The text appeared first on Boiling Steam.

1. We are in a stalemate

When Ekianjo approached me to ask about my Linux gaming expectations for 2021, I had to stop and think for a while.

It has been a good while since Valve’s Steam for Linux debut, Steam Machines, fully cross-platform Humble Bundles, partial support from big names like CD Projekt and their GOG.com, or Paradox and their Plaza. Since all those great ports released by Feral, Aspyr, Virtual Programming, and many less known porting heroes. Vulkan being pushed as a new industry standard and with it, the dawn of new fantastic tech, such as DXVK. Many of us prophesized the end of Wine era. Era of endless tinkering with configurations and workarounds. It was safe to talk about growth of the dreaded Linux percentage share in the gaming market.

And yet, just as unexpectedly as it started, it stopped. Less dedicated ports from big publishers, no fully-fledged support from SteamVR, GOG Galaxy, Epic Games Store. No interest from the biggest of fish: EA, Ubisoft, or Blizzard. Linux gaming only mentioned in context of Google products, Stadia and ChromeOS.

That is when we were pushed again into the loving arms of Valve and trusted companionship of Wine. Steam Play Proton has arrived. And with it? Support for thousands of Windows games, often right after their premieres. We could no longer count on porting studios but we could count on the company that made Linux their business over 20 years ago — CodeWeavers. And, as always, on ourselves. The community. Because it was yet again an enormous, valiant community effort that picked up Steam Play and carried it way beyond anyone’s expectations.

Look, I am not a doomsayer. But I am not an incurable optimist either. I think we all can admit that we progressed some, then regressed some. And then progressed again. It is true that the Linux gaming is right now in the best state it ever was. But Linux is still not easily available on gaming laptops worldwide. We are still less numerous than macOS users. And we are still hearing about games releasing on “PC”, and not “Windows and Linux”.

And that is how I am gonna begin this list. With a stalemate. The time of spectacular breakthroughs is over. And in 2021, I expect nothing ground-breaking to happen in the Linux gaming area.

2. More seamless Steam Play

Just like I said, Steam Play is the hottest topic these days and for a damn good reason. Even I, the old “no tux no bucks” principle follower, spend more time in Windows games (made easily available thanks to Proton) than in Linux native games. But Steam Play capabilities have been expanded. Thanks to the commendable community work tools like Luxtorpeda, Boxtron and Roberta make it possible to use not only Proton to run Steam titles, but also ScummVM, DOSBox, OpenMW, ioquake3, Arx Libertatis and many more. After trying it out I am completely sold on the idea.

And the Proton itself? From month to month its compatibility and efficiency grows.

All of that makes it pretty clear that Steam Play still has a lot to offer. And in this year I want to see what else that will be. Perhaps it will not be a proper support from the nasty anti-cheat mechanisms, constantly locking out Linux gamers from multiplayer sessions in Fortnite and PUBG. But I am sure that it will be worth waiting for.

3. Wayland gaming is a thing

Recently we have heard that NVIDIA is putting together final work towards supporting XWayland. The Wine project does not waste time either — Collabora has shared an update on this topic just days ago. Nearly 10% of GamingOnLinux users who participate on the website’s survey declared the use of a desktop Wayland session. Maybe you and your friends do not use it yet for gaming but I guarantee there will be someone on Reddit who does, sharing their experiences in the comments.

We might be waiting for Wayland for well over 10 years to truly introduce Linux displays into modern times. But it is happening and it is only a matter of time until all popular game engines and libraries stop depending on X server in their new releases.

Who knows? Perhaps 2021 will be the year for many major distributions (hi Ubuntu!) to push with new strength for Wayland.

4. Game development on Linux is becoming more viable

Thanks to the unstoppable power of open source, Linux was always a safe shelter for tech geeks, programmers, developers. People always were and always will be developing games on Linux. This list, curated by Cheeseness confirms it.

For many years engines and frameworks such as Love2d, RenPy, Cocos2d, MonoGame, FNA, Cube, HaxeFlixel, Irrlicht, Kivy, Twine, Ogre3D, libGDX, Torque were the weapon of indie developers. But only quite recently this group was joined by the popular in the industry editors. The famous Unity Editor is pretty much THE game making tool, holding majority of the game development market by sheer game numbers. Defold and Godot also have quite a lot of Steam, Google Play and App Store titles. And there are others, like Stencyl, RPG Maker MV, GDevelop, Visual Novel Maker.

The super-seller Valheim proves one more thing: Linux is not just a platform for developing games, it is a platform for making commercially successful titles. There is more to come, I am sure.

5. Cloud gaming on the rise

Like it or not, the gaming market’s latest craze are cloud gaming services. Big tech names: Microsoft, Google, Amazon are investing insane money in it. The streaming technology and the internet infrastructure around the world have matured a lot in the last decade. Many products, including Stadia, GeForce NOW, Shadow, Luna, Blacknut offer first class Linux support. So why not use it?

With to this technology, games otherwise unavailable even to Wine and Proton are playable right from the Linux desktop. In my case, they are, among many, Call of Duty: Warzone or Destiny 2.

If you are not convinced, you can easily give the streaming a go for free, by utilizing Steam Remote Play, Parsec, or Rainway.

And in 2021 I expect more of game streaming services to be available on Linux.

Special thanks to Ekianjo.

Introduction

Hi, I’m HardPenguin! A solo indie developer specializing in programming and QA from Warsaw, Poland. For my day job I currently work with Python in the corporate business world. In my free time, I work with C# and Unity.

Nearly two years have passed since I completed and published Sharp Turn, my first semi-serious game development project. What was supposed to be a small training task, started solely to learn a new programming language, ended up consuming three months of full-time work.

In this article, I’m going to highlight what went well during the production, what did not, and what other beginners could learn from my journey.

Sharp Turn is a text-based role-playing adventure in which the player becomes one of the three available characters: Barbarian, Enchanter, or Sneak. The main objective is to find and help your lost partner. In order to do so, the player needs to stay alive during randomly drawn encounters. Choosing the right action while making sure not to run out of food are both the priority. The game features permadeath, so if you fail, you start over — which is basically the core mechanic behind the game. Try, die, repeat, until you win.

Was it the first game I have ever made? No. A couple years ago I spawned a turn-based minigame in LÖVE (Love2d) using Lua. As a kid, I also produced a short title in RPG Maker 2000. Both of these were significantly smaller though, and much simpler to create.

Plans & beginnings

Why make a game?

First of all, a bit of context.

In November 2018 I was in between jobs. My goal was to keep working in the video games industry. I wanted to be as close to games as possible, preferably making them. However, the mismatch of my experience with the direction I took was posing a problem: so far, I worked “only” in a digital distribution company, not in a game development studio.

Although my previous job involved, among other tasks, resolving OS-level issues in hundreds of games released on our platform and doing overall QA for them, in order to become a full-time game developer I needed quite different expertise than the one I gathered for over 4 years there.

After browsing gamedev job boards for a couple days I could confirm what I already knew — Unity with C# and Unreal with C++ were the two most desired skill sets on the market. Since my passion lies mostly within small 2D indie games rather than large 3D productions, Unity seemed like an obvious choice for my next career step. I planned to learn it by creating a game, while enriching my portfolio at the same time.

I had some affairs with C and C++ in the past, but they were rather short-lived. Unity meant working with C#. To fill the gap in my knowledge I went to Tutorialspoint, which I recommend to any coder looking to learn a new programming language or framework in the most straightforward way possible.

Even though my friends definitely do not think of me as a Microsoft enthusiast, C# turned out to be an okay language. As a Linux user, I am grateful to the Mono team for their effort over the years to make C# easily accessible on this platform. Their documentation is also a pretty solid resource, along with the official Microsoft docs.

Long story short — one would think that after finishing all of the language tutorials I would just move on to Unity, right? Unfortunately, that is not what happened. For some reason, I decided it would be a good idea to finish a small game project utilizing C# in a more “basic” form first, just to strengthen the foundations I had learned.

After taking a peek at C# SDL (which looked outdated and abandoned) and somehow failing to notice the well-maintained SDL2# (I have nothing but respect for flibitijibibo’s awesome work!) I suddenly had another idea. I decided to create a game in… a GUI framework.

Inspirations

During my childhood years, I spent many blissful hours playing a niche freeware Polish game titled Biznes Filmowy (Movie Business). Made by the independent developer Tomiga, it was a tycoon simulation released in the early 2000s. In the game, the player was responsible for producing a Hollywood movie. This involved selecting the cast, picking the director, special effects budget, and managing the marketing campaign. The result was calculated by a rather mysterious RNG-powered algorithm that took into consideration properties such as acting level and popularity of the hired actors.

Considering that the game could be played in multiplayer hotseat mode, it was a big hit in my friend group. If you are interested in it, the Movie Business series has been translated to English and is still being maintained. You can download both games for free from the author’s website, or just the sequel from itch.io.

This bizarre, yet brilliant oddity of a game could be played in a Windows window only. It used regular GUI elements, such as buttons, checkboxes, radio buttons, and others. I suspect it was simply coded in MFC. And even though the limitations of this technology were apparent, the gameplay was super fun. It was Biznes Filmowy that inspired me to do something similar in Gtk# — and thus the Sharp Turn project was conceived.

My other significant influences were:

• Hand of Fate — which inspired me to use the idea of randomly drawn encounters (just like cards from the deck!) with multiple choices deciding the outcome,
• Fallout —which was utilizing meaningful abilities that affect how the situation unveils for the player,
• Tajemnica Statuetki —which presented the unique art style achieved by heavily pixelating photographs (which were taken during the vacation by one of the developers).

I would like to highlight Tajemnica Statuetki (The Mystery of the Statuette) since it is yet another curious Polish title. This first person adventure game was developed in the early 90’s by Metropolis Software. The studio founded by today well-known, at least in Poland, industry personalities: Adrian Chmielarz (currently of The Astronauts) and Grzegorz Miechowski (currently of 11 bit studios).

Ow, that hurts

Time management

When I began the project, the premise was pretty simple and the planned scope was small. The estimate was that the game would take me about one month to make. This included 3 weeks of coding in a language I did not have any experience with, and 1 week of creating some simple visual assets.

If you ever shipped any software product you can already see how wrong I was. This kind of thinking ended up being the biggest fail of the entire endeavour. Yes, the code took me 3 weeks altogether — just as planned. But the assets? 3 long months.

If you are interested in game development, and especially if you are still fresh in the subject, I am sure that you hear lots of advice from industry veterans. Have you heard this one? “It does not matter how much time you estimate your project will take — always plan for at least two (or three or four) times as much”. This pearl of wisdom turned out to be painfully true for me.

By the time I finished my project, I was no closer to mastering Unity than when I started. At this point I was running out of money and had to start a new job.

Choice of tech

The second pain point was certainly my bad decision to make a game in a framework not designed for this purpose. Gtk# is a solid toolkit for creating day-to-day application software, or specialistic and productivity tools. It offers a straightforward and logical way to send information between application windows via events. It blends in a nice, consistent manner with the rest of operating system applications on Linux and Windows.

But because of that tight system theme integration, Sharp Turn ended up looking differently on each machine it was played on. A most common example was the use of a dark window theme. Such preference was the cause for a bug causing some sprites blending in with the background.

The choice of Gtk# also forced the user to install additional redistributables, as without them it was not possible to even launch the game. This was confirmed by the early testers. I was not able to alleviate these issues at the time. Common game making tools such as Unity have already resolved the majority of such basic problems, which is a good reason to use them instead of reinventing the wheel.

Inconsistent art style

I could not afford to hire a qualified artist so I embraced the common saying: “beggars can’t be choosers”. By that principle, I downloaded tons of free photographs, sprites, and drawings that even loosely matched the medieval fantasy setting which I had in mind. After making sure they were explicitly listed on a permissive license, I processed them to match the project’s “vintage” vibe, and scaled the dimensions properly.

The end result is better than I expected, yet still it clearly shows lack of consistency. The culprit was likely the vast variety of source materials, as well as my limited abilities.

From this point, I would like to extend my thanks to Marcus J. Ranum and his freely available collection of fantasy stock photography. And I want to thank every artist sharing their work for free. I would not be able to finish this project without your selflessness and generosity.

Doing everything by myself

From top to bottom, I was responsible for: managing the project, writing the code, designing the gameplay and the UI, testing, debugging, inventing and writing the plot, processing game art and producing the music background.

That is enough work for an entire game studio, even if we assumed that one person did more than one thing. And while handling it on my own was satisfying, it was often frustrating as well. As stated before, I am a programmer. Not a graphic designer — despite my vivacious passion for making memes in GIMP. And certainly not a writer.

Work on the graphic assets took 3–4 times more than the code. For most of development time, my routine looked like this: Every day, I was waking up, eating some breakfast and ingesting some caffeine. After that came searching the internet for freely accessible images. Then I was spending hours in GIMP by scaling and tuning them, making them look “more retro” by using the “pixelate” filter, and reducing their color palette.

Still, making the visual art was tedious but easy. The biggest, most time-sinking challenge turned out to be writing. To this day, I am not sure how I managed to write over 40 unique encounters without going crazy. It was a demanding, creative work, very often conditioned (or, rather, limited) by the available sprites. One thing I know — I don’t want to do it ever again. Humbled by this experience, I can say for sure that my respect for writers has grown exponentially.

Not so bad after all

A chance to learn

From beginning to end, working on Sharp Turn was a big opportunity for self-improvement. Which is good, considering my long-term objective to get hired as a full-time game developer. Some things which I have learned, often from scratch, were:

C# programming language

Coming from a scripting languages background I expected to be eaten alive by a new language from dreaded Microsoft. In the end, it turned out to be easy enough and in fact pretty similar to coding in other languages I already knew.

My own limits

At work, being able to evaluate what I can do by myself and what I need help with is crucial. This project showed to me clearly that I am not a one-man army and probably will never be. It is good to get backup from outside, especially for my own sanity.

New organisation and motivation techniques

Even when you are working for yourself on your own terms, finding motivation to achieve your goals can be difficult. That’s why lots of tricks help. In my case it was meditation, caffeine, and splitting work into smaller bits. One particularly revealing discovery to me was “warmup”. Before facing a big, intimidating task, I first did a smaller, easier task, possibly one that does not involve much thought. After a while, I entered the state of flow, and more challenging tasks became much easier to do.

(Hopefully) better project management

After spending over three months on something that was supposed to take one I really hope I will not underestimate project scopes and timelines so grossly again. I hope. Task trackers and such help — I used Trello.

Publishing on itch.io

itch.io is truly every indie developer’s friend. It comes with a thorough documentation, easy to use interface, plus a bunch of publishing and analytical tools. Uploading builds from the command line tool butler to then run them from the itch app was a breeze! Perhaps this experience will translate nicely to publishing on other distribution services in the future, who knows?

Free art is a blessing

I can not repeat this enough: content posted online on permissive licenses is a gift from the internet people for the internet people. Authors of such content usually do that with the intent to help others. Keep that in mind and use it as help with your work: with changes, “as is”, or as an art reference.

Leveraging existing talents

As a kid I received musical education and played multiple instruments. This came in handy when I had to adding some music to the game. I fired up LMMS to create a couple instrument lines, edit them a bit, and then export them as a track. Sure, I did not create a masterpiece. But it is very important to note that creating music took me about a day, especially compared to how long it took me to develop other aspects of the game.

Testing

There are not many elements of the video game production process that are as crucial as testing is. It not only helps you deliver a better, more polished product thanks to bug reports, it can also be a source of opinions on whether the game is actually fun or not.

My prior experience with the games QA served me well here. Both organizing tests and collecting feedback with Google Forms went smoothly. Testers were mostly volunteers from Twitter, sometimes fellow game developers. Without their work and perspective from outside, many issues with the game would go unnoticed until after release.

Summary & advice

I would like to close this text by outlining some advice points which I have for other inexperienced independent game developers, now that I was able to produce a game from the beginning to end myself. First of all…

Know your strengths

Not only know them but remember to use them as well. My advantage was prior experience with coding and music making. If I decided to make a game emphasising on those two elements, I know that I could end up with a better end product with less work. Instead, I spread myself too thin and wasted too much time. Which leads me to the next point…

Outsource as much as you can

There are very few of us that can do it all. Once you identify your strong points, think what you will need help with. In my case: I know there are artists out there which are in the same situation as I am — being fresh to the gamedev but having experience in their own field. Bringing them in to your project will benefit not only your game but also their own portfolios. But how to do that as a solo beginner?

Ask other developers, artists & community for help

The indie development community is vast and usually very helpful. Some people will want to help you out of kindness, boredom, satisfaction from participating in a game project, or simply for a mention in the credits section. It is okay to reach out for volunteers. That was the way I was able to get my game tested before I released it to the public. Ask for learning resources or for recommendations of specialists that might be interested in joining your project. The internet shall provide.

Exchange favors

If you cannot afford hiring an experienced artist, a musician or a coder, you can also offer an exchange of favors. One of my Twitter gamedev mutuals was kind enough to provide proofreading for the story text. It was crucial in the context of a text-heavy game. In turn, I helped a little bit with the Linux aspects of his game. Voluntary cooperation ending with mutual benefit.

Remember: doing poorly is better than not doing at all

Sharp Turn development was a bumpy road. While I am not overly proud of the end result, I am very happy with the fact that I produced an entire, even if small, video game. In the end, I learned a lot and proved to myself that I am a capable game developer. Because of that I can say with confidence that it was a priceless experience to me. It could never happen if I turned my back on it or decided to move on to the next project.

What comes next

As I am writing these words, I have already ported Sharp Turn to Unity. You can play it in browser right here. You can also download an offline build for Windows or Linux.

Learning Unity is an awesome journey with loads of learning resources and documentation available online, totally free of charge. That is what I am gonna focus on from now on. Hopefully, each game I make will be easier to develop than the current one.

I hope you found this text useful, or at least, entertaining. Let me know what you think and visit my Twitter profile.

Special thanks to Egee and Sara.

Those who know me are already well aware that I talk a lot about cloud gaming. As a full-time Linux user with no dual-boot to Windows I am always looking for alternative ways to play PC games that are not supported on my platform of choice. That’s why renting a Windows server with a powerful, dedicated GPU for a monthly fee is a practical choice for me.

The concept is simple — I send the mouse, keyboard or gamepad input to the server and it streams the video and audio back to me with unnoticeable delay. It works really well, mostly thanks to highly efficient software solutions and fast internet connection, both of which are easily available these days. That way I don’t have to set up a physical Windows machine in my house.

My cloud gaming service of choice is Parsec. Their partnership with cloud provider Paperspace makes it easy for me to rent and manage a Windows gaming server right from their web interface. With Paperspace I can pay per hour, instead of paying for entire month upfront, which saves me money if I don’t want to play much that particular month.

However, I have recently noticed that a competing service called Shadow is expanding to new territories and slowly getting some popularity as well. For a monthly fee it offers you unlimited play hours, which would make it a better choice if you plan to use cloud gaming as a complete replacement for gaming on your local computer.

Which service is better?

Currently, both services cover my basic needs, which are:

• a Linux client app
• availability in all European countries

I am not going to go into differences between offered cloud machine specs (GPU, memory etc.) here. Those are listed on both Paperspace’s and Shadow’s websites and can be compared in a pretty straightforward manner.

What interests me much more is: which one gives you more play time for the same amount of money paid per month?

Renting a gaming server with Windows is significantly more expensive than paying for movies and music streaming services. In Poland, where I live, those are approximate monthly prices I would pay for an unlimited access to:

• big collection of streamable movies and shows, Netflix: $10–$15
• even bigger collection of streamable music, Spotify: $5
• cloud gaming service, Shadow: $35

First thing you may notice is that this is quite a disparity. Is it possible to lower this expense? Perhaps if we didn’t intend to use the cloud gaming service all month long? Well, I did some math on the subject and as it turns out… It all depends on how you’re going to use the service. Let’s have a look at how Paperspace compares to Shadow’s offer.

Hours by price comparison

For the sake of calculations, I am gonna use prices for Paperspace P5000 and Shadow’s UK offer. Exchange rates between currencies used here come from July 15th, 2018.

First, I took Shadow’s price as a starting point. After all, that’s the goal: to get unlimited hours for the best price possible. Right?

So let’s say I wanted to spend the same amount of money on Paperspace as I would spend on Shadow: £26.95, which in US dollars would be $35. Storage alone will cost me $7 (this value may vary depending on how much storage you want on your cloud machine). That leaves me with $28 left to spend on play hours that month.

How does that amount of money transfer to gaming time ? Well, that would be 44 hours monthly, or 1 hour and 27 minutes daily if we’re gonna play every single day using Paperspace’s server. It’s not tragic, but it doesn’t seem much compared to Shadow’s unlimited hours offer.

Test case #0 — Steam:

However, is that enough? Not if I wanted to play all my Steam games using only cloud gaming. My Steam stats say that I have spent 39 hours and 48 hours in all games played in last two weeks (14 days), which would give me 85 hours and 17 minutes spent in games monthly. That’s a much bigger need that what Paperspace would be able to satisfy for the price of Shadow’s monthly subscription.

So the verdict here is a NO for Paperspace.

But what if I didn’t want to do all of my gaming using a cloud gaming service? That’s where Paperspace’s pay-by-hour model shines. Let’s move on to another two of my use cases.

Test case #1 — Realm Royale:

Like I said in the very beginning of this text, I primarily use Linux for everything, including gaming. I don’t need to run all of my games using cloud services, since my computer does fine in most of cases. But every now and then, an interesting Windows game pops up and that’s where a reliable way to run it really comes in handy. For me, such a title is Realm Royale, a Paladins take on battle royale genre.

How much time exactly I spend on playing it? Turning to stats provided once again by Steam, that would be 26 hours and 34 minutes monthly. Not only is that enough, but I also end up with some unused value, and here comes the best part.

With Paperspace’s pre-paid model, you can use the remaining money for either more gaming during the same month or for play hours in the upcoming month! In this case, playing only Realm Royale on a cloud machine leaves me with $11 or 31% of original monthly cost available to reuse.

In this scenario, we can surely say YES to Paperspace.

Test case #2 — Sleeping Dogs:

Another good example from my life would be when I played Sleeping Dogs. It’s a singleplayer open world game about an undercover policeman in Hong Kong, that takes over 20 hours to complete, if you don’t count the DLCs and aren’t interested in finding all of the collectibles. Unfortunately for me, it doesn’t support Linux. Too bad, because I love a good police story!

Cloud gaming came to my rescue here. I spent about 21 hours on completing my playthrough for this title. The month I finished Sleeping Dogs I wasn’t interested in using any more of my cloud computer, so that month I used it only for 21 hours.

How much did I have left? Nearly $15 or 41% of original monthly price that I was able to save for my future gaming adventures with Parsec.

Once again, we can give thumbs up and confidently say YES to Paperspace’s pricing model.

Wrap-up

Like you could see, choosing a suitable offer between those two services largely depends on your use case. Doesn’t matter if you’re a Linux gamer or maybe own a low spec gaming rig, I wholeheartedly recommend giving cloud gaming a go. There are also other competitors in this area, such as LiquidSky or Vortex. The selection is more broad than one would expect and anyone should be able to find something that matches their needs.

The spreadsheet containing calculations described in this article is available here (it can be opened using LibreOffice). Feel free to modify it to see for yourself which service would be a better fit for you!

If you have any questions or would like to add your opinion on the subject, you can reach me on Twitter or leave a comment below.

Big thank you to the Parsec team for supporting Linux with their client app and to Parsec community for inspiring this post!

EDIT:

I just learned that Paperspace’s prices are going up by about 20%. Even though it is not exactly good news, the advantage of their flexible model is still there and savings proved by my test cases are still true.