Frankly, Whats the point of c3?

Now the question is whether you really need Unity if you are not developing 3D games

Short answer is Yes:

For the last 5 years I had fun making games in C2, but when it comes to publishing them, integrating ads, leaderboards, etc...I was always at the mercy of plugin developers. CranberryGame had a lot of plugins that I used in my games, but for the last year it disappeared. Some of them are not working anymore and he is not interested I guess in updating them.

With Unity there are always more options to choose from in case one of the developers dies or is getting hostage when going in Holiday.

I agree that plugs really are a big issue, and have been for a while.

I also think that Scirra is going to have to lead the way in fixing it. We need better support, and set rules over all aspects.

From behaviors with no license to store plugs that have zip documentation, and even less support.

I mean the thing is useless without plugs.

Time to move to Unity which is 100 more powerful and FREE. I have C2 pro licence, but I don't seethe point in upgrading my licence to yearly payment for the same futures as in C2

I would recommend Unity, but also Construct 2, and Construct 3. I pay for all of them, and they all have their strengths and weaknesses:

Unity

• $16/month for Unity Teams: 25 GB of cloud storage + 4-person Unity Collaborate
• $100 one-time for Steam
• $25 one-time for Android
• $100/year for iOS
• When we publish next year, we are planning to spend an additional $35/month to get rid of the "Made With Unity" branding.
• Tremendously flexible, and produces great results for 3D games
• Extremely bloated editor
• Creates extremely large packages upon export
• Unity WebGL export is useless for 3D games. Its exports are ridiculously bloated (2 hours to export + 1 GB for the project + hundreds of megabytes of memory use) and slow (5 fps for something I could get running at 60 fps in Q3D). We are hoping to make an AirConsole version of the game, and I'm dreading the fact that I will probably need to write a three.js or babylon.js version by hand. I could probably do a limited version of the game with Construct 2 + Q3D, but by then Q3D will be four years old with no updates.
• Bottom line: If you are exporting to Steam, iOS native, or Android native, and making a 3D game, Unity is a no-brainer. If you are making anything for HTML5, Unity WebGL is mostly nonfunctional for desktop web, and entirely nonfunctional for mobile web.

Construct 2

• $430 for a professional license (one-time payment)
• $600/year for web hosting to handle 50,000-100,000 pageviews/month (shared with all web projects)
• C2 is the best thing around for making HTML5 content. The vast majority of my user base is HTML5-only (education).
• Supports Q3D, which is the only reason I still use it.

Construct 3

• $150/year for a business license
• $600/year for web hosting (concurrent with C2 projects)
• The C3 editor works on my MacBook Air. This is huge for me, as carrying around a second laptop for development is really cumbersome.
• I do all of my 2D development with Construct 3 now. It is the right combination of fast and lightweight, and it produces relatively tiny exports.

For HTML5 export, Construct is WAY better than Unity. Since all of my current development income is from HTML5 work, switching to Unity is a risk. I am banking on the income I make from our first big Steam game to make Unity development financially viable.

Why am I currently using all three? Because each of them is the right tool for a different job. For my team, Unity is not free -- it is the most expensive of the three tools. However, the cost of the software in all three cases is much smaller than the cost of maintaining an LLC, attorneys, purchasing assets, etc., and utterly insignificant compare to the costs associated with the amount of time I spend at it. For a 1500-hour project, labor costs are around $75,000.

I agree that plugs really are a big issue, and have been for a while.

I also think that Scirra is going to have to lead the way in fixing it. We need better support, and set rules over all aspects.

From behaviors with no license to store plugs that have zip documentation, and even less support.

I mean the thing is useless without plugs.

This is one of Construct's biggest downfalls really. It is why Blackberry and Windows failed - lacking dev support. It goes two ways, Devs support Scirra and Scirra supports Devs. The bond is just not there like it used to be. This has some pretty damning impacts on current and future users.

cjbruce has the right idea. You use what is best for the job you are doing. For people with less experience, it is hard to know what is "best" I recommend plenty of research and testing things out. Don't start any big projects without testing the waters first of each option. IE: Do some prototypes, check out features and ease of use. Check out the support documents and user complaints/praises. If you just pick a tool at random, without any real thought . . . you are going to have a bad time. Once you are more experienced, you will notice you use a vast range of tools typically...dabbling in each one for certain tasks.