|
[This is a repost from my blog, doolwind.com]
I hear a lot of game developers refusing to use Unity for web games because of penetration. David Edery made a point of discussing this during his keynote at GCAP last year. Today I’ve decided to formulate my thoughts on why I disagree with this argument and why I think you should be using Unity.
Multi-platform development
If, like us, you’re making games for web and iOS then this is really a no-brainer. The extra development time is not something we can afford when there’s a perfect alternative in Unity.
With Flick Buddies, we developed the web and iOS version simultaneously without any extra effort. All our games in the future will see simultaneous releases for both web and iOS which will help drive customers from the free web version up to the paid iOS version.
The main argument against using Unity is that with such a low penetration rate (for arguments sake we’ll go with 1%) users will bounce from your site when they hit the Unity installation page. Taking the statistics (from Unity and others) lets go for the low end and say that 50% of users will bounce when they see the Unity installation page.
Rather than looking at it from percentage of users being lost, let’s look the other way around. If you’re developing an iOS game you can release a web version for relatively little development cost and capture 50% of the potential web market. If you were to instead create a flash version you would have the full development costs of porting the game to get 100% of the market.
Looking at the cost/benefit ratio the Unity version is a much better option. The only time this becomes a negative is if you have a highly successful game, in which case losing 50% of the market will end up costing an order of magnitude more than the development costs of porting to flash.
In that case, go for it! There’s nothing stopping you from later creating the game in flash once you know there’s a large enough market for it. Where the cost of porting the game will easily be made up by the 50% more users you’ll receive.
Another interesting point I’ve heard is that the more popular a game is, the lower the bounce rate. If you have a highly successful game that people really want to play, the act of installing a plugin will be less of an issue for them. This goes some way to negating the lost sales for a highly successful game.
Web games only
What if you’re making web games only? In this case it’s not quite as clear cut and we have to dig a little further. I’m still inclined to go with Unity for web only games for a number of key reasons:
- You can give a richer game experience including 3D. While 3D is coming to Flash soon, it’s at a lot lower level requiring far more development time and cutting out many less experienced developers.
- Much richer tools and pipeline. I’ve previously discussed how much I love the Unity editor and asset pipeline. This lets you get your games out faster and cheaper. This saving offsets the 50% bounce rate.
- There’s the opportunity to have a unified language running between client and server when using C# within Unity. This simplifies communication as well as allowing the exact same code to run at both ends if required.
The Minecraft example
I like to use Minecraft as an example of bounce rate when discussing Unity and Flash. Minecraft requires an .exe to be downloaded and run before the game can be played. This is far more intrusive than a browser plugin and yet it’s still had great success. With Unity developers can create something as successful as Minecraft with a lower bounce rate due to installation and have the ability to port to iOS and other platforms easily.
Flash was once like this
I remember when Flash was at a similar position to Unity, albeit with web pages rather than games. Many people said that you shouldn’t make websites with Flash as it required users to download a plugin if they didn’t have it. Not only did Flash reach the penetration rates it desired, times have also changed in Unity’s favour:
- Faster downloads mean there’s less of an issue in the download time of the plugin
- Seamless installation reduces the bounce rate with unsophisticated users
- Larger sites (like Kongregate) are getting behind Unity which makes it more trustworthy in users eyes
What can Unity Technologies do about it?
I saw a similar issue with Silverlight penetration and I find myself again thinking of inventive ways for companies to increase penetration of their plugins. This could be as simple as a referral program, offering 1-10c each time a user installs the Unity plugin on your site. This motivates developers to make awesome free games with a guaranteed return as an alternative to advertising. Unity can then buy installations for as little as $10k per 1 million users. It’s in Unity Technologies best interest for penetration rates to go up as quickly as possible and I’d like to see a novel approach to achieving this. While the penetration rate will naturally increase over time, it will take quite a while.
Conclusion
What are your thoughts on Unity's penetration rates? Is the bounce rate just too high for you to switch over?
|
The "make better games" argument stated lower in this comments thread is ridiculous imho, considering the potential user came to the website to play the game, so there is obviously intent and desire to try it, and they haven't played it so they can't even make a judgement yet as to the quality of the game. It is easy for even low-moderately tech able people (let alone extremely able folks like us) to get past barriers like these and minimize the issue, but don't overestimate the young and casual markets - that's all I'm saying. Don't even get me started on the amazing prevalence of ridiculously low end hardware and dial-up ping times we were seeing. Be aware of your target and measure the risk accordingly before you commit to a platform.
The fact that EA had licensed unity as their online game engine http://www.gamasutra.com/view/news/30571/EA_Signs_Broad_MultiYear_License _For_Un
ity_Development_Platform.php will help with the penetration issue.
So is JOGL now a realistic option? Or what does Minecraft in the browser use for 3d? Perhaps, Notch has paved the way for 3d in the browser!
I also think that grabbing a plugin doesn't stop people. My Wife loves Pogo and she had to nab a plugin to play, she just did it. She is in no way a techie type and I was actually pretty impressed that she just did it without calling me over.
If the user sees the content as desirable then installing a plugin will not stand in their way.
Exactly! I didn't want to openly come out and say this, but I have definitely heard this from a number of people. If you're game is no good then even the smallest barrier will bounce people.
A great move that Unity Technologies should do in this front is adding Steam integration out of the box for its next Unity release. I recall one major game portal willing to help others into doing just, on a case specific-basis - which included releasing the game in their portal as well, if I recall correctly.
For truly casual games the best bet would probably be waiting for Unity to a) double its user-installed base b) add HTML5 deploy option. Obviously option B would require that you also wait for HTML5 to get good performance and become truly compatible for the major browsers - IE especially.
So the question is, is there a forcing function behind unity, such as awesome exclusive releases? If not, people would probably just prefer downloading an running an exe.
At the other end of the spectrum, I like the idea of html 5 providing a cross platform solution. But thats just a pipe dream, if you've tried running an html 5 game on android, iphone, ipad, or google tv, you'll know that issues are still present.
But I do like the idea of HTML 5.
This is sounding a little "chicken or egg" to me. If more people use Unity exclusively (as we are) then there's a higher chance of a big hit which will drive up the numbers. I guess it depends a lot on general developer adoption. Perhaps, psychologically, I wrote this article to try and convince more developers to come on board to help drive up penetration as that will help our company in the long run :)
This pretty much says it all IMO. It's not like installing the plugin is a huge deal either.
Don't forget grandmas and mothers, they are a really big market share
I imagine though that if said game was a blazing feat of gaming genius, eventually the games sheer mass appeal would be so high that people would be willing to give it more time and effort to get the game to work, but the reality is that most games, even good well made games won't have that kind of critical mass. Most games must rely on as many people who come for the first time to play it in order to help assure a better chance for success.
It's possible people that already has the plugin, or that even installed it, is leaving in the middle if the loading time.
They advertise that they're "cross platform" but I personally feel their inaction speaks volumes.
I would like a representative from Unity to officially make a statement regarding this if it hasn't been done already.
We've had this for Windows for a while and we are very soon also adding Java Web Start support on OS X.
In terms of real numbers people have achieved with this. A couple of facebook games that have tweaked their install process graphics for their game. For example one game has achieved a 92% install rate from the moment of the user clicking the button to the plugin actually getting installed. 72% actually do click the button.
This is highly dependent on the presentation that you give to the user. Showing some images of the game, that you are about to miss or better a video. Clear instructions on what to do once the install button has been pressed, make a massive difference.
I definately agree with the other points though. Unity allows you to deploy to more than just the Web. Cross platform publishing to PC / Mac / iPhone / Android / consoles is a big deal. We had a lot of developers who also made some good money on the MacAppStore. The main point is that, it's always good to be able to push a game into many different channels.
Bounce rate is bounce rate once in the game, and will be similar whether the game is in Flash or Unity. i.e. There will always be a proportion of people that will not like it or not want to play it.
In this case I think the decision to go with Molehill is warranted - Facebook is a special beast. Highly casual, highly consume/discard.
"If the user sees the content as desirable then installing a plugin will not stand in their way." is all well and good for *talk*, but at the end of the day as Chris said in his comment right at the top "...casual audience bounced in the 70%-80% range". Results peak louder than words (assuming Chris accurately measured this).
That's about as niche of a reason as possible.
Every now and then I'll realize that I didn't have Unity installed, and that's usually when I'm seeking out tech demos. Who's actually accidentally ran into must play Unity content?
Seriously guys, high bounce rate? Make better games.
it's interesting how people talk about market penetration with flash, which is true, but flash still has to be installed. Its not like it's part of OSX or Windows...
I guess its just that people run into flash more often than unity... but users get flash because it is so prevalent and they will keep running into prompts to install flash till they do or not see web sites / parts of sites.