The last time Hackerfall tried to access this page, it returned a not found error. A cached version of the page is below, or clickhereto continue anyway


Welcome to this post-mortem. My name is Aurélien Regard, I’m 35 years old, and I make video games. Three years ago, I’ve decided to make a commercial game on my own: code, art, music. It’s called « The Next Penelope », a kind of 2D F-Zero who would have merged with a PC-Engine shoot’em up.

The game was released on Steam in its full version on May, 29th and collected great reviews. And when I offered on Twitter to make a post-mortem, I’ve received tons of questions. The result is a novel-sized note, with suspense, fun, numbers and real nerdy adventures, in case this story could be useful to others.



  1. - Why I’ve made a game on my own.
  2. - The first ideas, the differences with the final result.
  3. - Playtesters are my bosses, and make the planning.
  4. - Working in the shitter, going crazy, well, a typical day.
  5. - Spread the word about the game, to become a product?
  6. - To Hollywood… in Germany! (conventions, press)
  7. - Cash, bucks… funding.
  8. - Release day (under pressure)
  9. - The sales. (Newcomer on Steam? You gotta believe PARAPPA!)
  10. - The choice of tools, the HTML5 shock.
  11. - Hey, what about the consoles versions?
  12. - Conclusion (you should leave now, Sir…)
  13. - Bonus: short questions, blitz answers!


Three years ago, Arkedo studio (Nervous Brickdown, Big Bang Mini, Hell Yeah!, the Arkedo Series, Poöf vs the Cursed Kitty…) where I’ve been working as a graphic and game designer shuts down. I co-created it, but wasn’t the boss. Thanks to that simple salaryman status, I’ve earned rights to severance pay: it’s time to try something.

Between the bonus earned when signing “Hell Yeah!” with Sega in 2011 (carefully saved) and my unemployment benefit, I feel like getting in the club like Jay-Z. Champagne, make it rain, Oreos all-you-can-eat! The first idea is then to make a team of former colleagues and create a studio of three people.

Back to reality will be the very next week, opening the book “Create your company for dummies” (for real). Check done, my savings are nothing but a joke, we could never live long enough to have half a game done without having to seduce a publisher in the meantime. Fact is, Hell Yeah! (published by Sega) just proved me I’m pretty bad at enduring the pressure of a big publisher. Time for plan B.

As creating a real studio is impossible, why not give it a try before looking for a job?

I’m a graphic designer with a touch of control freak, I compose music as a hobby, and softwares like GameMaker or Unity allow newbies in coding to create games. Also, given my unlimited admiration to Eric Chahi, I’ve had this fantasy since a teenager.

On the paper, it just requires a high motivation, and to keep it on the long term. Alright, let’s do this.

Bonus, being alone, I think I can make exactly the game I’d like to play, without caring much about sales, as my fees will be reduced to the minimum.


From the beginning, I know I wanted to make a top down racing game. It’s outdated, the public is limited, but I miss this kind of games. Remember the PC world three years ago, most of racing games released then are all using realistic textures, Unreal Engine 3, and blood stains on the track. I want to put a bit of flashy colors in all this, and also a simple but not stupid story, because solo modes in this kind of games are often tasteless. I watch tons of videos to inspire me, like this one from Wipeout 2D, very nice except for the camera that makes one feel sick once in full screen.

> Pictured: Wipeout 2D, an official and nice adaptation, but with a maddening camera as soon as it goes fast.

The first step is to validate that I really can learn using tools like GameMaker, Unity or Construct. I’ve already tried them for fun, but can we really make a pro project with that kind of stuff? From the first days I get a prototype, deliberately made to end in the trash. It’s just for fun, and mainly really lame. No gameplay is particular, the ship is as hard to handle as a drunk cursor, no graphics, and even less sounds. But it displays moving pixels. Hey, it’s promising, I guess.

> Pictured : Awful squares for the prototypes. The most masochists of you can see them moving in this one year old documentary.

I’ll move fast on the next 4 months, beginnings of projects always are easy. They consist in making mockups for the design (see below), and a huge game design document, here to get out of the simple Micro Machines clone idea.

Every ideas to add tension, strategy and originality are put one on top of each other, to not only be a retro tribute, always in the risk/reward model.

It’s decided, on the gameplay side all weapons will consume the life gauge and it’s going to be the skeleton of the game. And there will be bosses too, it’ll be half-racing half-shoot’em up to add noteworthy moments to the adventure. Graphically, I’d like to avoid pixel art seen a lot lately, so I’ll go for a fake 3D in flat shading just like the first StarFox that I love with a feeling as pure as its aesthetics.

> Pictured : On top left, the very first design of the game, very (too) close from StarFox

I call this candid period the “letter to Santa”: it represents a dreamed version of the game. We think like an 8 years old kid, it’s really soothing. And at that time we still don’t know some ideas won’t work together once implemented.

By the way and for the record, 50% of the game features have been swapped out during dev time.

Here are some that have been prototyped and removed:

More seriously, it’s really easy to have nice game ideas, everyone has some all the time. The real challenge, it’s for them to be consistent and balanced once the pad in hands. Hence the next paragraph.


Being alone means you can trust your ideas enough to rush, but also means that you are going to lose yourself in your game, and not being able to know what matters anymore from what’s only details. The only solution are playtests. In a real democratic studio, it takes a lot of logistics. Everyone has to agree on if what they did is showable, and that an unknown people can try it. For me, it’s been a lot easier.

Every week I’ve been showing ugly stuff and bugged prototypes to anyone who was willing to help me.

I really need to insist on this given the amount of solo projects I see being developed almost in an autistic way, and that are only pushing the problems to later. Show regularly what you do, all the time. It’s motivating to observe that the game is interesting people, and it’s useful to force yourself to owe explanations. The person plays and speaks out loud. You say NOTHING, you let him struggle even if it’s heartbreaking. As an alternative, you can send builds to voluntary people, and they fill in forms to smash yourself.

> Pictured : The playtester the most settled and brave of all: my wife. Just under, the kind of form the remote testers filled. Trying to only have a people test once the game (at different stages of its development), I have 200 people more to thank for their time and generosity… Thanks again!

Of course, this doesn’t prevent form having a view of the game. For example, there was a need to decide when half of the people asked for a closer camera, and the other half wanted a faster game, which was cancelling immediately the benefit from having a distant camera to have time to prepare for obstacles. (Note: the solution would have been to move the camera away before them, and to bring it closer in straight lines to give back liveliness to the action)

For the planning, I had a global plan for the conquest of the world on Trello, and the main steps were naturally determined by the conventions, the early access release, or basically by my strict budget. But I made the content in a very linear way based on the playtests, world by world, moving on to the next feature once the precedent one was validated by my players. I imagine the annoyed face of a project manager working in a real studio reading this: “Well this is pretty stupid, we can’t observe a deadline and a budget working like this!” No, indeed, we can’t.

Anyways, working alone, all tasks were mixing together and my personal life was so tied to my professional life that any try of an organization was bursting every 3 hours.

The good news is that it almost doesn’t matter. We will come back to it on the money part. (It’s long a postmortem, so as well put some suspense in it!)


Making a game alone is allowing to work in an organic way: every part is made with simple squares and circles, then you need graphics, then you need sound, then playtests. It’s all simple, no meeting, straight to the point. The only thing to decide on is the moment when you take care of what’s outside production, like business emails, communication or contacting players. If I had to make an average on the overall production, it would be: 11am to 5pm on emails (it’s been up to 200 emails a day some weeks), then 7pm to 4am actual work on the game.

> Pictured : start of development, in a real place with real people. Little Aurélien pretends to be working in front of the camera but isn’t fooling anyone.

In terms of working hours, let’s make a quick overview of the 2 and a half years of production:

Every work hour being good to save on transportation time, I’ve been less and less present at the Pastagames studio (Rayman Jungle Run, Pix the Cat) where I’ve been generously welcomed, and more and more in my studio bathroom (“Oh, hi honey, you need the toilets?”). It’s been good fun for Twitter, but instead I wish I had a garage like in a Silicon Valley success story. Or just place for an office at home (welcome to Paris, where the flats are super tiny!).

> Pictured : “You will found on the left our meeting room, on the right the break room, all in the same sockets drawer”I know some people will raise an eyebrow reading about that berserk schedule, either saying “oh poor guy!” or in the opposite “and one more indie guy willing to make people cry, they’re so annoying…”.In both cases, please keep in mind that, yes, it’s true (the game didn’t release itself magically) but especially that no one made me do that. It changes everything, and I took a lot of pleasure in it!

A doctor would say that skipping a night out of two is stupid, that sleeping makes us more efficient. My opinion is, any way is the good one if it’s avoiding you going bananas.

I don’t need a lot of sleep and mostly I couldn’t close my eyes being unhappy with my day. So as well not sleep that night, eventually fight with delightful night hallucinations, but snore peacefully the night after. I found it way more relaxing than staring at the ceiling listing to-dos like Arya in Game of Thrones. It mattered even more as it wasn’t a punctual effort but a real marathon.

> Pictured : “…collisions bugs, blurry backgrounds, particle emitter stucked, teleportation going crazy, include translations, collisions bugs, blurry backgrounds, particles emitter stucked…

During two and a half years, the enemy isn’t tiredness. It’s to demoralize or that the pressure makes you a huge moron in front of your friends, your family or spouse who don’t see you enough already. Every way is good to stay cool and keep taking pleasure.

If the idea of working so much scares you, don’t ever go alone in a production bigger than a gamejam game: you need a team. There really isn’t any scorn in this phrase, and I’m not going to lie, I would be unable to do this again in the next weeks.


As this part lakes a bit of ties with the start of the post-mortem, let’s just go back to the timeline, it’ll be easier.

After a bit more than a year, I have something I think is showable. I’m not happy yet with the result (it won’t be the case before 2 weeks until final release of the game), but as I’ve decided to focus my life on it for a good time, it’d be good to know what people thinks about it. And to start working on communication, because here too, it’s going to be a first for me.

> Pictured : One of the two promotional artworks for the game. I really wish I could make more of them, and more regularly.

To present my project being alone is very, very different to the team presentation I was used to. The critics aimed at a studio don’t quote your real name, they’re directed at a bit blurry company, where each member is a shield for the next one. One can put their fingers in the ears, it goes away in the end.

At the time, I know I can endure the potential lack of interest of people for Penelope, but what will happen if someone hates me as a guy?

Starting to show a game in solo, it’s either accepting to become a bit of a fictional character to protect ourselves in case of troubles (hello-FibreTigre-from “Out There”-whose-real-name-is-unknown), or as we say in the big times of reality TV, stay true to ourselves. I picked the second one, knowing that it would hurt a lot every time I’d read the subtle comments of the Internet. I only needed to know how to make the first step.

> poster from:  Moma Propaganda 

Before publishing the first Penelope video, I’ve read tons of post-mortems, guides, and tips from other devs. I can already see myself sending copy-paste of my press release, but I don’t like it so much.

When a PR guy says his company is doing a great project, he puts the talent on the others. It works like a friend recommending a new thing to a friend. Once again, being alone changes something.

Here a deliberately exaggerated example during a convention:

“Do you want to try the demo? You should, the teams behind it are so passionate and talented, you need to see this, they did something great! Have a sit.”And then, alone…“Do you want to try my demo? You should, I’m so passionate and talented, you need to see this, I did something great! Have a sit.”

Hard to deal with for someone with a normal ego, and it gives the sad impression to be begging for attention. In fact, I didn’t really want to become the product instead of my game and be telling whatever it takes to make it be known.

So the first video will be a very explanatory devlog, pretty boring long, with my regular voice, without saying the game is good. It’ll be up to people to decide if it’s worth sharing or making news about it. With a bit of luck, the interest for the game will be more sincere.

At the time, I had 30 times less followers than today on Twitter, but I still hope one or two will tell others, and so on. And it works. With the years, people kept in mind the games I’ve worked on, and spread the word. The devlog is spreading in France way better than expected!

But, for the American part (so most players in the world) at that time, no results even with the production of a translated version. Really NO results. How worthwhile that I force myself to tweet in English since day one…

My goal was never to make a game for the money, but with a game in such a niche as Penelope, French-only sales are the insurance to LOSE money. Damn it.


So, my kamikaze operation ‘no official press release, no mail to the press’ is doing pretty well in France, but gives no results outside of the country.

I sent then ONE mail to an American website, specifically - I’ve been reading it every day for years. It’s the indie little brother of Gamasutra, and I follow the news of creators I admire like Cactus, Derek Yu, Pixel, Locomalito or Kenta Chô there.

More than a PR attempt, I really wish to know what the site thinks about Penelope. John Polson is the editor in chief and hosts a regular column on racing games that are “different” since some months. Awesome luck, I’m right on the topic. When he’ll leave his job some time later, he’ll offer to help with the game communication in the United States for some months. He’ll say tons of smart stuff like:

‘add some random content’/ ‘do whatever it takes to have a simultaneous PC/consoles release’ / ‘change the UI to let space for the streamers’ head’…

I listen carefully but it doesn’t always match with my modest jack-of-all-trades abilities, nor with my funds. Especially as my playtesters are already asking for a multiplayers mode not planned to begin with.

He’ll insist even more on something else: I need to find budget for a huge international event, for example with IndieMegaBooth, an organization which bids on future successful indie games and highlights them during events like PAX.

To sum up, I need to try my luck at the cool kids club of the United States, and it’s intimidating me given my shitty English.

Flash forward, they’re selecting me based on application file to their first big venue at Germany’s Gamescom, the biggest video game convention in the world, far ahead of E3 with more than 300 000 visitors. A booth in an event like this is really expensive and my budget was around 4000$ for all Gamescom. It required my (awesome) wife to add her savings to mine to make this doable.

From Gamescom, everything changes.

With the help of John, direct contact with a really nice public, journalists but also manufacturers, there’s talking about Penelope in every website of the universe. And in every language possible (Eurogamer, Destructoid, Rock Paper Shotgun, Kotaku, IGN, Gamespot…)

Seattle’s PAX, which I’ll attend thanks to Neko to promote another game I’ve been doing the graphics for (Poöf vs the Cursed Kitty), will push this trend to the absurd. Taking advantage of it, I’ll speak about Penelope too, and end in articles till Fortune. Which makes no sense but still makes me laugh today.

Even though, press articles aren’t what I remember from these events. Sorry for the silliness to come, it’ll sound very douche for someone who weren’t there, but for the first time I felt like I was part of a huge, generous, badass family. I was surrounded with people not only talented but also understanding, sometimes having the same troubles than me, helping as much as they could.

Unsorted, thanks to Alexander Bruce (Antichamber), C418 (Minecraft), League of Geeks (Armello), Mi-Clos (Out There), Boneloaf (Gang Beasts), Rocketcat Games (Wayward Souls), Yacht Club (Shovel Knight), the Twitch team, Nintendo Europe team, a lot more I’m forgetting, and, let’s be crazy, thanks to Candy Crush artistic director.

Thanks for helping me avoid doing bullshit on business or PR, for giving me hints to make my game better after trying the demo, for streaming me in front of thousands of people, for giving me incredible job offers that made me laugh, just saying “hey, it looks pretty rad!”, or merely for giving me advice on staying healthy in body and mind during these long dev months. I’ve already spent quite a time in my cave making my game, and in the opposite this has been an incredibly powerful and motivating time.

Special thanks to Olivier from Seaven Studio (Inside my Radio), who found in 24h the missing money to finish my game. Family, I said it.


(From now on, I can write anything I want, everybody switched off given the length of the text)

So money. Let’s start by funding plan: first development year has been funded by my unemployment benefit. Then, I had to make my company, needed to get a devkit for console. Since then, my savings took over. When I ran out, I fend for myself: selling all my consoles, my tablet, everything I could find. I also did casual jobs like making logos, and thanks to this and presales on my website, it was almost enough for two years.

You need to understand that being alone makes it a lot, lot easier on this topic. The amounts are laughable in comparison with the costs of a real studio, and it’s a lot easier to find way to pay the rent. When an 8 people studio burns 50 000$ a month, the same amount would make me last 20.

This is what I was referring to when I said planning wasn’t at the heart of the project: nobody dies if the games gets a month delay. The only real milestones are the ones I decide on. Or the one my wife’s death stare gave me, depending on the weeks. In any case, after two years, I didn’t have anything left for sale, and sleeping only one night every two didn’t allow me to have casual jobs anymore.

> Translation : 30 months founding. From start to end : unemployment benefits, savings and casual jobs, advance on Plug-In Digital revenue, fixed-term contract at Pastagames.

If by chance you’ve read the previous paragraph on Gamescom, you can remember that money fell from the sky at that time. As it happens, Plug-In Digital one. This team has something particular for me as there job at heart is distribution, not publishing. Well back then. In exchange with rights on the PC version and a minimal percentage, they offered me an advance on revenue, with which I think I can last 6 months being really wise on my expenses.

This solution is tailor-made: I don’t want a big publisher to push me on marketing, being afraid to explode under pressure. I also want to use my own words to not oversell the game.

This is why I said no to the yet very cool Adult Swim (tbhit was actually super hard to turn down something that cool) and to some other publishers at the beginning of the project. And, Plug-In lets me handle everything as long as I’m not doing stupidities. And with the down-to-earth aspect of their job, I feel like speaking to open and direct humans – plus they’re nice. Bingo.


There’s been two releases for The Next Penelope. First the Early Access (including Story Mode) then the full release, with additional solo content, plus a local multiplayer mode.


Penelope was my first game made on my own, it would have been dishonest to promise to have it perfect on the first attempt or to sell it directly full-price. I was also really impatient to get new opinions through Steam forums, my playtesters patience and availability not being infinite.

Following Plug-In advices, I put the release around 6pm, French time, to hit both France and United States. Bull’s eye, the info is moving really fast, and people looks happy with the result. They like it even more not being a dubious prototype but already a real game, even not a full one.

Obviously, there’s also problems. From the first hours, I can enjoy liquefying in front of the first Twitch streams. They highlight problems no playtester ever had: code running too fast on screens with a framerate higher than 60hz, or in the opposite missed collisions when the weak computer of the streamer is hashing chunks of video in the background, and causes a jerk right on bad time.

Even though, the early version is really well welcomed, so well that a misunderstanding is setting up: some people thinks the game won’t change before final release in terms of contents, and already give review scores.

They are good, but reading on the “cons” stuff about lack of content when we are just producing some for full realease is really weird… Right, the game won’t be so long, but they haven’t seen everything yet!

That the nature of early access, everything’s blurry, relies on trust, and with so many frauds on the market, I can’t really blame anyone for judging what’s here and not what’s to come.

By the way, it’s not the only problem with an early release, as we’ll see later.


Welcome in a troubled 29th of May. Really troubled.

To sum up, it’s the only day in three years when I’ve been thinking the idea of making a game alone is the worst idea I’ve ever had: everything is too mixed up, always, everywhere.

Here’s what I was expecting from the press: a discreet release for the early version and then a bit more noise for the full one. What happened was the exact opposite.

Without hoping reviews on the first days for a niche game like mine, almost none of the websites who did previews made news to announce the game release.

It required a change of angle a week later, and throw some not really elegant e-mails title “Hey, my game got more than 90% positives reviews on Steam and nobody gives a shit! LOL!” to get attention. Some days later, finally, reviews are coming and filling me with infinite joy.

This does confirm my dislike for classic press releases: crowd yelling there is always a more important event to cover for journalists and for youtubers. In my case it was the Hatred release, Kung Fury and the E3 coming three weeks later.

I’m not looking out for excuses, any week would have been the same, and I should have forced myself to yell louder. Or smarter. Or both, working on the assumption that someone who liked the early access would be happy to know the full game is available. Well, if I could do it over again, I would take the help I’ve been kindly offered for promotion. But with way too many unanswered important emails already at the time, how to be able to coordinate this help? A tricky, infinite loop.

>Translation : 1122 unread e-mails. My phone screen at the time.


You may know it, Steam’s NDA makes it complicated to publicise sales numbers. However, I can give you the numbers for all PC retailers as a whole, around 4500 sales since January 22 without ever lowering the price of the game. As a matter of comparison, all middle-sized games I created back in the time of Arkedo (and had lower review scores) systematically sold beyond 100.000 units.

Sounds low for a game rated 9 out of 10 on Destructoid, 18 out of 20 on (Note: the biggest european website), 93% of positive reviews on Steam, and YouTube Let’s Plays with more than 100.000 views? Welcome to the indie PC scene of 2015.

More details and further explanation:

Most of copies sold nowadays are thanks to sales, for the sake of making a good deal before the enjoyment of the game. It may sound depressing, especially regarding their already affordable price, but it is the truth : there are way too many good indie games available for PC, and not enough demand for novelty which the player will have no time to play right away. However, Penelope hasn’t participated to any huge sale or any bundle yet. Out of respect for the first buyers, I don’t want that to happen too soon, but later on it’ll bring in more money.

> Pictured : Steam’s home, almost a miracle, if not definitely one. Penelope will stay there several days.

If you’re looking to compare with other titles, be wary of the numbers you’ll read on the excellent, for instance displaying 25.000 sales for a game will often mean 20.000 sales wherein the developer only earned a few cents per copy, due to being purchased as part of a bundle.

In that regard, I have investigated more thoroughly by the time of the release. Before they were sold in bundles, games of a size comparable to that of Penelope had the same amount of copies sold, even being featured on Steam’s home and backed by a huge editor for the promotional campaign.

Before you drink your fill of antidepressants, let’s get to the next part right away. In fact, there are not just Steam and GOG (which refused to sell Penelope, because, I quote “it’s not a real driving game”) out there. A part of my income after that will come from other types of sales : releasing the game for a console yet to be announced on the other side of the planet, partnership with streaming services, etc…

In the end, there are high chances that what other people see as tiny deals actually represents for me an income greater than everything else. That’s where the distributor comes in, negotiating said things in my stead.

> Pictured: The retail version of the game, available on Indiebox. Cool people who said yes in just a few messages on Twitter. I’ll be honest, I’m very proud to have a retail version.

Don’t listen to the first generation of successful indies on Steam, nor even the second. I admire the quality of their games, and often for their personality as well. However, they unwillingly have a twisted perception of Steam’s reality.

“A good game will eventually come out of the water…” No. Each month that passes lowers your chance to get success on Steam, if you haven’t already managed to get your studio noticed during the previous years, or that your audience isn’t already present following a Kickstarter campaign. Try it out yourself: randomly pick a succesful game. For instance, Thomas was Alone, or the next game from Behemoth (Castle Crashers). In one case, I love the game, in another, has it not already come out that I already love it. And they definitely bring something new to the video game scene. But for both games, I’m fairly certain that titles of this kind would go entirely unnoticed were they made by newbies today. And since I waited until Penelope to timidly dare put my name on a game, I am exactly the aforementioned newbie.

> Pictured : It doesn’t seem to be a lack of communication either, this video not being the only one that was a hit.

Penelope is a niche game, and the door’s already closed if you game doesn’t already contain numeral elements of surprise, of random, and of choices, inducing interaction with watchers for YouTube videos and Twitch streams.

Between the freemium model, and the superiority of video compared to written material that generously spoils their content, the most linear type of games; skill-based ones will surely and slowly disappear from Steam. They will only meet limited success with the speedrun community, or the older players. (That’s a loss, these are my favorite kind of games.)

The only odd cases you’ll encounter will be streaks of luck comparable to those of Apple’s AppStore. And if you have a chat with these very people, the more honest will gladly tell you that they have no real explanation on their success.

Regardless, I’m still very happy about my situation. I did exactly the game I wanted, some people like it a lot, and I barely ever had that good reviews on any other of my games. Also, the game just came out. Not talking about winning a lot of money, but I paid off my advance on earnings, and I will surely, eventually repay myself. In fact, after these two and a half years, I feel extremely relieved.

It appeared to me very important to warn the next challengers to Steam’s great lottery; yes, you are reduced to be glad to sell only 4500 copies at the launch of the game, including Early Access looking at my own case.

The only viable alternative for a niche game without randomly generated zombies, rogue-like elements, or craft, it’s cross-platforming. Hence the importance of choosing your tools well, hence the two following themes.


I thus used Construct 2 to make my game. Quick reminder for those who’re unfamiliar with these kind of visual scripting tools in general, these are software that allow to code by blocks. If some are limited to a certain type of game (RPG Maker, FPS Creator…) and can be quite frustrating, others push the idea very far. To the extent of letting you realise a complete game without any imposed boundaries, and zero limits on the gameplay side.

Hotline Miami was made using GameMaker, Blizzard’s Hearthstone uses PlayMaker for Unity. These are only two examples, but a lot of critical and commercial successes put great emphasis on visual scripting, which nature allows to super quickly shape the logic of the game. And that’s only the beginning. Take a look at “Adventure Creator” if you like games like Walking Dead, or again at “Buildblox” if you are rich (damn, 2000 bills for a license!) and that you like games with physics for mobile phones.

My treasured Construct 2 has not yet any big commercial success to put in the spotlight. Lots of my developer friends were thus surprised when I chose it to make Penelope.

> Pictured : And understandably: my game is, to put it in a nutshell, a HTML5 website that (quite well) pretends to be a common game with a clever compacting in .exe. I do not really regret this choice when it concerns the PC version of the game. As I often will gladly repeat in public, or even directly on the software’s home:

Construct 2 is the smartest, fastest, and most proficient tool I’ve ever used. I think it is absolutely brilliant in everything, in its logic just like in its interface.

Long story short, I’m totally in love with it, and I have infinite respect for its creator Ashley Gullen. The software reads my thoughts, contrarily to other tools I also use like GameMaker (with which I had started to make the game, by the way), or Unity (with which I’m playing around quite a bit these days). But then, why the hell am I going to use Unity for my next projects? Fairly obvious answer in the next paragraph.


Being based on HTML5, Construct 2 does not allow releasing a game like Penelope on consoles. That’s a very bitter constatation for me, but as of 2015, support of HTML5 on consoles is in reality either clumsy or non-existent.

However, releasing your game on just one platform is a luxury most small developers like me cannot allow themselves. Let’s be clear as water: I missed contracts that would have drastically changed my financial life because of that. No, not because of Construct 2, but indeed because of the HTML5 upon which it is based, and which it depends on. From improbable tweaks to drastic optimization, without counting the help I asked from the gifted programmers I know. I thought I’d find solutions along the way, but despite my efforts, nothing worked well enough to be proud of the result. Then… what about the WiiU version I announced?

Being that obsessed about a WiiU version may make some smile (yes, I see you.) But I have a very peculiar relation to Nintendo players. Ever since the first announcement, they always were present. Every announcement moved around the specialized websites, and I rarely have seen that enthusiastic people on Twitter. It was out of the question to let them down, be it at least to thank them for their incredible interest. Hence…

> Pictured: It’s a relief: while I’m writing, the very talented studio Blitworks is currently totally rewriting Penelope from scratch to C++, and making a great port / remake on consoles. It’s a gigantic amount of work, and they will even publish it, prioritizing WiiU.

If you don’t know Blitworks, you should. Meet the studio from Barcelona that ported Fez, Jet Set Radio, Spelunky, Don’t Starve, Bastion, Olli Olli or lately Super Meat Boy and Crypt of the Necrodancer. In other words, a team of geniuses which I heartily thank for coming to rescue me.

Aany questions on the console part? Don’t hesitate to notify them of your interest through their Twitter account or on their website!


Back in the time of Arkedo, I worked on commercial successes, but which reviews were good, but not exceedingly so.During these three last years, I worked on a critical success, but of which sales will be, a year from now, ok but not exceedingly so, with the condition of multiplying small contracts. And not too count too much on Steam.

It could be frustrating, but in fact, I have no regrets. It was really great. I learned lots of things in three years, more so than in the ten last years, and it allowed me to meet tens and tens of incredible people, be them developers or players. To put it in a nutshell, I’m not even afraid to go on, be that on new projects of niche solo games, or projects of games amongst a studio.

It was tricky, but I had a lot of fun.


Even if I were writing a novel, it would be impossible to answer all the questions I’ve been asked on Twitter, upon announcing this postmortem. Turbo catching up session!

Q: Do you think you sold less because of a disabled female protagonist?A: It is 2015 and I’m happy to believe that this has nothing to do with it. I’m even more so happy that very little people ask me that.

Q: Is it just me, or didn’t you say to me at some point that Danny Baranowsky would do music for the last boss?A: It is true, but at the end of the day, he has been too busy working on Crypt of the Necrodancer to be able to do that on time. Definitely my bad, as I only dared to contact him 2 months before the Penelope release. I sincerely hope we’ll be able to work together on another project. Q: In hindsight, solo or with a team for the next game?A: Currently, I’m working for my friends at Pastagames (Rayman Jungle Run, Pix the Cat…) but I have no idea what I’ll be doing in September. I have lots of ideas of games I would like to do alone, but the idea of joining a team (and of getting a salary) seems equally pleasing after having been an hermit and in a somewhat precarious situation all this time. If you are a studio and want to talk about exciting projects, you know where to find me!Q: Any hints to work faster?A: Not keeping the PSD sources of your Photoshop files is a good start. Flattening all your layers is a rather hardcore process I like to impose on myself. It forces me to instantly decide whether the image is good or not. It prevents me from pondering on a detail that no one’ll notice for ages. Of course, I never do that when I work for others, in the prospect that another designer wants to pick up where I left off.

Q: Have you had help from the french government?A: No, a one man company cannot get help from the CNC or similar things. (Note: CNC = National Centre for Cinema and the Moving Image, French governmental and culture institution). These are meant to help the creation of several employments. Q: How does advance and follow their planning whenever they’re feeling in a crucible ?A: By publicly announcing a release for a demo or an event. It’s crazy how fast you work once stuck in a corner.Q: What software did you use for the music of the game?A: I am absolutely not a legit musician, so pretty much everything at hand. Synthesizer apps on tablets, like the iKaossilator or the excellent Figure, and Ableton Live on the PC, with sometimes commercial licensed sample banks to go faster (but have much less fun and feel a bit ashamed.)

Q: Can we talk about your game’s lifespan?A: Yes, the game is short. Most people take four to five hours to complete the story mode (without counting the additional solo missions and the multiplayer, then), while some badasses wreck it in two hours. It first horrified to see this, but I then remember than small linear games never meet the correct length. They are either too short when you love them, either too long with boring moments when you’re tired of these. An even more so obvious impression given that most games base themselves on randomness to dodge that problem. It’s probably an actually very good solution for a next solo game.Q: Steam refunds, do they hurt?A: Not my game anyway. There are very few of them.

Q: Redbull consumption? Do you happen to have any arrangement with them?A: Hah, I expected that one. Several a day, everyday for two years. The article on Penelope or the interview on the subject of F-Zero on are only due to an enthusiastic journalist, which I by the way wish to thank. But no, I have no acquaintance with this brand. Too bad, I would have done a lot of saving!Q: What do you think of Construct 2’s performances?A: They’re really impressive, given that one knows what generally makes a game laggy or doesn’t, all 2D engines regardless.Q: The most annoying thing to do on the game? The easiest thing?A: The teleportation during races. I really struggled so that the calculation of positions was correct on the less powerful computers. There are numerous invisible triggers on the tracks, and if the computer lags right during the teleportation, it causes lot of hardly manageable cases.

The easiest part was the composition of the music score, which I had never done before, and was a breather at night, with my small headset on my ears.Q: I’m curious to know if the stress during production was rather financially-induced or technically-induced.A: Neither. The stress came from the will of doing something good, and feeling like shit if I failed after spending so much time on it. I was also very stressed by the idea of letting my wife down, as she supported morally and sometimes logistically all along the game’s production.Q: What are your sources of inspiration concerning the scenario, the music?A: I do not know if the game’s story deserves to be called a real story, since the constraint was to make an interesting story, but that could be absolutely skippable, and that was short, for players who really only just wanted to blow things up and race. The everything in a game that you can play in whatever order. It kinda limits the possibilities, but inspiration, I think it’s pretty obvious: Ulysses 31. As for music, anything that’s catchy and close to distasteful.Q: I would like to know the time spent in polishing the marketing assets, and the time spent polishing the game.A: Given that the game advanced by huge blocks rather polished indeed (cf the part concerning playtests), I skip the end of the question. Concerning marketing assets, I would have loved having the time to polish, but the final trailer you can see on top of this page was made in 48 hours, as I had to deliver it urgently for online retailers a few weeks before release.When I had more time, I came back a few times on some artworks, like this illustration on which I’d redone the face six months later as the style of the characters had changed in the mean time.

Q: How did you work up all your abilities (technical, artistic, game/level design) throughout the years?A: The answer is clichéd, too much repeated and used, but true. “By doing things and finishing them.” Even if the result is terrible, you still end up being slightly less bad in a domain when you finish a little something. It’s only by working it til the end that you can base yourself on that experience to become better next time. It’s worth all theory classes, tens of unfinished prototypes, and YouTube tutorials of the world. Another honest answer would be that I’m getting old, and I had the time to look over the shoulder of all the people I’ve worked with to steal their tricks.

That’s all folks!

Thanks for reading that novel-sized text ! I’m still at your disposition on or Twitter (@AurelRegard) if ever you still have questions !

> ENGLISH TRANSLATION: MANY MANY THANKS to Chloé “@BeignetSugar” Agostini and @wplanetary, who spent a lot of their time to translate this huge postmortem from french to proper English. If you notice typos, that’s because of me (I made some edits here and there), not because of them!

I would also like to warmly thank Peter Wingaard Meldahl, the author of the game Teslagrad, for his precious advice concerning this post-mortem during the Gamelag at Barcelona. Another dude who kindly took some time to teach me things. Yet another.

Continue reading on