your fellow survivors. Become their new protector. With this you will be helped by one of the remaining working exoskeletons of the MGE-OP-14, in which you can clear the remnants of the base from the monstrous experiments of its owner that have broken free.
So, restore the base and retaliate !!!

Creating objects and characters

This part is unlikely to be a discovery for someone, but still I will explain how the work on the 3D game looks like. You have a list of objects you want to draw. The screenwriter and / or ideological mastermind, knocking fingers over the keyboard into the blood and working with the mouse until the power cable is worn, collects refs (pictures, photos or video of what the object should look like) and give it to the artist. Of course, you can not search, but limit yourself to a dry description of two words, but then you can expect anything from an artist. In general, remember, a telepathic artist is a rare, almost extinct creature, so take the trouble to explain to him what you want from him.
The artist draws a concept, and shows it to those who are deprived of artistic abilities. After several shipments for revision, the modeler takes over. A good modeler needs just one concept. Give less experienced and in front and in profile. The model can be high poly or low poly. Respectively high and low polygonal. And, interestingly, good lowpoly modeler beast is more rare and valuable than highpoly. In addition to the model itself, it must also marshall it (i.e., create an empty texture-blank, something like a sewing pattern) and give it to the texture maker. The text leveler naturally makes the working texture so that the model looks like concept art (in addition, it can throw a normal, a dump, a speculator and other mapy). Then, ideally, the model is transferred to the rigger that creates the rig, i.e. the skeleton of the model, and the engraved model is transferred to the sensitive hands of the animator. As a result, the programmer gets the model, textures, animations and inserts them into the game.
So, a concept artist, high- or low-poly modeler, text level maker, rigger, animator. At least 6 professionals are needed to create art for a 3D game. Of course, some tasks can be combined and confined only to an artist, modeler and animator, but still this is a lot compared to 2D games, where you can do with just one artist;) Examples of how our characters were created are contained in the spoilers above.
Fortunately, we were lucky with both the artist and the modeler. Both were professionals in their field. But the animator did not have enough money. What I regret, of course. But we were lucky, because The Mixamo service, recently purchased by Adobe, has come to our attention. I advise it to all indie developers for several reasons. Firstly, it allows you to buy already shattered 3D character models. Secondly, the animation to them. Third, it allows you to upload your models with a rig, animate them on the site and buy the resulting animations (moreover, such parameters of these animations as speed, arm span, level of hand position, etc. can be calibrated). Fourthly, if you only have a model, then you can try to use the authoring Mixamo. Moreover, it works quite well with humanoid models. Unfortunately, the humanoids among our characters are few, so most of the rigs were ordered from a freelancer, but the service itself showed its best.
For example, we decided to add a small hand-to-hand fighting game to our game, while it took only 2-3 days to create animations and a tree of behavior of Mecanim, and this was due to the abundance of enemies.

A little tip. About fake screen

If you have an artist and the project is not quite primitive, then spend a few days creating fake screen, i.e. gaming pseudo screen so that all team members imagine the end result. Place there the main objects of the environment, the hero, if that is the interface. It will save you a lot of time in the future.
Here he looked like this:

Classic mistakes. Unnecessary chips that are not included in the game.

Pixel-Art mode

Initially, there was the idea of ​​displaying a minimap of the level in the form of a good old pixel-art card, with the ability to switch to 2D game mode. So that players tired of 3D bells and whistles can enjoy old school. Initially, as in the original, most of the map was occupied by various rocks that needed to be drilled and blown up.

4 breeds differing in texture, color and strength, and divided by height into 3 types.

They, along with the other elements, were redrawn in pixel art style (on the screenshot, of course, not all game objects).

Not bad, but something is missing. Volume, to be exact. To make it easier for the player to determine the strength of the blocks, we added to the color and brightness another texture of different size and a falling shadow of different lengths. Those.tall blocks gave the highest shadow, while if it fell not on the ground, but on a block of medium or low height, then its length decreased. At the same time, the shadow falling diagonally looked best, and therefore stones of different heights could be below and to the right of it. In general, you understand that drawing several thousand sprites of the card itself would add a few thousand more shadows of sprites, rendering the height of neighboring objects and all this would take away a lot of CPU time. Therefore, a special shader was developed to transfer all these calculations to the GPU. As a result, on the basis of the current map stored in a two-dimensional character array, a texture was created that goes to the input of the shader and gives such a playing field to the output. And all this for one DC.

To work on these I had to use a couple of freelancers (pixel-art and writing shaders). All this beauty took me almost a month of working time (maybe more, now it is difficult to remember). This includes drawing up the TOR, debugging the shaders, discussing, coordinating and inserting it all into the game. All this corresponded to the original mechanics, where each object occupied one cell.
But then, it was decided to decorate the levels with various decorative elements and add an island structure. To transfer this adequately to pixel art was already too time consuming and did not guarantee a result. Therefore, the 2D mode can be seen only at certain levels and in a very limited mode. Although who knows, maybe it will be possible to make a branch exclusively in this style :)

Server part

The server part has also been burned. Initially, server logic was developed for the original mechanics. Moreover.It was finished and ready for load testing, but after adding decorative elements to levels, there were no resources left to refine the server logic. In the meantime, it was decided to focus on a single game, so that we have pvp in semi-finished condition.

Base location

Instead of a banal GUI, we decided to visualize it as a base location. Only instead of the Normandy from Mass Effect, we have equipped an underground laboratory. When clicking on any device, the camera moved to the screen with the appropriate interface. Not that it was completely in vain, but it took about a month of work of the whole team.

Inevitable evil. Interfaces

The best interface is the lack of interface. Unfortunately in our case it was unrealizable. Therefore, the work on the interfaces took months and this is not a joke. Months of development and implementation. Routine work requiring only perseverance and concentration, which can not be avoided.

Unexpected discovery. Project management is really a job.

Guiding the work of the development team is indeed a job that takes up all working time. If I had the opportunity to return a year ago, I would either find a programmer instead of myself, or someone in the place of a project manager (most likely an experienced game designer). Do not be mistaken that you can easily combine the work of a programmer / artist / modeler and manager. It takes a lot of time from you, and you will not have time to look at the project entirely. Of course, such a manager should have experience in some field, so that if necessary he can back up his colleagues.

Error or not? Breaking bad.

Filling the RPG part of us suffered. After all, we counted not only on the single-player campaign, but also on the constant online battles. And for that, it was necessary to give the players a good leveling capability. In addition to the class tree, the choice of a profession by players opened access to special weapons and skills.
3 classes, 6 subclasses, 19 skills.

Our exoskeleton conditionally consisted of 10 slots for equipment of various types and levels.

Up to five types of equipment per slot, which could not only be bought at the store, but also produced at the base, after studying the appropriate scheme.

In addition, almost any equipment could be enhanced by inserting a modifier into it.

There are also a lot of modifiers, and many of them not only increase the parameters, but also add new mechanics to the game.

I have already mentioned about the weapon, about two dozen original and about a dozen new bombs on the way (freezing, light, magnetic, etc., intended to fight against our mobs).

Do you still have little pumping? Then for you the whole underground base, from 15 types of premises, each of which can be pumped up to the 5th level. Here you can develop the extraction and production of resources, get access to shops, combat modes, pvp, explore the schemes of weapons, equipment and modifiers.

Total:

• 3 classes;
• 6 subclasses;
• 19 skills;
• 49 types of equipment;
• 50 types of modifiers;
• 75 types of base modules;
• 20 types of weapons.

The question of whether we realized it for nothing or not for nothing is open and it is too early to judge this, but the fact that implementation took a lot of time is a fact.

, . .

If you have a person who does not know how to program, draw, model or write music, but wants to take part in the development, then it does not matter. Let him deal with public relations. Let him out of the bushes, like a real paparazzi, watch your work by arming yourself with a long-focus lens and fixing your every word. Let him compile reports for the faithful fans about your current work, publish game screenshots and videos in your VKontakte groups, Facebook, Twitter, and even in instagram. Spread test builds, collect reviews. All this needs to be done throughout the development to know if you are going in the right direction, but no one has time for it. In terms of personnel shortage, each article for your VKontakte group takes your precious time, and after all, fans are such people,who want to receive news if not every day, then at least every week. At one time, we actively led our group, and our wonderful scriptwriter / level designer released a couple of posts a week, but how much time did it take for him? In general, everyone can find a useful lesson.

Council for growth. Conference.

Conferences and various jams are not just parties, but also a place where you can get a lot of feedback from professionals in the gaming field. So if you have the opportunity, then participate in them as often as possible. Personally, because of my place of residence, I am deprived of this opportunity, but for metropolitan developers this is a significant competitive advantage.

The harsh truth. The end of development is only the middle of the road.

Promotion and creation of promotional materials on labor costs is not much inferior to the development itself. Creating an advertising video, game art, game descriptions, advertisements, reviews, mailing and advertising in groups, takes a lot of time working professionals.

Oil. Butterfly Effect.

Much of our life is due to the butterfly effect.
... a butterfly in Africa sat on the nose of a giraffe, the giraffe sneezed ...
... frightened gazelle, gazelle imprinted on rhino ...
... and the rhinoceros blindly ran into the phone booth ...
... somehow called OPEC and said ...
"Immediately increase loot!"
... the price of oil is falling
... and the dollar doubles against the ruble
and here we are looking at how our budget shrinks no worse than shagreen.
Conclusion: if you are counting the budget in rubles, then agree on payment with everyone you can in rubles, and lay out currency items of expenditures with a margin.

Content generation. Is the bet on the player justified?

The original was an excellent map editor. Simple, clear, effective. So even in the prototype a year ago, a similar map editor was present. There was even the ability to upload your own maps to the server and share them. The bet was made on the fact that the players will be able to fill the game content themselves. Create your own interesting levels, share them, etc. It was easy and paid for itself by the speed of level development. However, if the usual game block occupied exactly one cell, was of standard size and was easy to encode with one symbol in a two-dimensional array, then with the addition of decorative objects, the situation changed. They could be installed at any point, in any quantity, change their size and rotation, they could overlap each other, could interact with the player or not. To this, you can add various scenario objects, such as quest triggers, mob generation points and resources, etc. All this had to be made to the editor, to render objects for icons (and these are several hundred objects), to maintain the constant ability to download maps to the server. Together with the complete serialization of the whole environment, the editor took as a result another month or two. Of course, the editor really came in handy for our level designer. There was no need to learn Unity itself. Plus, the system still works. He can easily edit the current scenario map and update it on the server, or save it locally and immediately go through it without making changes to the build. On the other hand, the resulting editor is somewhat monstrous. After teaching the designer the basics of working with Unity, you could also create new levels without your own editor. They would be harder to test, they would weigh more, but the choice of tools would be wider. Baking lighting, fine-tuning effects, and more would be available. Again, later it was even easier to serialize all of this data, although the maps would weigh more and the players would not have access to the editor.
The conclusions are not surprising in general: the complexity of the game mechanics, and in fact the change of the development paradigm, led to the complication of the accompanying infrastructure. And although the players still have the opportunity to create their own content, but because of the increasing complexity of the process, you cannot count on a large number of such players. The editor will probably remain just a regular development tool.
Was this all a mistake? Of course not. But if we immediately chose a different path, the result could have been no worse.
So is a bet on a player justified? It all depends solely on the project itself. Your condo

Something the article turned out too voluminous, but I didn’t even really talk about the technical part. Looks like you have to interrupt. In the following part I will tell about actually technical side of development on Unity. If there are any other questions, then leave them in the comments, I will answer short ones at once, I will try to highlight the rest in the next article.

PS By the way, we passed Greenlight in a week :)