Greenlight has been broken for a while

Usually I tend to stay away from news that just broke, mostly because it seems so click-baity to write about it and often people are just panicking for no reason until a few days later when things are explained to them again more clearly. But this time, I just had a thought and wanted to tell it to whoever may listen.

News broke that Valve will be shutting down its Steam Greenlight system in favor or a new system called Steam Direct. With Steam Greenlight a developer had to pay an entrance fee of $100 and could then submit as many games as they wanted. From there on, the game was put into a queue with all the other games in the Greenlight system and people could up or down vote them, as well as leave comments. If a game reached the top 100 of the queue, there was a chance of your game being evaluated and Greenlit – the game was ready for release on the Steam platform. Steam Direct removes this voting process, as well as the random factor of getting picked or not. You fill out a form, you pay per game a fee and if everything is in order, you get your game onto Steam. It’s simple and “direct”, and a process one might have expected from the start.


Originally, Greenlight’s intention were really good, but similar to how people tend to find and use exploits in games, they also found and used exploits for Greenlight, until it became the norm and the system was left broken. The voting system for Greenlight is basically a weak form of a scheme used by drug dealers and other shady sellers: “If you can bring me X amount of buyers, you’ll get a shot for free”. The voting system was supposed to be a useful thing for both Valve and the game developers. It was supposed to be a place where you can drive attention towards your game, all the while getting it up the ranks and eventually published. But since Valve is not evil™ they didn’t force the voters to any commitment, which is where the exploitation began. Game developers soon realized that all you needed for passing Steam Greenlight was to get a lot of people to just press that “Yes” button. It doesn’t matter, whether those people actually like the game or not, they just need to be convinced somehow to press that button and optionally leave a comment.

Besides the issue of getting low-quality games onto Steam, the bigger issue for Valve is, that these boosted games won’t generate enough revenue, because even though a few hundred or thousand people said “Yes, I’d buy this game if it were on Steam”, only a very low percentage actually bought it once released. So Valve is left with your $100 and 30% of a couple of sales, but has to provide a very highly available service, fast connection, lots of bandwidth, update procedures, dealing with complaints, refunds, etc. In the long run, this will not work out for Valve, which is in my opinion why Steam Greenlight has been broken for a while and is now being shelved.

In most conversations I have read surrounding Steam Direct, the main focus has mostly been on the mentioned Dollar values. “$5000 is way too much!”, “$5k will kill small indies!”, “So expensive, I live in a 3rd country!”, you get the idea. While it certainly is true, that games of the concept mentioned above will have a hard time justifying a higher entrance fee, those are the kind of games Valve has little interest in publishing on Steam to begin with. If your game doesn’t sell, your game isn’t generating revenue for Valve to cover their costs. Since the voting system will be removed, game developers aren’t required to provide a pool of potential buyers anymore, as such Valve runs a higher risk of not getting any sales in at release. To counter balance this problem, they have to raise the entrance fee and restrict it to each game. Remember that when Steam Greenlight first came out, there wasn’t even an entrance fee, those $100 were only added after the fact, due to people just submitting anything and everything, as such it acted more like a spam-filter. The new fee, however large or small it will be, will have to pay for the actual services that Steam provides, but again Valve is not evil™ because they’ll most likely let you recoup that fee, by not taking a cut or a lower percentage cut for the first game sales. So if you are smart as a game developer, you’ll simple write that fee off as an investment, then try and bring enough buyers to the Steam platform and you gain back the invested money.

In conclusion, I find Steam Direct a lot simpler and clearer. You’re no longer at the mercy of people voting for you. Over the years I’ve voted for many Greenlight games and rarely made a purchase afterwards, as such I’m glad that this broken system is being replaced. The discovery phase and additional exposure you could get through Greenlight might take an initial hit, but I’m sure Valve will introduce a different section if it’s really missing.

Let me know what your opinions are on the topic, by either leaving a comment below or tweeting me @DarkCisum.

SFML News – Week 40 (2013) – 18 (2014)

Believe it or not, but here’s a new SFML News post. About half a year is it since the last news post, to all the regular followers of the SFML News I apologize. Writing the SFML News is a time consuming thing and with my internship my focus has been heavily shifted toward web development, but as you may read further down, SFML hasn’t been forgotten.

Given the huge gap, I’m sure you’ll understand that I won’t be presenting all new projects, but well see for yourself, the list is still huge. And because of that I decided to put the other news bits first.

Continue Reading “SFML News – Week 40 (2013) – 18 (2014)”

SFML News – Week 30-31 (2013)

Whenever I collect topics for forum post, you can be sure, that I’m missing one or the other by accident. If I notice it, I’ll make sure to include it in the next news.


Mega Man

Are you a fan of Mega Man? Then you should already know of this awesome project, otherwise what are you doing? Go check it out! ZackTheHuman is putting quite an effort into it and personally I think it totally pays off. Unfortunately I haven’t gotten around to build it, so I haven’t played it myself. It’s definitely worth keeping an eye on that project.

Continue Reading “SFML News – Week 30-31 (2013)”

SFML News – Week …-29 (2013)

Although I’ve once wanted to make weekly updates on SFML News, the amount of content go more and more, while my time got less and less, thus I eventually stopped. Since there have been many new posts on various projects in the last few days, I decided to make another quick update and present some text and images and because there is so much nice stuff, I’m not going to limit myself on a fixed date in the past. Unfortunately I can’t write about every single project, otherwise this blog post would never end!


Academic project: Simulation of Scorpion & Mice

This project was presented on the forum for the first time on the 7th if July, it’s an academic project from the Swiss university EPFL and created by the Mac OS X guy behind SFML, namely Marco Antognini aka Hiura. Now the aim of the project was to simulate full life cycles of mice as the prey and scorpions as the predators. The hard part of the project seem to be lying with programming the neural network, rather than getting something displayed with SFML, but in any case it’s quite nice to see SFML being used in notable universities and projects. With this project the use SFML wasn’t just temporarily and won’t stop now, but Hiura says that they’ll be using their SFML based setup in the future, so the next class might be actually learning SFML!

Postmortem – One must die

Postmortem has been first shown over a month on the SFML forum, but it went into Closed Beta just recently and now even tries to get the green light on Greenlight – although I’m still wondering how they’re going to get a free game on Steam…
Here’s a quick description from their website:

Postmortem is a FREE Indie narrative-exploratory game that will stress your moral compass this August 15th! Think The Walking Dead meets Home and The Last Express, with a dash of To The Moon – a perfect mix of exploration, conversation, meaningful choices, discovering clues and some puzzles.
You are an agent of Death sent to take ONE life from a cast of influential and ambitious characters at a charity Gala, in a rich and complex setting of industrial-revolution, conflict-torn country.

They are of course running their own website and a nice dev blog. It’s nice to see, that they’ve already gotten an article on US Gamer. Personally I think it’s an intriguing idea and definitely worth checking out!

Colonies – A Retro Sandbox Survival MMO

Not a new game presentation either, but they’ve now gone into Closed Alpha phase and since it looks really well polish, I just had to include it here. As the title suggest this is going to be a sandbox game with RPG elements and it should be an MMO or at least O(nline) as well. You should be able to settle, create a village, and defend your friends from the horrible creatures who lurk in the forests. You can change the landscape, farm, or hunt in order to survive the conditions, etc. Whenever the word “MMO” drops people get suspicions, since it’s hard to get done right and needs a lot of skill and time, but feel free to follow Jungletoe’s dev blog and don’t forget to stop by their own community.

Basis: Bone and Sprite Integration System

Some of you might have noticed the Spine project from a few month back, which also has support for SFML and I’ve even helped on Kickstarter. While Spine costs you some money Basis doesn’t cost you anything. Both mentioned applications are for 2D animations of skeletal and sprite creation. Although the tool is free the development still needs to be financed, thus they’ve started a Kickstarter project as well and they could really use some help! A download can also be found on the Kickstarter page.

Council of Torment – 2D dungeon crawler with a rich storyline

Although Council of Torment is still in rather early development it still deserves its place here.

Council of Torment is a top-down 2D dungeon crawler inspired by Ultima IV, with a very rich storyline, gameplay elements that’s been under development for 10 months.

From the description on the forum it seems that a big portion of the development will also go into the story. There’s not much more to talk about for now, but it’s certainly a project to keep an eye on.

Rock, paper, scissor.

I think, I don’t have to explain how this game works or I at least hope everyone knows “Rock, Paper, Scissor”! Although there still seems to be some animation issue, it works fine and can be interesting to try and figure out, if it’s based on some pattern or simply randomly generated. Not only can you download the game itself, but if you want, you can also look at its source code.

Metanact – Filesystem spaceshooter

While nobody has ever heard of Schnommus aka Seb, he seems to have been reading the forum for quite a while now. Judging from his progress on Metanact so far, he also seems to know quite a bit on programming. Metanact’s goal is to explore and conquer your own computer’s file system in a 2D environment. So while playing, the game will go through your file system and pick out various file names and represent them as enemy. While this game reminds me of other “file system” games, which would actually delete the files you’re shooting (e.g. Lose/Lose), Metanact is not such a game. It does not do any harm to your file system. You can find out more about Metanact on their website or even help Seb getting a bit more money together for some more nice content on indiegogo.

2D Platformer using SFML and Box2D

It always amazes me how some is able to create such a rather nice game as their first project. I still haven’t managed to get my own little platformer running (most because I end up doing unnecessary stuff) and others just go ahead and get it done in their first project. Though it seems, he’s using the Public Domain tile-set, I’ve found yesterday as well – visuals are something that can blur the actual vision between good and bad.


One of my favorite game in the current list is GoPlanets. The idea is rather simple but much fun! You’re red and can send out a certain amount of ships to other planets to take them over. The numbers on the planets display how many ships you can send and based on this number a given percentage will be taken away when sending out ships. The number increase over time automatically, so the longer you wait, the more ships you get, but keep in mind that the computer will try to win everything in the meantime as well. Although the current AI isn’t really challenging it’s still very fun to play. I really hope to see some further development on this project!

Project Blastorium: a Bomberman-esque game

I don’t think anyone will really remember my attempts on a bomberman clone for one of the SCC. I got some result, but it was very messy and I think I’ve never got it really finished after I’ve missed the deadline. Project Blastorium on the other hand is very well polished and introduces some different weapons to the playground, which seems rather interesting. You can follow the progress on the game on the dev blog or just in the forum.

Black and White and Colors

Molyjam 2013 a game jam, where you have 48h to start from scratch to the finished product. They seem to have been three people involved in this project and their final product after only 48h is just mind blowing. It’s simple, but very well polished and the game idea is very good. It would be totally awesome, if they’d go and add some more content, i.e. new levels, since you’ll get to the last level in just a few minutes. You can get the Windows binary from the Molyjam site or read some more information on the forum.


SFML Game Jam

This discussion just popped up on the forum, thus it’s still very hot and it seems many people are tuning in on the idea – even Laurent would be excited to see something like this.
The goal would be to have a contest going for a specific amount of time (24/48/72h) to start and finish a game using SFML. How, when, where and why is what the discussion is all about, so don’t forget to give your own voice!

Clipboard and Open with Default Application feature for SFML

Originated form the GLFW 3 thread FRex went ahead and proposed the two features. The discussion is still open and if you really want this feature, you should definitely go and write there why you’d like to see this. Personally I think this could be nice, but since it should be relatively easy to implement on my own, it’s not something that should get on a high priority list.

Changes on SFML

There have been so many commits and issue discussions, that I’ve totally lost track. You can however always go and look at the commit history yourself!

SFML News – Week 12-13 (2013)


Updates on the Nightly Builds

My SFML Nightly Builds were quite a few commits behind last week, mainly because I wanted to keep the matching SFML version for Thor and thought I could get the automated building system to work with Thor. Unfortunately I didn’t manage to do so in time and thus I only update the latest version section and left the previous section with the older version. Now that GCC 4.8 got released this week, I’m thinking about providing binaries for yet another compiler, but I’m not sure which one I should use. What do you guys think?

Kroniax 5.2, Preview & Source Code

AlexAUT is not resting and keeps polishing his game Kronaix. We’re not at version 5.2, which brought a working Challenge Mode with online highscore and fixed different bugs. He also moved the updates to Indie DB and thus might get some more people to play Korniax.

I’ve posted the last weeks LP not only here, but also on the forum thread and AlexAUT asked me, if he could use my LP as a Preview. I gave him of course the permission, but also offered to create a ‘stand-alone’ preview, which he was even more excited about.

Besides the new official release and the preview, AlexAUT finally published the source to GitHub and with 5 pull requests, he already got some contributions from Haze, iostream and me. Haze made sure, that the source code will compile on Linux, iostream made a conforming Markdown ReadMe and I’ve mainly cleaned up the file naming and structure and added a CMake build system. Feel free to check it out and contribute to it.

Kroniax on GitHub

Open Hexagon’s Linux Port Released

Vee has been working hard on Open Hexagon last two week, but since he’s not that experienced with Linux, he has been deploying Open Hexagon for Windows only. Luckily he got flibitijibibo to continue his older ports. Thus releasing Open Hexagon 1.84 for Linux.

But with help from my side and even more from Aster’s side, Open Hexagon now uses CMake to build and thus is way easier to deploy on Windows and Linux and we might even see a Mac version, if anyone is willing to test it and compile it for Mac. With those new changes the time was right for version 1.9, which adds a new level pack and a few other changes and fixes, see the full ReadMe for more information.

Thor’s Particle System in Action

Tank has posted an example video on a small game he’s been working on, to get back at C++ again, after a longer time doing web development-only for the company he’s working for. It looks quite nice and demonstrates how easy it is to get a good effect with particles. If you’re looking for easy way to start with particle effects, you might want to checkout Thor, which of course offers many other handy things.

Mini Crown

kaB00M who has been or still is working on a Mana fan game called Seiken Densetsu, posted his newest creation Mini Crown. It seems to be based on Vanillaware’s Dragon’s Crown upcoming RPG/Beat’em up game and is available for Windows. The game is still work in progress, but it’s still nice to see that people are working on projects with SFML.

Platform – The simple platformer game

The author of yet another platformer called Platform is santiaboy. He has been on the forum only since the beginning of this year, but has now already released his first SFML game. The controls are quite fluid and from the technical part it seems well done, but one of course still notices that it’s a very rough version and needs some more polishing.

Personally I’m quite impressed how far he got his platformer, I always failed at the basic concepts or the collision detection, got frustrated and left my approaches to a simple platformer lying around on my hard disk.



Will SFML 2.0 ever see the daylight?

Believe it or not, but the people who’ve been with SFML for quite some time, are waiting on the release of SFML 2.0 already for around three years. I’ve started with a bit of SFML 1.6, but since the ATI bug wouldn’t let me properly execute my applications and since SFML 2.0 already came with a few more bug fixes I switched over rather quickly. The SFML 2.0 back then, wasn’t anything like the SFML 2.0 we have today. We were still using CamelCase for the functions and nobody had ever heard of sf::RenderTexture, sf::Vertex or a few other classes. The API was way more similar to SFML 1.6, what made the switch easier. The big graphics API change was introduce around 1.5 years ago, which was a good day and basically turned SFML into what it is today. The opinions of the community got split when changing the function names from CamelCase to camelCase. Some hated it, some loved it and others found it just ridiculous to even do such a change. Then nearly 1 year ago, we got a release candidate, which made everyone believe that a release would follow within the next few weeks, but as we all know, this hasn’t happened.
So getting back to the title question, will we ever see a SFML 2.0 release? The answer is yes and looking at the milestones on GitHub, we can only see a one open issue for 2.0, where Laurent stated that the tutorials are complete and only parts of the website need to get updated even gave, he even gave a rough ETA, “next month” – you better keep your word Laurent!

New SFML Logo

Unfortunately nobody has commented here on the blog about the last logos, but the discussion on the forum keeps constantly going. Haikarainen posted a new logo, implementing some suggestions by others. The font style looks really well, but personally I don’t see any reason to use an animal for a logo. The most interesting part is though, that Laurent has kind of pointed out the most fitting and we just might see a basic pentagon has SFML’s new logo. Obviously not everyone agrees and thus Nexus has introduce an idea with arrows and made an example. But one should also not forget the two contributions of jabza. Let me know in the comments what you think about the logos.

Changes on SFML

Unfortunately there have been no changes to SFML this week.

SFML News – Week 11 (2013)

Although I probably shouldn’t start again with something new and rather try to get more of those computer science articles ready, I still had this funny idea of releasing ‘news’ articles on what has been happening in the SFML community. I’ll try to release weekly articles, but I don’t promise anything and I don’t even know how much I’ll write for each article.



As promised by AlexAUT, Kroniax 0.4 got published last Sunday and it adds a complete redesign of the GUI and a few additional levels. Although he mentioned that it would take longer before the next version, he managed to get 0.51 (download page) out yesterday.

But who’s AlexAUT and what’s Kroniax? AlexAUT joined the SFML community in October last year and has been most likely busy ever since. Less than a month ago he released his probably first bigger game with SFML. It’s some kind of a side-scroller, where you have to maneuver a white triangle, representing a ship, through a maze of blocks. To make things more interesting the velocity of the triangle as well as the strength of gravity varies from level to level or can change within each level. Next to the described Arcade mode, the latest version 0.51 features now another mode called Speedchallenge, where you can change the velocity on your own, while you still have to beat the level. Since it is called ‘challenge’, AlexAUT added an online highscore, so you’ll get some aim to fight for.

Kroniax is extremely addictive and makes a lot of fun. Since I had so much fun with it, I’ve even made quick Let’s Play:

Open Hexagon

Legendary Vee has done it again and released not only one new version of Open Hexagon but five; starting with 1.8 and ending with 1.84 (download page). The most noticeable change are the online highscores. Everyone who’s playing with the official version in ‘official mode’, will automatically submit their scores to a server and get their current position and the top 8 places. But that’s not the only new thing; you now get a full menu for various options and a nice pseudo 3D effect. As always you can find the full changelog in Vee’s detailed ReadMe. The minor updates were mostly bug fixes, as well as security and performances updates.

I actually wanted to make another video on Open Hexagon, but my microphone on the headset died and I don’t have a replacement yet. The built-in mic of my notebook is not really useable for such things, but you can still checkout one of my older videos on Open Hexagon:

NEAT Visualizer

The NEAT Visualizer is a project, which visualizes an implementation of the Neuro-Evolution of Augmenting Topologies (short NEAT) algorithm. The NEAT algorithm evolves neuronal network structures and weights them simultaneously. Such a network is capable of learning some simple tasks, which the visualizer should make prettier for the eye. The used algorithm doesn’t have anything to do with SFML, since it’s a generic, but the visualization uses SFML underneath. Based on NEAT lolz123 has also written a small but fun game as Ludum Dare.

lolz123 released a new video and a new version of the NEAT visualizer this week. You can look at the video here:


New SFML Logo

Nearly two years ago Laurent opened the discussion on the forum for a new logo. Back then we already assumed SFML 2.0 would get released anytime now, but well we are still waiting, more on that in another post though. After the thread died out for a while it’s back again and every week we get a few new suggestions. The ones of this week were in my opinion quite strange and not very well suited as logo. On two of them it’s nearly impossible to see the letters ‘SFML’, especially if you don’t know already that they should be there. The other one is textured which is not useable at all for a logo. But maybe I’m not artsy enough to understand this, so I’ll let you make your own opinions:

Will SFML support OpenGL ES sooner than expected?

OpenGL ES is as specification for a platform and language independed 3D programming. It’s mainly aims embedded systems (thus the ES) such as Raspberry Pi or devices that run Android or iOS.
If SFML were to be ported over to OpenGL ES, it could potentially open the world to a whole new set of platforms and thus new games and applications with SMFL. In general Laurent wants to get SFML to OpenGL ES, but in the past this was not due for anytime soon. What could then change his mind? Well slotdev started a discussion on the forum about having a DirectX backend for SFML, which would involve even a bigger change to SFML, but luckily some guys have created ANGLE, which basically translate OpenGL to DirectX calls and thus enables one to write OpenGL applications for platforms which support only DirectX. The downside to this is though, that ANGLE implements an OpenGL ES specification. Since slotdev and a few more people are commercially working on games for casino machines, Laurent stated that he’d be looking into getting OpenGL ES support sooner than plant, if they require it anytime soon. The discussion was then moved to private chats and we’re kind of left in the dark what they decided on.

What do you guys think about elevating the priorities on OpenGL ES? Should Laurent tackle other more important issues first before rewriting huge junks of SFML’s code? Let me know what you think in the comments below.

Source code changes

Personally I feel like we got a much more involved community in the past few weeks, but I’m not sure whether this is just because I’m looking more closely at the commits and what’s going on, or if it’s really because more people are submitting pull requests and Laurent actually accepts them. It also kind of feels like, Laurent got a bit more open about contributions after the forum thread with the title: “More Commuity-driven Development”

  • We received a fix from one guy, who had problems with some events on Arch Linux with Awesome WM. – 560b741
  • As noticed by our great Nexus in a forum post, the explicit for the sf::Text constructor wasn’t needed anymore. – 5c46daa


I hope you’ve enjoyed reading my news post and actually got some news out of this. If you feel, that I’ve missed some parts or just got any kind of feedback, please leave a comment down below or contact me by means of PM, Mail, Twitter, IRC, etc. I’d really to hear some opinions.