Actually going through with finishing making a game?

[Babar]

How do you do it?!

I'd like to call on the expertise of the more prolific among you, who regularly complete games, mostly on their own (browsing the MAGS wins, I see people like @4KbShort and @Ponch and @ddavey1983 and we miss you slasher!), and ask for your specific process, with actionable steps I can take to follow them and also complete a game, for MAGS or longer term, or shorter term.

My process usually goes like this:

• See a topic or theme or contest that interests me (man, I wish I could go back to my longer term personal projects at some point!), or often doesn't REALLY interest me, but I want to challenge myself and make a game, so I try anyway, often twisting the theme
• Coming up with a game idea that fits the theme- this is usually my first point of failure, or the thing that unreasonably takes up 50% of the time of the competition
• Trying to map out the puzzles/plot progression for the game- this is usually my 2nd point of failure, and probably the most common one, even happening if I progress to steps beyond this one. I guess there could be multiple reasons for this, mostly involving me not wanting to do a simple use tomato on cook->get sandwich->use sandwich on guard->get key->use key on gate->... type of puzzles for due to them being boring, and then as a consequence, often trying to innovate TOO much ("Maybe I can make a match-3 minigame!") or even getting caught up in side-issues ("maybe my interface can be a super accessible only spacebar and nothing else!" "maybe I can implement a state machine!" "maybe I can try an iMUSE type system!")
• While theoretically I'd like to have completed an entire progression of the story/puzzles before this step, that almost never happens, so I figure I'll come up with the basics of the previous step and try to move to this next step- actual implementation: usually involves setting up the game system (how the UI works, the intro and start menu and icons) or doing a couple of the BGs or main character (although I usually start with a placeholder character), while still working on the previous step. I haven't recently progressed far beyond this step.

Help? Please? Step by step, how do you fine people who do it, do it? Do you often feel like you're intentionally implementing or planning something substandard just to "get it done" (or just get deadlocked and not complete because you couldn't do the substandard thing, nor come up with the awesome thing)? How do you limit yourself and prevent your idea from ballooning out of control? Conversely, how do pyt in that one or two "cool", "different" things so that your game is not just another boring checkmark on the list?

[cat]

I admit, I'm not prolific at the moment, but I've finished a few games in the past.

Some people have a story to tell and think a game is the right medium for it. For me, it's the other way round, I want to make a game and have to come up with a story for it. That's why my games usually don't have a deep story. Instead, I focus on the game world and try to fill it with people/objects/obstacles that would fit in there. So the world grows in a somewhat organic way. I think this results in puzzles that feel relatively integrated. (Or sometimes I just think it would be fun to code a catapult/maze/music minigame and find ways to integrate it in the game )

Another thing that works for me is to bounce ideas with other people for puzzle design. This can either mean working on the game together with someone or just talking to a friend or colleague.

[RootBound]

I've released 4 games this year, so I feel justified hopping in here. 

For me, I try to stick to the following:

1. If it doesn't REALLY interest me, I don't make it. Making games is too much work to put into something I'm not excited about. That pretty much guarantees I will lose motivation before finishing. Don't underestimate the importance of this. Wait until you have an idea that fires you up! You'll need it. 

2. No matter how much I want to start  programming the game, don't start until the puzzles or gameplay are at least mostly worked out. Things will go more smoothly (and importantly more quickly) if you to know what puzzles to put in the rooms before making the rooms. Be patient with the planning. That's when you can revise instantly because you aren't locked into anything yet. It's MUCH harder to change course later. Enjoy playing at this stage. BUT...

3. Keep the game as small as it can possibly be. Then make it smaller. All 3 of my MAGS games have had only 1 room of gameplay. Not every game can be like this of course, but if you allow the game to sprawl, then "finishing" keeps getting farther away. Combine rooms if you can. Stay focused on the core of the game, and if you have time later, add extra stuff after the game is fully playable with just the central elements. Or release later versions once version 1 is out.

4. Accept that at some point it will stop being all about fun and will become work. The goal is to have enough of the game finished by that point that the prospect of being done is enough to help you power through the work part. This is more likely to happen if you follow points 1 through 3. Put in the time even when you're not in the mood. That's the most difficult stage of the whole process, but for me, it makes or breaks whether games get finished.

5. There will be snags. Expect it to take 3 times as long as your first reasonable estimate. No matter how many games you've made, something that seems simple will end up very complicated to implement. Be ready.

Not sure if all that is motivating or discouraging, but I hope it helps! Other people will definitely have different approaches that work better for them.

[Kastchey]

I probably shouldn't speak for myself because your issues sound all too familiar but I can share what Slasher once said. He had made himself a "generic game template" - title screen, basic GUI, locations map, save/load, settings, on hover text label, some generic functions etc., that he would use for most of his MAGS games. This would save him tons of time setting all the basics up.

[Cassiebsg]

I usually get an idea inspired by the topic. Then I might use a couple days just figuring out the story in my mind, once I'm satisfied I start (if I'm confident I have the time to finish it). Though, considering that I have a bunch of unfinished games, I might not be the right person to answer this either. But if I do start, I make a commitment to myself (and to others, if I'm not working along) to at least delivery a working version of the game (I'll then call it Demo, if it's not fully finished as I wanted). This means I will make the time to make it happen, even if I have to sacrifice a few nights sleep.

I'm though happy with the type of puzzle you describe, if it makes sense and progresses the story line. Not that I wouldn't love to have mind blowing genius puzzles, but I accept my limitations in puzzle design and story telling.

Also, keep the UI and such simple, since the point of MAGS is to make a game, not the ultimate UI. I usually take the BASS template, add a few of my "usual" stuff from previous games, and go from there.

Then it's art, sounds, music, story & puzzles.

Make sure you have the start and the end of the game done first, the add as much as possible of the "middle". This will guarantee that you'll always have a "finished" game ready to deliver, if you're happy with it that way... if not deliver it as demo and work on it post MAGS (I have a problem when I get to this point though).

[Ponch]

Hi, Babar. I'm flattered you asked my opinion on this. 

First, I usually start from one of two places. Either I have an idea for a story (or sometimes just a character) that I really want to tell. The main appeal of adventure games for me has always been the character and the world of the game. The other starting place is when I have an idea for a puzzle or a particular sequence that I just want to explore. Maybe I just want to make a crude first person shooter, or a top-down scrolling game, or something like that. Both of those starting points require me to really WANT to make that game.

After that, I build a crude version of the game as quickly as I can, just to see if the idea is going to work as well as I want it to. I typically make a list of the rooms I am going to need. I scribble them down on a sheet of paper along with the room number for each. I group them by the "scene" they will be used in. "Bar 1" "Bar 2" etc. Then assign numbers to be used when I actually put them in the game. I try to leave a few room numbers unused just in case the area takes a few more rooms to realize than I thought it would. For example, all the rooms in the bar will use the room numbers 30-35. The next area, say "Office 1" "Office 2" etc, will start their room numbers with room 40. That leaves me a few spots to plug extra rooms in where needed as the idea matures. Other people may plug the rooms in in whatever order they made them, but this makes the game harder to "see" as I'm working on it. And if I'm going to stay focused during the unpleasant "grind" phase of indie game development, I need to make this as easy for my brain as possible.

I develop each area of the game at a time. First the character's home, all in crappy MS Paint art. No walkcycles, not proper sprites yet. Just enough to have the guy slide around the room, interacting with lots of crudely drawn places and things. Much, much later, I'll go back and draw everything properly. Once the home is done in this rough way, then I'll move on to an equally rough (but playable!) build of his office. Then the bar. And so on.

In fairly short order, the game is playable, start to finish, even if it may look very bad. Most of the dialog isn't there, the jokes aren't there, nothing is present except the necessary framework to hang everything on. I develop it the rest of it later.

Once all that is done, then I go back and start coloring in the world. Depending on what kind of mood I'm in that day, I might do backgrounds. Or maybe I'll do sprites. When I'm at my most awake and focused, I write dialog and narration. If I find that I have suddenly thought of a good joke involving the broom closet the office, I have the extra space in the room numbers to add another room at his job to allow for that without disrupting the flow of the ordering system that my subconscious demands, for some reason.

If the game has a lot of puzzles (like my most recent Barn Runner game, for example), then I try to work on those only when I'm sure I'll have lots of time to code and debug them. I HATE having to close a game with buggy code. After a few days away, I feel like I'm starting from scratch trying to chase the problem down again.

Once all the rooms, items, characters, and words are done and ready to be seen by the public (i.e. play testers), then I do the cutscenes last. Those things add a nice layer of polish to a game, but they're the most expendable part of my design philosophy. It's great to have them but it's easy to work around them if they have to be edited out for time (for MAGS, usually).

The most important thing is to schedule time to work on the game. I know sometimes this just isn't possible, but if you plan to devote a little time on a regular schedule, the game will eventually come together (assuming you're not racing a MAGS deadline).

Don't wait for inspiration to work on a game you have already started! That wonderful rush you get in the beginning of any project will be there at the end too, if you can manage to complete the game. If you're not feeling inspired when your daily appointment with AGS arrives, then do the grunt work instead. Polish the background. Smooth out that animation. Edit the dialog a bit. Do something, just so something gets done and you're one step closer to the finish line. On those days that you're really excited to work on your project, make the most of that energy. But don't let time get away from you just because you're not feeling it that day. No one will finish your game but you! Don't let that little dream die!

This is how I approach actually making my games, in a nutshell.

None of the above guarantees I will actually complete the game, of course. I have lots and lots of half-finished games in my wake. But it usually gets me to the finish line (MAGS notwithstanding, of course ).

Anyhoo, I hope this was helpful. Thanks for coming to my Ted Talk. 

[heltenjon]

I probably shouldn't speak for myself because your issues sound all too familiar but I can share what Slasher once said. He had made himself a "generic game template" - title screen, basic GUI, locations map, save/load, settings, on hover text label, some generic functions etc., that he would use for most of his MAGS games. This would save him tons of time setting all the basics up.

He would also re-use assets, such as characters with finished walkcycles or talking animations, now being cast in new roles. He would see it as an actor now playing a new role, and not be concerned that he would have to make this the same character and that all the connections fit. Famously, he used the same character sprite with different colour clothes for a set of goons in the same game. The art was just a means to an end to him.

I recognize Babar's work process, though I'd usually breeze through 1-3 and never get started on 4, because that involves art, which I have little to no talent for. I guess what can be read from what you write is that you may be overly self-critical. The puzzle sequence you describe may be perfectly fine if it's told right, or with some small twists. (Or maybe I'm not critical enough about my own puzzles.) Even in a game with great puzzles, there are usually some things to do that are fairly easy. Which is nice, because it makes the player feel good, as there is some progress, even though there are puzzles that seem impossible to solve at that point.

[Danvzare]

Your process seems good. Your problem seems to be the amount of time you're allocating to the steps.
So let's go through them one by one.

• See a topic or theme or contest that interests me (man, I wish I could go back to my longer term personal projects at some point!), or often doesn't REALLY interest me, but I want to challenge myself and make a game, so I try anyway, often twisting the theme

Nothing to add here. This is a perfect starting place.

• Coming up with a game idea that fits the theme- this is usually my first point of failure, or the thing that unreasonably takes up 50% of the time of the competition

For a MAGS game, this part needs to be done as quickly as possible. You need to brainstorm with yourself, throwing every idea you can at the wall, try to elaborate on what you like, see if it has legs, if it doesn't look at another. For a MAGS game I'd say that this should take you no more than 24 hours. If you can't settle on an idea in that length of time, either give up or choose one. Remember, you don't need to make something that's perfect, just good enough. It's more important to finish something than to start something. And if you're stuck this early in the development process, it'd be a better use of your time to work on something else. After all, if it's not working, why force it.

Of course with a non MAGS game, you'll want to spend much more time on this part. Anywhere from a week to a full month. (Any longer than that, and obviously the game is going to be long enough that you'll be able to figure the rest out while you're making it.)

• Trying to map out the puzzles/plot progression for the game- this is usually my 2nd point of failure, and probably the most common one, even happening if I progress to steps beyond this one. I guess there could be multiple reasons for this, mostly involving me not wanting to do a simple use tomato on cook->get sandwich->use sandwich on guard->get key->use key on gate->... type of puzzles for due to them being boring, and then as a consequence, often trying to innovate TOO much ("Maybe I can make a match-3 minigame!") or even getting caught up in side-issues ("maybe my interface can be a super accessible only spacebar and nothing else!" "maybe I can implement a state machine!" "maybe I can try an iMUSE type system!")

Mapping out the puzzles and plot progression isn't too difficult. Just write down a series of bullet points for your idea, then iterate in your elaboration. Fill in the spaces between, how do you get from one bullet point to the other. The puzzles dictate the plot and vice versa, so they're more or less made simultaneously if you're doing it right. For a MAGS game, I'd recommend getting this done within a week. If you're struggling to figure out how to make something into a puzzle, then talk to someone. Ask them "How would you pump up a tyre without a pump?" or whatever your puzzle involves.
If it's a longer game, then I'd recommend at least a month, with a constant ongoing process as the project proceeds.

As for wanting to innovate. If it's a long game, then that sounds good to me, do it. But for a MAGS game, like I said before, it's about finishing not starting. So keep things simple, get a feeling of what you can and can't do in the timelimit. If you know it'll be difficult to implement such a feature in the time you'll have, just shelf it for a future project. There's nothing wrong with not innovating. After all, you can innovate later.
Besides, when you get to this stage, you're already pass the point of adding innovating ideas. That's something that you should be considering during the previous step.

• While theoretically I'd like to have completed an entire progression of the story/puzzles before this step, that almost never happens, so I figure I'll come up with the basics of the previous step and try to move to this next step- actual implementation: usually involves setting up the game system (how the UI works, the intro and start menu and icons) or doing a couple of the BGs or main character (although I usually start with a placeholder character), while still working on the previous step. I haven't recently progressed far beyond this step.

There's nothing wrong with playing by ear and just improvising as you go along. Adventure games seem to be quite a good genre for doing that.

Something that I've done a couple of times is having planned everything out, and just putting it all in at once. Quickly drawn backgrounds, huge rectangular hotspots everywhere, no objects, no talking, and walkable everywhere. Basically, a working prototype that's only completable if you know exactly what you're doing. It took me a week to do that for the entirety of Black Friday II, and that game isn't exactly short. So it goes to show how quickly you can get something made if you know what you're doing. The problem is, you need to have planned everything out, and also be willing to slowly replace all of the parts you implemented in the prototype.

Alternatively you could just implement everything one section at a time. Like an entire room, or an entire puzzle, or an entire chapter. Making sure to polish and refine every aspect before moving on. In other words, you can choose either the modern "Agile" method of coding or the old fashioned "Waterfall" approach. I personally don't have a preference. You usually feel like you're making more progress at a quicker pace with the "Agile" method, but it takes the same amount of time.

Help? Please? Step by step, how do you fine people who do it, do it? Do you often feel like you're intentionally implementing or planning something substandard just to "get it done" (or just get deadlocked and not complete because you couldn't do the substandard thing, nor come up with the awesome thing)? How do you limit yourself and prevent your idea from ballooning out of control? Conversely, how do pyt in that one or two "cool", "different" things so that your game is not just another boring checkmark on the list?

Substandard? Never. The first step to finishing anything is to accept that it's not going to be perfect. Never aim for perfection. Aim for "good enough" and you'll be fine. In my book, "good enough" isn't substandard, if anything it is the standard. 
As for limiting yourself. Well, I just shelf it. If I come up with a great idea, awesome, I can put it to one side and do it later. You don't go needlessly adding onto something. You came up with a plan, and now you need to enact that plan. If there's time, perhaps you can add that awesome idea you just had. But otherwise, put it to one side. You're not a child, so concentrate and focus.
As for how to put that one or two "cool" or "different" things in your game so it's not just another boring checkmark on the list? Like I just said, you take something from that list of awesome ideas when you first start the project. You want to make a match-3 tile game? Don't try to add it into the game you're already making. But do make it the main premise for your next game. (Like with the chasing scarecrow segments in my Black Friday II.)



TL;DR
Don't aim for perfection. Just aim to be good enough, and quality will naturally come.
It's better to cancel a project that's going nowhere and begin a new one, than to struggle to continue one that's going nowhere.
And most importantly, finish things. Even if it's cancelling it, scaling it back, or just releasing what you've got. Make things "done" and soon you'll be able to look back with pride at everything you've accomplished.

[Rik_Vargard]

You can go through any obstacle on your way to finish a game if you really love what you're doing.
And you really have to fkn love it because the challenges are big and many to get there.
It's so overwhelming, so many aspects, so much to do.

How do I do it, as a one-person studio... well like you said, step by step.
I have that "global brain", so I "see" the game as it could be when finished.
Thus I also see all of the things I have to do, and that's where it get's overwhelming.

Once I know where I want to go (I think it's the most important starting point), I try hard to calm down my brain and get more "sequential".
I have ALL those things I have to do. Sometimes I take too much time sitting here and thinking about how this and how that.

So at one point I'm like OK, what can I do >right now, because I will have to do it anyway.
Like "Damn it I have that plot/story/character problem, I have to figure it out."
Yes I do. But right now? I don't feel inspired right now.
Let's put that aside for the moment and focus on what I really can achieve right now for that game.
Changing the hair cut and color of a character and that's enough for today?
Great, I made one step. And that step kept me (and my brain) in my "game universe".
So while I changed my character's hair color, I got an idea about that plot moment I was stuck in.

As you can see from the replys, the ways are as diverse as there are creators.
I would be one to say to aim for perfection and greatness because it's inspiring.
I like to think that it's better to have a big picture that you can scale down than a small picture that you want to scale up. Because, you know... pixels.
The tricky part of that is that you have to know that you will have to scale it down at one point, and then actually do it.

And then: Finish things. Just-Finish-Things. You started it, finish it. "learn and move on", but finish the project, whatever "finishing the project"  means to you at that moment, I really think you have to "act it". It will bring you peace of mind and global experience points for your next project.

[Stupot]

The very few times I've actually managed to finish any projects (tiny though they were), they all had the same rare magical alignment of free time and motivation, which is something I struggle with.

Another thing is this: the ones I've finished all were planned out with pencil and paper first. Everyone works differently, but I think having them basically complete in a notebook really helped me to focus on just transferring it to AGS without adding any extra crap. Of course I changed some minor stuff during the transfer (dialogue etc), but all the rooms and puzzles were basically sorted.

(Also... I don't know if anyone's noticed, but I've never made a walk-cycle in my life. What a time sink! It's either been "first person" or the guy just stands there while you click about.)

[4KbShort]

@Babar  , thanks for the mention. I dunno about prolific, but I'll offer up what I can!
So first off, I'm autistic and my attention span can vary a LOT on things. I also procrastinate a ton! For this reason I like structure, rules, and time limits! For that, I'll focus on the game jam aspect of my design process. Since the normal process is "Get hyped for an idea. Work on it until I run into the first hurtle. Give up."

So for me the game jam process is like this:

1: Check the theme and see if it piques my interest. This is identified by my brain throwing out a bunch of mechanics, images, completed game art, etc into my frontal lobe at high speed. If this doesn't happen then I'm going to have to "work" on the theme and I've got other things to work on. This should be fun, not work.

2: Identify and pick ONE of the many things my brain just threw at me. I find one that seems interesting, new (to me), and easy to implement. Usually this is a game mechanic, art style, sound design, or general "feel" of the game such as scenes and genre. Like if the theme gives me a horror vibe then I'm making a horror game. (The RoadTrip) If it lends itself well to a text adventure, and I want to know if AGS can do that, then I make a text adventure (The Delivery). I don't try to focus on things like "Oh, this would be a great game for a Tower of Hanoi puzzle!" That's minutia and that's for later.

3: From here I do a lot of thinking before I code. This is procrastination, no lie, because I work harder the closer to a deadline I'm at, but I do THINK about the project before I start. I let me brain wander around it and look at it and all the holes in it. I try to identify key points in the game; Point A, the start, Point B, the end, and any key scenes my mind wanders back to such as two characters meeting to reveal "the twist!". I try to make a path from where I WANT to be from where I think I should start. So if I want my story to end at the top of an Aztec Temple in South America I need to figure out how to get a British Archeologist from his vacation in Alaska to that temple.

4: Once I have a "path" or enough of one to where I can see where I'm going from screen to screen I start things up and begin to work. At this point NOTHING is set in stone. Everything can change in a moment because if I stick to close to the loose ideas I've come up with I'm going to get stuck. So instead, as I work on the project, I make adjustments to avoid or fix "stick" places. For me, getting stuck on something can kill a project quickly because I get bored working on it. So if I can't get my character to walk properly then fine, he doesn't walk he just warps from screen to screen. Moving on. (I did this for The Errand to see if it was something AGS could do. It can. So now I can use it if I need to in the future.)

5: For stories I just kind of write whatever fits the theme. That sounds kinda lame, but what I mean is that if I see a theme I like and it throws ideas at me I'm less focused on the story than the "feel". IE: A horror game needs horror elements or I want to make a text adventure, what would require a lot of descriptive texts over pictures? If it's a space game what kind of interesting things can happen in space? I don't focus on names, or backgrounds, etc, I just plop a character on screen and have them do things to move forward. Sometimes that builds a story. See The Passenger for a grand example of the situation creating the characters because the only idea I had on that one was "It's on a plane!".

So the list doesn't get super long here are some pointers that have helped me work on projects big and small for years and each one has made me a little better at overcoming my issues and pushing out more games and projects than I ever thought I could:

"Perfection is the enemy of completion." - Adam Savage
K.I.S.S.: Keep It Simple Stupid
Complete small steps to get to your goal.
Build the world first then lock the doors.
Puzzles should be no longer than three steps deep.
A cliche story told in a unique way is a unique story.
Have a "Base" game template to start new projects from.
Recycle old assets/Get free assets from online to save time.
Use only ONE unique mechanic in each game jam game.
If something stops you from continuing, drop it. Come back later or find a different route.
If you can get to the "The End" screen then it's a game.