This is an update to the Mobile Game Engine Popularity Index I published 2.5 months ago.
Game Engine Popularity on Stackoverflow.com
Since November’s chart I made sure the tags I’m most interested in (cocos2d-iphone, cocos2d-x, sprite-kit) were properly applied to all questions.
Furthermore I update at least the past 6 months for each tag, which caused the graph to change noticeably. This may be due to policing the site, where users remove tags or retag questions. The net result is that there are now recent months with fewer questions for some engines compared to November’s chart. In particular this flattened the up-down curve of cocos2d-iphone. Continue reading »
Topics: Overall interest in cocos2d is waning. Unity and libgdx fighting for 1st place. Sprite Kit skyrocketing from day one.
This sums it up. Now for the details. Since my last game engine popularity measurement I tried to improve and streamline the process.
Game Engine Popularity on Stackoverflow.com
For stackoverflow.com data I’m searching for tagged questions created in a specific month. The search query looks like this:
[cocos2d-x] is:question created:2013-11
I repeated this for every tag for every month as far back as there were questions with the given tag. Continue reading »
For the past two weeks I’ve been running a Cocos2D Developer Survey. As of today, 236 developers started the survey and 189 finished it completely. That’s 80% despite the many questions they had to answer.
Here are the results with my observations. I started the survey also to see if I was on track with KoboldTouch, and whether certain assumptions hold true. Specifically I had a hunch that cross-platform development is only perceived to have great value or appeal. Let’s see if I was right.
Click on each image for full resolution.
Who are you?
I was very curious how many cocos2d developers consider themselves to be hobbyists and indies compared to professionals, who either work for a mobile developer or are taking on freelance jobs as one.
Almost half of those who answered the survey are hobbyists. Nearly 30% consider themselves indies who make a living making mobile games. This is great! Continue reading »
The previously command-line-only TexturePacker tool now has a nice GUI. It’s called the “Pro” version and for a reason. So far, I’ve been using Zwoptex for creating Texture Atlases, but I’ll certainly give TexturePacker Pro a try.
At first glance, what I liked was that it told me when the Texture Atlas was “full”, eg. not all sprites could fit into that Texture Atlas. And the always present list of image filenames on the right side is a welcome feature. Although Zwoptex does have this list as well, it’s almost rendered useless because it’s a seperate view option.
Zwoptex’ price has changed since I checked last time, it’s now at only $14.95. TexturePacker Pro costs $17.95, it’s command line version $9.95 and the upgrade from command line to Pro is $7,95, so you’re saving a whopping $0.05!
A couple words on pricing
I find both tools are underselling themselves. Zwoptex was initially at $24.95 and even that price seemed “cheap” to me, given how much trouble it saved me and how much faster and more memory efficient it made my projects. I would say that a price range of $30 to $50 would be more than fair for those tools. I can imagine that the TexturePacker command line version at $9.95 and now the Pro version probably forced Zwoptex to adjust its price.
Problem is: this isn’t a market where people choose their tool based on a $3 price difference. Also, it decreases margins for upgrade prices, with TexturePacker upgrade already below $10. Those low-end prices incur a proportionally large amount of transaction fees, paid to the eCommerce vendor, making them less viable. Since this is also not a mass market, but a niche, it would be wiser if both upgraded their prices back to reasonable regions, around or above $30.
Likewise, Particle Designer is also absolutely undervalued at $7.99. Those low prices undercut the value in tools, and make tool development less and less attractive than it already is. And if anything, cocos2d and the other engines need one thing above all else: tools. And good ones!
Be it Ricardos mysterious “world editor” or the Physics + Tilemap IDE (website) or the other game editing tools currently being worked on - if any one of them is being released as a commercial product but sold for less than $30, I’m going to be very mad at you! If it’s less than $60 I’ll still be mad at you, not very, but still mad.
Rant
Seriously, this isn’t the App Store! It’s developers you’re selling to, and they do value useful tools. Just because cocos2d is free doesn’t mean that all tools surrounding it need to be cheap (or free for that matter). Take Sprite Manager 2 for example. It sells for $75 (per seat!) and isn’t all that different from Zwoptex or TexturePacker Pro. If it were a standalone app it would probably pale in comparison! You might argue that SM2 is for Unity, and their developers are less price sensitive - I don’t think so, and some are even more price sensitive because they just made that huge investment in the first place.
In general you could say that those developers who invest several hundred $$ into a game engine (and toolset) are simply more serious about game development. You do have those enthusiasts working with cocos2d as well, but they’re probably outnumbered by a lot of hobbyists and “I’ll give it a try” kind of people who simply join because of the fun involved, and because the only investment in cocos2d is time and they got plenty of that. The question is: do you want to be kind to the hobbyists, or do you want to build a sustainable business? Plus, Zwoptex and Texturer Packer are also being used by Corona developers.
Now you may also be arguing that a cheaper price allows more developers to enjoy the tools and we’re a friendly bunch and not a commercial, greedy corporation. Sure. But those prices do devalue everyone’s tools, so if anyone wanted to build a tool that takes maybe not just 1-2 months to build (initially), but maybe 4-6 months or more before it’s going to be useful, those price points are not very encouraging to start such a project. That doesn’t mean it won’t happen, but I do know that those people who could pull this off, are generally terrible at doing business. And easily influenced by comparable prices, punching a few numbers, and then getting on with their next train of thought which probably involves solving an obscure programming problem.
And the kind of tool that needs this time investment is the one I’ve been looking forward to since I first started working with cocos2d in May 2009. Back at the time I was convinced that by the end of 2009 cocos2d would be having a game editor or at least something to build the GUI and screens with. I wished for a fully-fledged, professional game level editor, with a physics editor, sprite animation builder, asset management, and whatever else you can dream of.
Now, if that ever happened, it shouldn’t cost less than $100. If you can provide killer features like Box2D integration or scripting game logic, ask twice as much. And offer Lite and Pro versions with a variety of feature sets to make the most out of what developers need and what they are willing to pay extra for niche, but very useful features if you need them.
As summarized by DaringFireball, Apple has loosened their restrictions of section 3.3.1 of its iOS Developer Agreement:
In particular, we are relaxing all restrictions on the development tools used to create iOS apps, as long as the resulting apps do not download any code. This should give developers the flexibility they want, while preserving the security we need.
Previously, the only programming languages allowed to write iOS Apps were C, C++, Objective-C and Javascript. This has now been removed. For cocos2d developers nothing changes, except maybe that you can feel more comfortable embedding a scripting language like Lua into your games. As long as you don’t allow the Lua scripts to be changed by users, or download or otherwise modify/replace bundled Lua scripts. That wasn’t illegal before, however, yet after the change in section 3.3.1 it put a lot of doubt and worry into developers looking into using Lua. So you can now feel much more comfortable using Lua in iOS games, for example by using iPhone Wax.
The removal of these language restrictions is essentially good news for Unity developers (read their statement), and those who wish to develop iOS Apps using a Flash cross-platform compiler and also those using Corona Game Edition, which is entirely Lua-based. And speaking of which, Corona offers developers to purchase Corona SDK and Game Edition at just $99 until only September 15th, after which you’ll have to pay for each product seperately and the price goes up to $249. Just in case you were eye-ing it.