Unity Game Engine to Get Official 2D Game Support and a Built-In Ad Solution | Hacker News Reader

Unity Game Engine to Get Official 2D Game Support and a Built-In Ad Solution

m.techcrunch.com

108 pointsevo_942 comments

Note that many 3D games need 2D components for menus, HUDs and general UI. Most Unity developers were paying an extra 100-200 dollars for 3rd party plugins that brought in better text handling, sprite atlassing, drop shadows, etc. Even if you're not planning to make a 2D game it's a big upgrade of the tool.

fwiw the most popular 2D libraries are less than $100.

2D Toolkit - $65 (http://u3d.as/content/unikron-software-ltd/2d-toolkit/1Wi)

NGUI - $95 (http://u3d.as/content/tasharen-entertainment/ngui-next-gen-u...)

Uni2d - $90 (https://www.assetstore.unity3d.com/#/content/3826)

All of these have more features than what Unity just introduced so I don't see their usage declining anytime soon- until/unless Unity makes some serious improvements to their built in 2D library.

I noticed some cool features in their unveil - such as polygonal sprites rather than rectangles. This means it's cheaper to fill each texture to the screen, and it means the sprite atlases can be packed tighter resulting in less draw calls. They also changed the 2D UI so that you cannot select a sprite where it's alpha is 0 under the cursor, which makes it way easier to move and manipulate in the editor. Finally, they added an animation framework for the sprites which is powerful.

Free is powerful

I feel sorry for the guy who made the 2D Toolkit [0]. It's a very solid asset, but with a new built-in official competitor I think it might die, or at least slow down over time.

Would it be too cynical to think that this is part of the point of the asset store? To let independent devs run thousands of experiments at their own expense, let them all run, give them their 70%, then just reproduce and integrate the very best of these assets with the next version of Unity?

I suppose from the Unity team perspective, it looks like they can't do anything right: don't add enough features and people will complain about over-reliance on the Asset Store - but add too many features and Asset Store publishers complain.

There's probably one of those 'yeah well, sounds obvious' lessons in here: don't make your product overly dependent on somebody else's platform. Easy to say that, of course.

[0]: https://www.assetstore.unity3d.com/#/content/908

FWIW, all the "2D extensions" asset store publishers have known about Unity's upcoming 2D features for a while (have been part of the alpha testing etc.). Yes, their toolkit sales will probably drop, but that's what you get when targeting a missing feature of a product - sooner or later the product will add that feature. They are smart folks so I'm quite sure they will find something else to do though; or will build even fancier extensions on top of base functionality.

"I suppose from the Unity team perspective, it looks like they can't do anything right" - I don't think we have this problem. If there's a big missing gap in features, we will close it sooner or later. Yes this might upset some asset store publishers, but that's how things are. On the other hand, many of them will have had several years of really, really good sales of their packages.

(I work at Unity. YMMV etc.)

Yeah, you're probably right - they knew the risks going in, and it's not like they haven't benefited from the Asset Store.

Ultimately I'm willing to bet that improving the core product will mean there'll be more opportunities for asset store publishers, not less.

So, obviously you are right and this is an issue with any platform. But I do wonder if there was a better way. Couldn't you in many cases offer these people a job at the company? I mean, I guess Unity is hiring a lot anyway.

Well, if you look at it... RageSpline guy (Juha Kiili) is working at Unity (on the same 2D stuff for 4.3), and NGUI folks (Michael Lyashenko & Philip Cosgrave) are working at Unity on the upcoming GUI system.

That said, some people do not want work for any company; they just want to be "independent". And that's ok.

I think 2D toolkit is a great asset too, hopefully they got something out of this and will keep on doing good work.

But I think what most people are missing is that the real game changing 2D news is the new 2D physics engine (box2d). Doing sprites has been a tiny hassle compared to using the 3D optimized physics engine we had before to do accurate 2D stuff.

Some day they might even support proper wireframe rendering (without third party plugins)! Code based .unity scenes for easy version control? Anyone? Unity3D, what a love hate relationship I have with this thing. I will admit though, that the love grows with every day! It still boggles my mind how few issues I had developing a major project with Unity3D+C#. No stack overflow or segmentation errors, hardly any crashes. The ability to play and pause in the editor is just as great as the multi platform ability.

Unity scenes already support text serialization in the YAML format.

Have you tried merging those scenes? Even if the format is structured and in plain text, it does not mean that a human can manipulate it as a patch. The format is so convoluted and full of magic numbers, it would require a lot of experience to really merge them; or support from Unity itself, and then the choice of serialization is not really that important.

You can manage team work somehow with prefabs. But I would like to see something like Havok Vision Engine's layering system, where every layer of a scene is it's own file, thus with file locking you can work on a single scene as a team. For example, you could split the scene in layers by developer, and then have someone integrate them in proper layers. Or you could just split specific work between people with lighting layer, static layer, collision layer and so on.

Unity supports wireframe rendering since 4.0.

Thanks! Didn't know that.

A little backstory about this- the most popular 2D plugin for Unity was NGUI (now I would say 2D Toolkit and NGUI are equally popular, but 2D Toolkit has more momentum).

The guy who created NGUI was hired by Unity about a year ago to work on their new built in UI / 2D system. And, this is what he's created. It's still massively lacking in features compared to the 3rd party 2D plugins but I think this is just a start and we should expect to see a lot of great improvements over time.

These are the 3 most popular 2D plugins for Unity, and they're all less than $100:

2D Toolkit - $65 (http://u3d.as/content/unikron-software-ltd/2d-toolkit/1Wi)

NGUI - $95 (http://u3d.as/content/tasharen-entertainment/ngui-next-gen-u...)

Uni2d - $90 (https://www.assetstore.unity3d.com/#/content/3826)

Personally I use 2D Toolkit and I know for a fact he has plans to make his plugin use the new built in 2D Physics system and I'm sure he's looking at any other ways he can squeeze better performance and useability out of the new Unity 2D features.

I would love to stop using a 3rd party 2D plugin but I don't see it happening anytime soon.

I wonder if the timing of the release is tied to the release of iOS 7. There is a Sprite Kit 2d framework in iOS 7.

Upcoming 2D tools in Unity 4.3 aren't related to SpriteKit of iOS7. We support more platforms than iOS, you know.

Since Unity is targeting iOS, Android, Blackberry, WinPhone, Mac, Windows, Linux, Web all equally, I don't see a big correlation there. Also seeing Android's global market share, iOS is basically a fringe target for me these days, just slightly above BB and MS in terms of relevance.

With their Asset Store, they can easier gauge what Unity users need so badly they'll even pay for. Unity is used a lot for 2D games, yet is very 3D-centric, so there are many popular and I presume profitable 3rd-party extensions to make 2D first-class in Unity. So it makes a lot of sense for them to cater more "natively" to this segment of users.

Interesting - there are hundreds of millions of iOS devices out there and their average user spends more money on apps than the Android counterpart.

Are we already at the point where Apple is irrelevant to the app vendors?

I don't think so, because larger mobile operating system share (especially global, where you have to account for the Chinese and Indian markets) doesn't necessarily mean more people will actually be interested in your app, can find your app, or can run your app on their phones. Your potential market is larger, sure, but out of that larger customer pool how many will be able or willing to download and use your app, especially if it's something performance-intensive?

For a lot of people, the smartphone they buy is already fully functional for their needs out of the box. They get an Internet browser, media player, e-mail client, and some other utility/PDA features that are bundled with the OS. A lot of people will add 'service apps' like Facebook and Dropbox to access services they were already using, and then stop there. There is less of an incentive for people to go out and look for apps to make their smartphone more useful, like a 1990s computer purchaser going out and buying Microsoft Office and Quicken for their new otherwise-useless Windows box, and so I would argue that marketshare by itself is less meaningful a metric for comparing mobile and desktop platforms.

At least for big game companies iOS is still by far the number one platform. Many high profile games still aren't available on Android, Clash of Clans or Plants vs Zombies 2 or Kingdom Rush Frontiers just to name a few I've been missing in the last few months.

Now it's competing with Cocos2D.

Yeap, except it's a complete different league. Cocos2D-X (which is the real cross-platform equivalent) is an absolute nightmare for beginners. Last time I checked, there was no documentation except in chineese, and you wouldn't even know where to look when you wanted to do some networking.

Unity truly is a marvelous piece of technology and ecosystem.

I actually bought "Cocos2d-X By Example Beginners Guide" and it was very easy to get up and running with a demo application. The samples included with the framework itself show off a features and nothing has been all that complex, and I have made a bunch of 2D test applications already (XCode + iPad simulator)

Thanks for mentioning this book, I've wanted to dig into cocos2D for a while but the lack of a good beginner book kept me from getting too far. I picked it up last night and it's exactly what I've been looking for. Awesome thanks!

The biggest problem I found with Cocos is trying to google for something. If you're developing for iOS, then it's fine. But if you're looking into, say HTML5 canvas, finding a working JavaScript example for something is like looking for a needle in a heystack.

This is one of the main reasons I gave up on Cocos and used EaselJS for my last project.

I think soon you will have good news about this from the cocos2d-x side.

And MOAI, which is far better than Cocos2D, imho, although requires a bit of effort to truly 'grok' properly so that you can get things happening ..

That said, you can't deny that Unity has the toolset market pretty much cornered. If someone came up with some editors for MOAI, it might be an interesting world indeed ..

Except there is no way to create an editor for moai.

Moai is a very low-level framework. Basically just a tad bit higher than OpenGL, + lua bindings

Thats really not true - its quite possible to build an editor for MOAI .. in fact, I would wager that it'll be easier to build an editor for MOAI now, than it was to build the environment for Unity...

Any opinion on Moai vs CoronaSDK? I know people who use Corona, but not Moai.

Moai is opensource, lower-level engine that appeals more to those that want full control and customization. Tooling is mostly "bring your own editor/ide" (Intellij has a good Lua plugin[1] and there's also Zen Braine studio[2]). Personally, I would rather use it over Corona, but those aspects are appealing to me, but could be off-putting to others. You'll be looking through the Doxygen documentation and perhaps even the C++ source at times to figure things out in Moai. The plus side is much of the design/best practices of Corona Game Examples (or any other Lua Game framework, like LÖVE) can be used to model your Moai development, omitting the API differences.

Corona is more for those that want a high level game engine and all the tools for it built in as well as more beginner friendly (larger community, documentation, ease of use). The drawbacks are limitations from the source being closed and in the past there were performance issues for Android/iOS, but I have not tried Corona lately so it may have improved in that area.

[1] https://bitbucket.org/sylvanaar2/lua-for-idea/wiki/Home

[2] http://studio.zerobrane.com/

A lot of people seem to be dissing cocos2d here. The iPhone version is actually really good, it's the tech behind many successful iOS games such as Kindom Rush, Badlands, Megarun, Scramble with Friends, Tiny Tower, Feed Me Oil, Tap Zoo, Zombie Farm. Many of which have reached the top grossing list. How many 2D Unity3D games have reached high up on the top grossing list? Is there any other game than Bad Piggies?

Also, now with Apportable, porting to Android is a breeze. It's possible to have a running port within a couple hours.

And another good thing with Cocos2D is that you get the full source, allowing you to do basically anything you want and get good performance out of it.

Not sure about Cocos2D-X, but now thanks to Apportable, it's now possible to stick to Obj-C with is really nice.

"How many 2D Unity3D games have reached high up on the top grossing list?

Quite many, I'd say. Looking at top 50 grossing for iPhone right now, I see Temple Run 2, The Hobbit, Kingdoms of Camelot, Subway Surfers, CSR Racing at least; and probably some others I don't recognize. Many others have been in the top charts at various points in the past.

Are these 2D games?

Not sure about Cocos2D-X Ricardo Quesada, Cocos2D creator, is now moving from Zynga to Chuking Technologies (Cocos2D-X): https://news.ycombinator.com/item?id=6348565

And he is proficient in C++ and Objective-C.

Indeed, this will make Unity a direct competitor with many cross-platform 2D game SDKs. Here's one list (on a site I maintain):

http://www.mobilechameleon.com/toolkits-for-making-2d-mobile...

It was already, for a really long time. There are already quite a few capable Unity3d 2d frameworks, and a lot of companies created their own in-house solutions anyway.

Having some solid experience with Unity, have to say that it was already great environment for creating 2D applications. Only sprite management and object alignment support had to be reinvented from the scratch, but some very solid addons are present on the market for pretty long time. If I recall correctly, the author of the most popular one, NGUI, was hired by Unity to implement native 2D support.

Funny. I was googling for exactly this earlier today. I love the thought o using such a mature engine and tool as Unity but I mostly make 2D games so this is very welcomed news.

Still waiting on that PS Vita support they promised.

Reply on news.ycombinator.com