Xeno bits

[French] Belote World Tour – 1 an après

Un point sur Belote World Tour 1 an après sa sortie, profitons en pour balancer une avalanche de chiffres et de graphiques.

Dans l'ensemble c'est plutôt positif et la seule chose qui me chagrine c'est que les ventes de Windows Phone n'aient pas encore vraiment décolées.

Commençons par les téléchargements:

Selon les dernières informations disponibles la version gratuite a été téléchargée 19572 fois sur la marketplace (en vert les mises en avant sur la marketplace)

La payante a été essayée  1090 fois

Et achetée 571 fois

Et au niveau du classement cela représente quoi ?

Belote World Tour n'est jamais sorti du top 100 des jeux gratuit depuis qu'il y est rentré il y a un peu moins d'un an

Classement jeux gratuits

La version payante ne peut pas en dire autant mais cela reste honorable surtout face aux jeux xbox live

Classements version payante


Les téléchargements c'est bien mais est ce que les utilisateurs sont contents ?

D'après les notes obtenues oui, on remarquera que ceux qui la payent apprécient plus l'applications ;)

Notes de la version gratuite

Notes de la version payante

Les joies d'une application sociale

Le jeux implémente un système de highscores, et sur 4 classements différent 3132 scores individuels sont présents.

Et la publicité ?

Principale source de revenues sur cette applications, après un an le nombre de publicités affichées d'après pubcenter serait de 8 267 248.

 De l'utilisation de Flurry

J'ai implémenté Flurry dans le jeu en fin d'année dernière, maintenant je m'apercois que les résultats sont à prendre avec des pincettes tant ils n'ont rien a voir avec ce qui a été rapporté par Microsoft.

Nombre d'utilisateurs selon Flurry

60000 utilisateurs depuis le mois d'octobre alors que microsoft m'annonce moins de 20000 téléchargements ?

Une autre donnée intéressante, le nombre de sessions de jeu par mois, celui reste croissant ce qui est encourageant. (ne pas prendre en compte le mois en cours)

Nombre de session de jeu par mois

De même si on regarde ce nombre par jour, on arrive à observer les habitudes de jeux des gens, notamment que le samedi les joueurs de belote on visiblement autre chose à faire ;)

Nombre de sessions par jour au cours du dernier mois

Enfin quelques données plus techniques, comme le nombre de sessions par type de device au cours du dernier mois, sans surprise le Lumia 800 domine

L'opérateur préféré des utilisateurs de windows phone, Orange avec pres de 50%

Ou encore le % de gens qui ne sont toujours pas passés à Mango (18.3% quand meme !)


Pirate Cat – Post mortem 1/3

Back to 2010 I was working on a WP7 game as I thought it was definitely the platform where to go to have a chance to be noticed as a one man structure. Moreover I was already familiar with XNA thanks to previous professional experience, and I knew that 2d development would be pretty fast and simple using such tool.

Now few month later, in january 2011, I realised that I would not be able to complete this game anytime soon as I had a full time job during the day and only the possibility to work on it on the evening and during week ends. So I decided to give a try to a simple game concept that I wanted to prototype. In my mind it was pretty easy, take 2 of the most popular mobile games, mix their mechanics and see what happens.

That’s how I started the development of project Icarus (now known as Pirate Cat), the game was supposed to be a mashup between angry birds (you shoot things using a slingshot...) and doodle jump (you try to go as high as possible...).

The game is now released since one month and here are the things I learned during this journey.

During the creation

Screenshots of the game at various stages

What went wrong

  1. Project duration

        It started as a 2 weeks project and ended 1 year later (with some breaks in the middle but still).

  1. Time taken to validate the gameplay:

I never thought it would be so hard to mix popular gameplay elements and get something fun to play. The first idea was to shoot birds to make feathers spawn, then use the accelerometer to grab the feathers while falling, but I realised that if you fall too much, you’ll never get really high, and you’ll just spent your time falling, then try to keep flying, then falling again without even being able to reach the point where you first felt. So the first major change was to boost the player up to his highest position after being able to refly.

Also first the player had to fill ¼ of his energy bar to refly, but here again it wasn’t fun to fall then get a bit of energy, refly, and refall immediately after, so I fixed it by requiring to fill the entire bar.

Last major change to initial concept, I introduced a limited time, because I noticed that the game was becoming boring after few minutes without a time limitation.

  1. Time taken to polish the game:

I knew that I wanted to make a game as polished as possible, else it was pointless to release it, but I didn’t thought it would take so long, I would say the polishing itself took me 60% of the total development time.

  1. Budget exploded:

At first I defined a small budget for as much arts as possible, and I will do myself what was missing. In the end it costed me 2 times more that what was planned but I ended with professional quality assets so it was totally worth it.

  1. First users feedbacks:

So you have your great idea, you make a prototype that you find fun to to play, then you put it into stranger’s hands and discover that your game totally sucks and that nobody will understand how to play it. That was a bit of a shocker but it helped me to realize all the weakness I needed to address before releasing the game.

What went right

  1. Standing out from the crowd:

At first I wanted a game colorful with vector graphics like most of the other games on app stores, but the key point is to stand out and be noticed as something different, that’s why I decided to create an early cartoons themed game, I’ve not much noticed such design except in console disney games so it was worth trying, the bad point is it may look boring at first as it is black and white.

  1. Spending time on the icon:

I would say that the icon is the most important piece of art of your game, do a simple test, launch an app store, one that you don’t know if possible, look at the top 100 of any category, and try to find which one you would like to try, that’s basically how most of the users are proceeding when looking for an app, if your icon is not good enough they will not even read the app name nor check its details. Few tips to have an attractive icon:

    1. Keep it simple: do not use a screenshot of your game as icon, often the face of the main character or enemy is enough.
    2. Make it appealing: if your main character is a bit lambda use a kick ass enemy instead, if your character is a girl, use her sex appeal.
    3. Try try try: do as much drafts as possible, then compare them yourself and remove the ones that you don’t find worthy, next ask to someone that know nothing about your game to chose their favorite at first sight, and keep iterating this way until you have something really sexy.

  1. Organisation:

I do not use some advanced work methodology, all I do is keeping a google doc spreadsheet up to date with 4 columns: features, bugs, done and canned. But it works pretty well this way. Also do not be afraid to focus on the core game mechanics and slice any feature that may looks good but will take too much time to complete, to do so setup early in project key milestones like alpha / beta / gold and always stick to it even if it means sacrificing cool features, it will always be possible to add them later.

  1. Tests:

I can’t say enough how important it is, test your game, and more importantly put it in new people hands as much as possible. I found new issues with each people I gave the game to try, and even if not all were relevant it really helped me A LOT to improve the game and make it enjoyable.

  1. Using dev art:

Often when starting a project, I remember how bad I am regarding art, and this tends to discourage me to continue any further as the game will always look like crap. But my new moto is, don’t be ashamed of dev art, spend some time on it to get something close to what you are looking for, but not too much time. And if you succeed to have a fun game even with ugly dev art, now it’s time to look for some skilled artist that will transform this into a killer app.

  1. Finding artists:

I wanted someone with some experience to do this, it’s definitely easier to sell a beautiful game with poor gameplay than an ugly game with great gameplay, we’re talking about mobile gaming here and most people judge a game worthy after checking only an icon and the few available screenshots.

I would have been able to find some friends of friends that were in art school or this kind of things but I chose to go on various 2D arts forums and post my request.

In order to get quick answers and be able to compare the prices easily I defined a basic xls file including 5 or 6 kind of assets that I would need for the game.

After receiving many offers and comparing them I was not able to find someone that would fit in my budget and have the required amount of skill. So I decided in a second attempt to browse the various portfolio available and send them my request, I finally succeeded to find someone this way.

  1. Final result:

Not much to say here, of course there will always be things to improve, fix, or add but if you are proud of the results it’s a really good point!

Coming soon : part 2/3 after the game creation



Only one day

So yes it took one day for Pirate Cat xap to appear on "suspicious" chinese forums, even considering that the game is actually free.

At least I included some anti piracy trick in it (thanks to @hermitdave for this article : http://invokeit.wordpress.com/2012/02/06/wp7-app-piracy-and-how-to-deal-with-it/ ) so hopefully if someone download and install the game from other source than official marketplace he should not play more than the trial content.

Still I was the first to throw rockat Android and their massive piracy issues, but as I can see WP7 is not exactly a crime free land, at least pirates do not repost your games...  (for now ? :x )



Pirate Cat released

At last the game is released and you can now grab it for free before the start of June!

Check out the trailer


Project Icarus has an official name

Things are going pretty well on project Icarus, it should be feature complete this week end and I'll start working on polishing and looking for beta testers.

I finally take a decision regarding the name, it will be :

Pirate cat : The flying pirate

Like in old cartoons, the hero has its own name "Pirate cat" and this specific "cartoon" episode (even if I don't know if there will ever be other episode) is named "The flying pirate"

Now regarding the progress, I've got most of the assets for the tutorial, I'll now work on improving the texts, I want to keep it short but ensure everything is clear enough as it will introduce the game context.

Shooting tutorial

Few more screenshots from the game:

I've made some improvement on the aiming mechanics, I realized player was not aware of the slingshot strenght limit, so I decided to draw the aiming mark in yellow once the limit reached and make it rumble. Hope it'll work better.

Aiming improvement

Baloon full of feathers sent from the boat

The steam engine

Rain and full moon

Speed boost



Windows phone in France – some fresh stats

Thanks again to Flurry analytics  and Distimo I was able to extract some precious data from Belote World Tour usage and extrapolate a bit on the current state of Windows Phone in France.

The game itself is currently ranked 146 in french marketplace Top Overall and 61 in Top Games, I've made the chart below, representing the downloads and ranking between 1st and 25th february, in order to find how downloads impact ranking  but it does not appear to be that simple!



The daily downloads are not really decreasing, but my ranking is, I guess it's some kind of Microsoft magic in order to avoid apps trusting the top ranks, or maybe user base is growing fast and so is the downloads / rank ratio but I really doubt so based on the previous months evolution.

Now regarding the devices, it's really tight between Nokia Lumia, Lg E900 and Samsung Omnia 7, but remember that Lumia was launched in november, it's clearly the fastest growing.

One last about carriers, Flurry does not seem perfectly able to detect them right now but I guess that the second one is SFR, no Free in sight yet, but maybe it's included in Orange.

Now once again all those stats are provided by third parties, and I have to admit that I have noticable differences between distimo/app hub and flurry downloads stats for example, but it still can help to get an estimate on how are going Windows Phone in France.

Edit: One last important figure that I forgot to mention in the original post is the OS version repartition, it's interesting to see that today more than 1/4 of the devices are still not running Mango.


Next project – things are beginning to take shape

I've started to work on this game since 1 year now and all I can say is that things were not going really fast so far.

But I managed to find a talented artist and hope to finally finish and release it within 2 month.

The game is about a pirate, cat pirate more precisely, that have to reach the moon during the night.

To do so one of his fellows constructed a pair of wings covered with feathers, the downside is that feathers are not very well fixed and player will have to constantly shot birds he encounter in order to keep flying.

So yeah at first sight it sound a bit like a shoot'em up but in my mind it was never really related to this genre.

This week I started to implement new assets for the hero, those are not finals and will receive minor changes soon.




For those curious to see the thing in motion, check this video, a bit older:



Ad providers comparison

When I started to introduce ads in my game I was a bit puzzled by which provider to use, windows phone is a pretty young platform and options were at that time (end august 2011) pretty limited for none US developer like me.

  • There was Microsoft own solution (Pubcenter) which was available only to US developers.
  • Google had released a "beta" sdk for AdMob
  • And there was various other "minor" providers that looks good on paper but I was not able to find any user story about them.
So I decided to experiment, which was a big hazardous as I already had almost 4000 downloads at that time.

Downloads states between 08/25/2011 and 09/25/2011

I've put 3 different ads in game,
  • 1 during game (MS Pubcenter as I've seen on their forum that I should be able to receive my payment once the platform would be open to international developers, more on this later on)
  • 2  on an unskippable 5 seconds screen after each game round, Smaato (http://www.smaato.com/) and Inneractive (http://inner-active.com/)
What, no Google? yep, after trying to use their sdk I discoverd that it simply ruined performances of my application, so it was a no-go. Last time I checked their SDK was still in beta and I was not even able to run their last provided samples on Mango. I guess that WP7 is not a priority for them.
Once done needless to say that users were pretty upset after downloading an update including so much ads, and I've got many bad reviews during this period, sacrifice was worthy, in the name of science!
After one month of experiment here are the figures :

Inneractive stats

Inneractive figures

Smaato stats

Pubcenter impressions

Pubcenter figures

Let's try to see what's going on here :

  • Smaato ecpm are crappy : 0.075 ??? seriously
  • Inneractive ecpm are a bit better, but their sdk report 1/5 of displayed ads compared to other providers, I contacted them regarding the issue and they advised to use exclusively their SDK to solve the issue, that looked a little bit too magical  to me (and risky considering time needed to revert this choice, although I can use ad rotator  (http://wp7adrotator.codeplex.com/) to ease the process) .
  • Pubcenter ecpm are not stellar but at least they are the best, even if I had no idea when I would be able to get my money.

In the end I kept only the pubcenter banner, removed the unskippable ad screen, and user started to be happy again.

Few month after this experience here is my opinion about those providers:

  • Smaato will not let you earn much, but at least today they are the only ones from who I received money. Their sdk is pretty good and site is the best I tried so far, especially regarding analytics.
  • Inneractive tools are easy to use, I was easily able to contact their support and they spent some time to help me solve my issues by chat, but I was disappointed by their answer to my displayed ads amounts issue.
  • Pubcenter sdk is the best for sure, their site not so much (I need to use IE in 32 bits version in order to use it...). I finally succeded to setup my account in order to be paid at the end of november, as they were opening pubcenter to new countries including my. But I still have not received any money from them so far.... I heard many times that I had to be patient with microsoft (same thing for paid version of my app...)

Hope this will help few people out there.


Belote world tour on the marketplace

The Pro version can be found here:


And here is the free one:



Belote World Tour

Belote World Tour is our first game released on WP7.

It's a card game based on very popular Belote, popular only in France unfortunately.

First free version was released on 13th of july, it helped us to improve the game based on users feedbacks and include more features.

After some iterations we judged the game was ready to be released as a paid app at the end of August, in the meantime we started to include some ads in the free version , but more on this soon in a dedicated post.

I do not have to complain regarding the attention the app received, I launched it with absolutely no PR and it succeeded to reach pretty easily the top 50 of free apps (thanks to the popularity of the card game itself and the fact that it was the only app representing it), it stayed there during few month and since october it's slowly going down in the rankings.

After 5 month here are the download stats:

Free version

Not much to comment here, except that considering its ranking in the marketplace we can only conclude that WP7 devices are not that popular in France so far.

Paid version

This one is a little bit more interesting, first we can see that it's nothing compared to the free version, no need to explain that, but in the meantime at the beginning of november we added a new game mode, exclusive to the paid version, and sales started to become relevant, thanks, in my opinion, to the existing users of the free version, seeing new locked feature and ready to pay to unlock it. Moreover the price doubled from 1.49 to 2.99 during this update(for science purpose only ;p) but it did not seemed to be a blocker issue for a lot of players.

Now for the numbers lovers here a few more stats regarding the application usage thanks to Flurry analytics, a very convenient tool to understand better how people are using your application.

Median session length : 6.2 minutes

Daily active users :

Session frequency by day / week / month