Mika Mobile Our Future with Android

"From a purely economic perspective, I can no longer legitimize spending time on Android apps"
http://mikamobile.blogspot.com/2012/03/our-future-with-android.html

Some discouraging words from Mika Mobile on Android development.

Some of my interest in supporting android have actually been based on Mika Mobile's "success" with Battleheart in Android, but seems it was a short term thing. They claim Android only amounted to 5% of their entire sales for last year.

Although to be fair, they had Zombieville USA 2 hit the #1 spot in the iOS App Store, that’s a level of sales few will ever see. It is like playing the lottery and winning the jackpot, then saying working at a gas station is not way to earn money.

Note I'm not accusing them of anything, just trying to see some light of hope in the Android, even if only for developers that have not hit the jackpot in iOS.

What worries me the most is the notes about modifying shaders and other bugs in multiple devices, despite their games being made in Unity. Makes me fear diving into the general Android market. I’m still going ahead with it but… fear is taking over.

 

This? From you? I'm surprised.

These stats from AndroidLib should brighten your day.

http://www.androlib.com/appstats.aspx

Androdians are cheapskates. lol

Only about 250 paid apps have more than 50,000 downloads, and only 906 paid have between 10-50k. Still good money, but your game would have to be really good to make that kind of money.

Some think that a free app + micro-transactions is the way to go, but once again, you'd need a really awesome game for people to spend money. Look at the numbers above 5K for free games.

And ads? Ah, I think we all know the answer to this one.

We all kind of know that trying to make a big money maker on Android is like trying to win the lottery. And why not try? It sure beats just sitting around the house watching TV or playing video games.

Just keep doing what you're doing and stop worrying about it.

 

Contrary to what many may think, I do wish very very hard Android to be a viable platform. Also, I do got to hope I can make the near $400 I spent in hardware! (albeit I can always sell it later on ebay at a small loss.)

Can’t just go too blind, though. As Mika Mobile notes, support can become a big issue. If the investment is not worth it I may actually just avoid general Android release and go Nook/Kindle exclusively (even if the profits are small.) I may go free with ads for general android... some money is better than no money.

 

Awesome news! Testing of the waters is appreciated, so the rest of us can be confident that staying out of them is the right choice. It's a shame that the decision seems to be based around money, and not ideology, though.

 

You just love seeing people dive into shark infested waters, don't you?

 

Its bussiness , with Android cost skyrockect when it comes to testing and QA( plus Bob might have some ultra low budget phone and he can't play so he's emailing support all day looking for a refund, ect) , all for a market that doesn't want to spend .
My plan is to just make this game for Andriod until I can get a mac( since unity makes it sooo easy to go multiplatform, why not do both).

I'm looking into adds, aside from the Kindle I don't think you can sell games on Andriod( this is only because Amazon does a good job of putting a full content consumption center in front of people with money to spend) .

 

I can think of one very small change Google can do to their app store that would make the place profitable in a heartbeat:

Hide free apps.

If the app is free, only allow people to download it if they have a web link that takes them to that section of the store.

Maybe ad a setting, hidden in a config box very well tucked away that allows users to see free titles in the search results.

Right now, I can see how it can be a big issue with so many free games everywhere (tossed there for free by kids that got dad to pay 25 bucks for the account.) Just imagine, if you entered a strip mall. And there was a good restaurant with decent priced food, but 40 fast foods all giving food away for free.... would you bother going to the one good quality restaurant? or would you just keep tasting the free food until you are too full to pay for food, even if you ended with a bad mouth taste?

To be honest, this would even be on Google's interest. If people download an app for free, Google gets nothing! Why give the apps any highlight? Just highlight paid apps!

 

Hmm, then how do I give out the demo version of my game.

They could do what apple does and reqire a credit card just to enter the app store( I very much hate this, since not everyone feels safe with their CC info on their phone).

The kindle is your salvation !

 

Extend the 15 minute refund window to 25 minutes. That should be enough time to decide if you like the game. You download, try it for 25 minutes, if you don't want it you return it.

That said I got to put this out there: if you are interested in a 99c game, and you NEED a demo.... what is wrong with you!? If this was an Arcade, you would change 20 bucks into quarters and toss those coins left and right! 99c for a title you are interested in should not be a big consideration. The only reason it is, is because people are used to see free stuff all over the place.

 

Well for me I'm putting my game out to see if people like it , plus considering this is my (solo) first game I don't expect people to pay for it.
Plus your telling everyone to abandon ad-based gaming...

 

I gave ad based gaming a shot. It's a horrible horrible business model.

Free to play games with micro-transactions may be a real concern though, although I feel they also cheapen the user perception of the market.

 

On the brightside, gamestop is investing a bit in Andriod gaming, so they might open up the market a bit .

 

Hmmm last I heard GameStop was planning was to make their own tablet with their own app store. Not sure if that will ever pick up, though.

 

Android is a huge and growing market. It is, however, very fragmented. So if you don't have a large team to optimize for all the various phone specs, it might not be as appealing as iOS where you can focus on just a few devices.

 

AndroLib is slightly misleading with their "free vs paid" apps -- the stats really should be divided by the number of apps in each category considered as well, because I think the issue is that there's just a lot less paid apps at those numbers.

What is a problem though is fragmented screen sizes and varied computers. If, as a developer, only had say three screen sizes to develop to, and on top of that Android would provide a DETAILED list of what OS's are used, and what download games, and what their specs are, I could make an informed decision, buy my "low end" test phone,and only publish to anything >= to that. It would cut things down tremendously.

They have shot themselves in the foot with their "free to do whatever you want!" mentality, and as a result the phone market is destroying their apps.

 

Great insight blog, thanks.
And now I'm cold as ice about releasing it on Android too... 20% time on Support, holy S***.
If it is such an amount of time for a (great) sprite based game, I don't even dare to imagine how much more it is for a 3D game ...

also, this graph is truly depressing :

 

Apple does not require CC to enter or download free apps. App Store also accepts other methods like gift cards you can buy from local stores while market supports only few countries and a lot less payment methods.

 

You only have to support something that people love and want to make better. If they don't love it or it's buggy in a bad way, they'll just uninstall it. There are soooo many free games on Android, that people can plow through it like junk mail. Install, play a few minutes and uninstall.

Fragmentation and different devices can "potentially" be a problem. It's not true in all cases. It really depends. You definitely have to do your homework and testing. Try taking a look in the comments section of popular games and apps and see what people are writing. That will give you an idea of what you're stepping into and what you have to consider before putting your game out there.

 

The Android market is growing really fast and fragmentation isn't as much of an issue as everyone is saying... it's all about your development pipeline and how you build your games. If you plan properly at the beginning of your development, you can make a solid Android port of your game within about 2-3 weeks which is pretty amazing.

Sad to hear that Mika's are not making any money on the Android platform... Our last title Mike V: Skateboard Party HD is selling twice as much on the Android platform than for iOS. We didn't have any promo or advertising help from Google too. On top of that the Amazon Android sales are almost as much as the ones on the Marketplace too. Our Xperia PLAY compatible version will be released soon.

It would be non-sense for us not to release an Android version of our games. Even Lite versions, no matter which platforms you release on, you are increasing your user base which leads to cross promotion and letting more people know about your titles. We will be reaching over 10 million users across all platforms next week. Android and other "upcoming platforms" are definitely part of our business strategy.

 

I posted on my blog about this too. While it is good of Mika Mobile to release info, I can't help but think perhaps he should have tried freemium (in app) or adverts. Do you have any response to that or thoughts?

 

You know, I was thinking more about this and I think there is an issue of expectations...

I remembered one important point about Mika's post: he says that 5% of last year income came from Android. The fun bit? The Android market tells you how many copies of apps are installed, within a bracket. Both of his android games are within 50k installs and 100k installs.

One of his apps is 2.99 and the other is 99c. If we assume he never made the game available for free (this would mess the entire estimate entirely) then this means he earned between 140k and 278k from the Android market. Not an insignificant amount. But as I noted, he had his latest title, Zombieville USA 2, hit the #1 spot and that likely skyrocketed his income insanely on the iOS Front. Actually, if the income I calculated here is right, then that would mean that his iOS income was between 2.64 and 5.3 million dollars.

Those are insane numbers, and so large that they would make the 100k income not worth the extra support hazzle.

Mind you, this is a huge guess and is entirely invalidated if he even had a free day, since those free windows would insanely boost installs.

 

Keep in mind with unity porting can take DAYS( this assumes you can forgo sleep ) between Android and IOS , we don't know how much porting cost Mika Mobile

 

Mika Mobile uses Unity.

His issue, though, does not seem to be on porting troubles but mostly about installation issues (something that I know is a real problem first hand with an old Android phone I have laying around) and new devices showing up that suddenly don't play nice with certain shaders (not sure if he uses stock shaders)

 

To be fair, there were a few zombieville clones on the Android market because it took him so long to port over. One of the clones is free and has over 5,000,000 downloads, so why would anyone want to play the original?

The original is great and fun, but the free clone is good enough for most Android users.

 

Well, since I'm starting with Andriod, will porting be easier, and much thanks for all the GUI tips, I'm going to put out my game on April 1st!
One free level, lets see what happens !

 

Well I can see why he wouldnt bother with it, he is making millions of units per sale per app. Now if you arent really I really cant see how a day to port it over cant be worth it (but then again you can imagine there time is worth more then mine).

I think if your like most of us and dont sell say over 50k units then it cant really hurt to port to droid.

 

Well I'm still going to go ahead with my port. I already spent 400 bucks in hardware so may as well. But I don't think it will take me just a couple of days. I have some heavy changes to do for interface handling, plus had to gut the game left and right to get it under 50MB.

I also plan to take the chance to do some game restructuring, but that's theme for my other thread on the project.

At the worst, I'll be able to make some blog post and some kind of press release with market comparisons out there, and that on it's own may bring some attention to the game in all the platforms.

 

He says in his article he didn't sell that many less copies, he just didn't make any money after considering the time he spent trying to support all the various android devices....and in hind-sight a year later, 20% of his labors went to supporting android devices, which made up a total of 5% of his profit.

As he says:

I just tried to play my first android build on my phone and it can't install it (20 meg build, basic level + tilt controller)..........The phone's a couple years old but I figure it's a good baseline to target as S*** would run faster on something new.

 

I dont see anthing there noting that they didnt sell many less copies. The only reference at all at proportions is the 5%, but honestly, the 5% is not that far from what I heard some quote as their iOS/Android proportions.

It does have to be noted too, though, that most his income last year must have come from Zombieville USA 2, a game he never released to Android.

 

Maybe if we call his name 3 times in front of a mirror he will come to share his thoughts ? :O

 

I have iPhone and my wife has an Android, and the main problem with the droids for games is the Android Market.

Compared to the iPhone and iTunes/App store, it sucks. With the iPhone its a simple process, boom, a few seconds and you're in a able to buy and download any app you want.

With the Android .... man it was a headache. Go here, go there, create a google account, it has a tendency to mess up with new google accounts, things became more difficult when it randomly decided not to associate the account to the phone, took forever to get it resolved. When it was finally able to work, it wasnt user friendly as compared to the iPhone/iTunes.

I'm sure many would have abandoned the process once it got difficult, but I was persistent to see how it compared.

So they need to re-vamp the whole thing so it works easy peasy right out of the box, and not expect users to jump through hoops.

Only then will Android be better for the developers and consumers.

 

They are now putting droid games at 50 cents rather than 99 cents. At least with apple you couldnt go lower than tier1 99 cents
https://play.google.com/store/apps/category/ARCADE?feature=category-nav

 

This is like a week long promo to hype Google Play( the intergation of the music store and the market)

 

I have not developed for a mobile platform, but I want to add my opinion on the Droid market and mobile games. Please don't let that previous statement turn you off to my ideas. I have been studying game development for the past 8.5 years I have done much research on marketing, user trends, piracy and traditional vs pay to play vs free to play business models. I do realize that theory only goes so far and I am not claiming my words to be the be all end all. I just want to partake in this discussion to add my opinion and hopefully learn more from those who have been in the trenches.

Hopefully I still have your attention and respect, so here it goes:

I find it interesting that Droid is such a difficult market to penetrate as many have described. I understand finding baseline hardware and working up from there to ensure it works on all devices with all resolutions, but really, Droid seems to be making mobile development just as involved as PC game development. This must be hard for iOS developers to adjust too since iOS is a walled garden it makes it easy to make a game work on one device and every other new device. Droid requires a complete different way to think about making games. You can't assume screen size, you can't assume a processor. You just have to work from whatever weakest device you want to support.

Now here are some interesting statistics about smartphone use in the US:
http://www.comscore.com/Press_Event...uary_2012_U.S._Mobile_Subscriber_Market_Share

Droid dominates the market with 20% total share more thaniOS. So really there should be many more users that are willing to pay right? (yes I know that is not necessarily true, but lets just assume that with such a larger user base there are up to an equal percentage of paying customers in each market.) So why is the market so unprofitable?(besides having to support more hardware/software types) Perhaps its because of the marketing practice for games in this type of market. With that said I would like to address opinions similar to the one below:

I think developers need to change their marketing tactic to counter this, not throw their arms up and say "well its not worth it", or "if someone wants my game they will pay for it". Sure $0.99 is cheap, but I also hear devs cry out that its hard to support all the hardware droid runs on. Wouldn't you think the users are aware of this as well? That they might have paid for apps before only to have them not work properly on their system? Or worse had them work but then crash or freeze their system when they reach the most complex part of the game? A demo would allow weary customers try something they like without a worry that they might pay for an app that doesn't work. Sure, they could file for a refund etc, but if the user thinks its just $0.99 and not worth the hassle, how many times do you think they will risk losing a dollar before they decide to stop taking risks? And how soon would they stop taking risks in a market that has so many free alternatives?

Also, I would like to note that Arcades are dead in the US. I think people got tired of spending so much money when home electronics offered the same experience for a flat fee. Its about convenience. The larger barrier to entry for your customers will result in fewer customers who will bother to enter.

If you make a game for Droid and you think free games are cutting your potential market share, why not give the players a taste of your game so they want to buy it? Just give them a part of it. It could be the first three levels or a small test level. It doesn't have to be the full game. Strip out some parts of it and ship it as free. Get your product out there so people know what it is and that its worth their money. Even if you don't want to take the time to make a demo and submit it etc, make videos of gameplay and youtube "commercials" to advertise the game to the user so they see what they will get. I don't think you can simply put your game for sale in a market with so many free alternatives and not have anything to draw customers to you. To me, that spells market failure.

 

Another reason to give a free demo is to let each user determine if the game works on their phone .

I'd be pretty pissed to have to request a refund since the app wouldn't let me test it before buying !

 

I disagree with their assessment. Especially, with the release of the XAPKs (expansion APKs) and potentially Unity support or community support for them in the future. The size limitation is one of the most limiting factors when developing Android games with Unity.

 

How much do you reckon it cost to host a XAPK server for a modest amount of traffic ?

 

There are many hosting services out there that can currently host data for your Android apps. (http://aws.amazon.com/s3/#pricing) The problem hasn't been with the pricing as much as basically needing to develop your game with the external asset bundles in mind which have their own overhead involved. Our game didn't benefit from asset bundles as we have a huge number of dynamic assets that need to be loaded and there isn't any semblance of rhyme or reason on which assets can be bundled together. So that means that we needed to make an asset bundle for every single dynamic asset in the game. It was hundreds of asset bundles, and generally a pain.

However, the whole point of these XAPKs that were finally released is that Google will host them. You won't need an external server hosting the files. They'll be integrated into the Google App Marketplace (or Google Play as it is known now) and make purchasing the app better. The Marketplace's refund timer will start AFTER the extra APKs have been downloaded. I spent the majority of my time today looking at their examples, and with the Unity APK packager that Floky made and redirecting the game to use that external APK I think we should be able to trick Unity to think that it is reading assets locally like Resources.Load does. Which will help my game, and many games, a lot. Sure there will still be a lot of device compatibility issues as there are a lot of issues generally with that on the Android platform. However, with XAPKs we're closer to having the platform be more on par with others.

 

Well put. A great example of what you are talking about is "Game Dev Story". I heard about the game through a friend, tried the demo, fell in love and for the first time felt compelled to purchase a mobile app. (I've never done this before on my iPod touch). The problem with androids is that some of the older phones (mine for example -- the HTC Desire) weren't made for gaming. They have a tiny internal storage, and once that's full you are out of luck. Many apps won't allow you to simply place them on the SD storage (which I have in abundance). I really hope this has been addressed in future phones... when users start to update their phones after their contracts run their course... things in the market may change?

 

In IOS there is 3-4 devices on droid there is over 500, you cant even test a fraction of them. So either you are only going to 'reject' you know work or risking the wrath of the people who get pissed off rate it a 1 star if doesnt work on there phone and I guess ask for a refund.
http://forum.unity3d.com/threads/103608-Devices-that-get-bad-reviews-on-the-Marketplace

Look at plant vs zombies on droid. 900+ 5 star ratings and 600+ 1 star rating because it doesnt work. Most of them seem like the same devices but people dont even check the comments before buying.
https://play.google.com/store/apps/...251bGwsMSwyLDIwNiwiY29tLnBvcGNhcC5wdnpfbmEiXQ..
Samsung Galaxy, Motorola Droid X, Samsung Galaxy S

 

Seems to me that compatability is a major, major issue for Android. I have a Nexus S Android phone. This phone actually endorsed by Google and supposed to be a reference platform - so its not supposed to have any additions which could cause problems. I'm still finding that lots of games which don't need much graphical power, like Sim City, aren't available because the developers haven't tested on that device so don't want to take the risk.

It looks to me like there's an opportunity for Unity here. They should create an Android Benchmark and compatability tester which you can then download and run on your android device. It would then give you a benchmark speed reading and a shader compatability reading - lets call them a "Unity Mark". Unity would then keep a database of the "Unity Mark" of all android devices. When you're developing the game you could then set Unity to simulate a particular target "Unity mark" in the same way you can get unity to simulate different PC configurations.

When you publish the game you then generate a list of compatable devices which have a high enough unity mark, rather than having to test on each one.
With a bit of luck device manufacturers would pick up on this and start building their devices to perform well at the benchmark and publicise the benchmark for their new devices. This would be good for consumers - if you want to use your phone for games you would pick a high unity mark one. It would also be good for manufacturers - if they're trying to persuade you to replace your old phone its much easier for them if they can say "It has a unity mark three times as good as your old one".

 

Great idea, much better than the suggestion of making demo's for everything.

 

You cannot quantify a complex system by a single number no matter how many digits of precision you use. There is a reason a single graphics card, when reviewed, will generate ten pages of data with conflicting trade-offs.

It is not just a particular model of phone, but the hardware revision, OS revision, provider installed software/crapware, is it rooted, was it over WiFi or 3G or 4G, what permissions are enabled or disabled, what extensions have been added, and finally, "yeah, it's Android, nobody F***ing knows why it crashed that one time, it just F***ing did."

Programmers tend to focus on the cost of programming. "Well it only takes a day or two to port to Android when using Unity so the cost isn't very much." The cost of programming a video game, over the entire lifespan of the video game, forms only a small percentage of the entire cost of designing, marketing, supporting and upgrading that video game. If most of your budget (time or money) is focused solely on programming, your game is going to flop big-time**.

Disney closed its interactive division at one point because it was only making around $30M a year in net profit. Too much management combined with too much risk for not enough return. Think about that, $30M in profits but its not enough. That kind of number is like finding 10 cents on the sidewalk when you make $200K a year. Yeah, it gives you warm fuzzies but it doesn't even move the needle.

**Please don't quote Minecraft because 1) It's a single data point and you can draw any line through that you like and 2) A lot of budget (time) was actually spent on promotion, marketing and getting people to pay attention to it.

 

If you make $200K a year...you wouldent want $30million?!?!?!?!?!

 

/me pinches your cheek... "You are so cute when you fail at reading comprehension. Yes you are! Who is? You are!""

 

Yep, that is the reason i get "D" in every english thing i do and im rather sure that is just because my teacher dosent want me to fail 21/40 sound suspicious?, i get it now though

 

Some people are just wired differently from what is "acceptable" or testable. I am pretty terrible at English too.

 

Interestingly enough don’t you live off of the proceeds from writing? (at least you blog said that) wouldn’t that be a difficult career for someone that is "pretty terrible at English"?

 

Sure. I quite agree. But that doesn't mean that benchmarks and automated compatability tests aren't going to be better than nothing. I'd be interested to see how well the Unity PC "benchmark" system works in practice. If I set my PC Unity project for the lowest spec, what are the chances that that spec will actually cause a shader problem or crash?
Also at the moment I might develop a whole Android project using a shader which only works on my development phone and then only find out during QA. If this was build into the Unity environment I would know from the start and find an alternative.
At the moment, Android manufacturers don't have a clear incentive to make their devices compliant with Unity's requirements. Having a clear benchmark or set of tests would give them something to aim at. From a buyers perspective most Android devices are pretty similar so passing a benchmark or not would be a pretty strong differential so both manufacturers and buyers are likely to take notice.

 

A benchmark would be entirely useless and do more harm than good. Best case scenario, the benchmark does nothing useful. Worst case scenario, it tells users that their phone is fast and should handle it, they buy it and are hit with compatibility issues.

IF Unity was to do anything about this the best thing they can do is attempt (and doubt that would be easy) to detect every GPU feature you are calling from your shaders and game, plus other hardware requirements, and then keep an internal database of what devices are confirmed to have those GPU features. Use this database to generate a filter file the Android market can use to hide the app in devices missing features.

This would not be easy for many reasons. Even if they were able to generate the list of GPU features used, Android devices are famous for incorrectly reporting support positives for certain OpenGL features. That's what you get when you let hardware manufacturers build the OS.

Even if it was handled right, it may lock you out of handling certain devices in a different way. There would be no way for you to say “if this device is missing this feature, then use this other shader instead” without Unity simply flagging the feature as required.

We will just have to live with it until Google decides to standardize hardware requirements (never) or Android fades away (more likely than the former option.)

I like to think the included shaders have been widely tested in Android, so for now I'm just trying to stick to using the stock shaders and not go fancy with any with special shaders and FX. May not make me imune, but I cross my fingers and hope it minimizes incidents.