Experience using the game designer Game salad
This review is an experience gained through trial and error. A few months ago, I myself was looking for such information, but in vain. Therefore, I hasten to share with future seekers of easy ways to create mobile games, lovers of "simple solutions" and boxed products. I disassemble the visual constructor of games Game salad.
I acted in this project as the “manager” and the author of the project, but the programmer had to figure it out. Below is its text.
GameSalad is a constructor. Scenes are built from actors, within which rules are compiled for behavior in various conditions and attributes are added as desired. There is a certain set of blocks for changing the state and behavior of an actor. So in general, the system provides a sufficiently large freedom for manipulating the actors themselves. But the biggest inconvenience is connected with this.: There are some insurmountable restrictions for controlling one actor by another. More specifically, for actors who spawn during the game (rather than manually added in advance), there is no way to influence them from the outside, only the rules established inside them. This inconvenience becomes a problem as the complexity of the game increases.
Another drawback that could significantly compensate for the previous one is the lack of inheritance. Actors could inherit the rules from the ancestor and in addition possess their own, it would be very convenient. And when constructing the scene, the actors could be invested in each other, this would provide significantly more opportunities for interaction between the actors. In general, I was surprised to find that the scene does not consist of a hierarchy of actors, but simply from their list (divided into layers).
GS - the system is completely closed. It does not support plugins or extensions. It does not have any import / export mechanisms. If something doesn’t suit you in the functionality, you will have to put up without options, you can’t expand in any way.
Documentationdeserves mention too. In addition to the documentation, there is a collection of tutorials that describe individual functions. Of course, the articles describe general cases, so nothing is explained in detail, the options are mostly not described. Another way to get information is through the forum. The method is very effective, developers and users respond essentially, discuss, provide links. The forum has a huge amount of information that is not found anywhere else in the documentation. So you have to use the search, it often bears fruit. But still, looking for the information that is often needed and for many is somehow wild . Could collect FAQ or popular topics. Ideally, include in the documentation the information that users most often need, judging by the forum.
As for the game in general: GameSalad has only one method for sharing the result in the game - TweetSheet (send a message via Twitter) - and it is only for iOS. Several integrated advertising systems, but no banner management capabilities . All controls in the advertising systems themselves and nowhere in any quick launch tutorials with GS.
Summarizing the essence of the program. Suitable for creating simple games with simple mechanics. Scalability is low, a large and complex game cannot be made. On the other hand, the program does not require any initial knowledge and skills. All blocks have a clear description and understandable action. So for people not familiar with programming, a good option. Programmers need a strong reason to use it (for example, the speed of development).
Summarizing my attitude. I appreciate the fact that the program makes it possible to create games, having only the desire for this, no special knowledge, skills, even talent are needed. But if the structure of the program is a little more complicated, not so straightforward, it could offer much more opportunities without complicating the process. But I will not hope for a speedy development of the project, given its age and degree of development at the moment.
I think that you need to choose a program based on its functionality. Bugs are a business that comes and goes. You can usually get around them, waiting for the next update, where they will be corrected.
For those who decided to use the GS, I give a list of technical flaws
GS has a version for Windows and for OSX. For Windows, the current stable version is major number 0.9, for poppy - 0.12. Not because they have their own numbering, but precisely because the version for Windows lags behind in functionality by 2 versions. This is a lot: there are blocks that are long outdated and already excluded in the mac version, and some of the mac version are missing. Now in the nightly builds (which require testing and detecting bugs), the Win version has the number 0.12.1, while the next version is already being tested for the poppy version - 0.13.2.
In addition, there are bugs in the work of the wine version. A lot of bugs. In the interface, in saving some blocks, in previews, in opening projects created on another computer, crashes and more. There are simply too many to list. But they are enough to solve: using the wine version is completely impossible. If you want to use this program because of its advantages, the only option is the poppy version.
The only positive point in the version for Windows is that for my taste, the interface logic is better organized there.
Mac version problems : in the formula editor text strings are enclosed in quotation marks, but the editor itself does not allow printing spaces. You have to replace hyphens, or underscores, or come up with something else trickier if you really need to.
A more serious problem is that GS crookedly signs android apk. I used the certificate store created earlier in Windows, and the signed file refuses to be installed (without a clear error message. Just “the application is not installed”), even from the market, on devices with OS version lower than 4.4. I decided to sign the file from under Windows with the same certificate. So maybe the thing is in the android-sdk poppy version. Here I can only say that the SDK is in both versions of the latest version.
Signing an android apk file using GS directly from a browser is only possible on a Mac. Why, I don’t understand, but the web page gives a warning. Although there is an apk-signer in the GS version of the wine, I haven’t been able to use it.
The end of the quote.
I’ll supplement from myself, as from the “manager”:
- It was wild and painful to understand that we are throwing one feature out of the intended functionality. On the forum of this editor, people wait for 2-3 years from developers for the implementation of the “share” function, social solutions and other solutions vital for the game.
- We did not set a goal to analyze thoroughly all the pros and cons, but only share our observations.
- Advertising management mechanisms and their wretchedness have made us a choice: to spoil the game for people or refuse advertising. We did not spoil the game, but this moment killed all hope of making fast and high-quality cross-platform games.
In total, we spent a month and a half of work, 17 thousand rubles for the editor (bought in dollars at the wild exchange rate of autumn 2014) and created 2 games. Not to say that the products are a failure, but without the necessary elements and without advertising. Further prospects for using this solution are vague.