Okay, so a further update to this. I managed to fix this issue, but I don't understand why I needed to do this.

On start of layout, I have to set the position of the walls using ViewPort values. Then, having to subtract half the size of the wall. So, ViewPortLeft(0) and so on.

It's like the anchor behaviour is running too early?

Thank you, Geoff. I did have that initially actually. The problem is my right wall seems to get cut off, despite the fact it is being anchored.

You can see top, left and bottom are anchored properly. The right wall seems to be just out of view. This is how I am anchoring it.

I am using scaling inner. So, perhaps I am doing something weird here when I anchor my right wall that isn't obvious, or something else.

I found someone else describing a similar issue with anchor and scaling here: construct.net/en/forum/construct-2/how-do-i-18/anchor-sprite-top-left-113871

The Nintendo Switch is emerging as a boon for indie developers. I read on VentureBeat just today that YoYo Games is teaming up with Nintendo to create a special edition of GameMaker Studio 2 which supports the creation of Nintendo Switch games: https://venturebeat.com/2018/03/08/nint ... is-summer/

It seems Nintendo is more than happy to work with indie game developers and bring more games to the platform created using tools like GameMaker Studio.

In light of this news, is Scirra planning on implementing support for the Nintendo Switch and is it a priority?

> The idea is simple, at the end of each 12 month period of a subscription you can use the last major version without an active subscription. So if you purchase Construct 3 and then after 12 months cancel your subscription, you can continue to use it but receive no new updates.

>

This is exactly what I was talking about in my earlier reply. I would have no issue with a subscription system if I could keep using updates up to the point I stopped paying. That's essentially what Microsoft also does with Windows and I never heard anyone complain.

Speaking of which, I'm starting to think Scirra is actually going this route and they were either vague or misunderstood by everyone. I haven't really read into it much, but this just makes too much sense compared to they just want a subscription system that cuts you off entirely if you stop paying. I suspect this because of things like the editor works offline too.

I hope this is the path Tom and Ashley do take. If they don't, I'm definitely not going to buy into the subscription model at all.

Tom Ashley

The fix for the subscription issue has been solved by another company in the development space (not games development): Jetbrains. They had the same controversies moving to a subscription model and they introduced perpetual licences: https://sales.jetbrains.com/hc/en-gb/ar ... k-license-

The idea is simple, at the end of each 12 month period of a subscription you can use the last major version without an active subscription. So if you purchase Construct 3 and then after 12 months cancel your subscription, you can continue to use it but receive no new updates.

If you're planning on supporting native iOS and Android exports like Clickteam Fusion 2.5 offers, then a subscription fee is not unreasonable (at least to me). That alone would be enough of an upgrade for me. The current means of exporting to those platforms in Construct 2 pales in comparison to Clickteam Fusion.

Technically, Construct 3 is quite impressive. I see people going on about how it looks, failing to acknowledge the Scirra team are the first out of the event-based game maker tools to create a browser editor. Even more impressive is they're not using any frameworks or libraries, it's all just vanilla Javascript with CSS and HTML. A lot of work has gone into this and I think it's fantastic.

It definitely needs some work and absolutely needs some "wow" features missing from Construct 2, to warrant spending $99 a year on Construct 3. I think Scirra knows they need to beat Construct 2 in features, ease of use and other aspects. If Construct 3 offers native exporters for iOS and Android without needing to use horrible third-party solutions like Cocoon or Cordova, I am definitely sold and will day one purchase. That alone would be a worthy upgrade feature.

I am curious to know if anyone has built a tapping based game using Construct 2 before. When I say tapping game, I of course am referring to games like; Tap Titans, The Simpsons: Tapped Out, Simcity, etc. Not only tap focused, but time management games where you perform time based tasks that could take X amount of time, then provide you with a resource.

I am in the midst of building a mobile tapping like game, but curious if anyone else has any pointers or source code who have already done so.

I can gladly help you out with this. I will send you a PM now, and if nobody else already has the job, I would love to help. You are right in that it is simply more than just sending data to a remote server, you need to do it securely to prevent anyone from submitting data to your server.

At present we have HTML5 based export options for Construct 2 games. While this approach somewhat works using Crosswalk or Ludei, we all know there are certain performance limitations (especially when it comes to physics and heavy use of imagery). Some of you are probably aware Facebook have something for creating truly native iOS applications in the form of React Native.

Through the use of React Native you write applications through a nice API that has similar syntax to React.js. I had an interesting though, would it be possible to use this approach within Construct 2 and make it so we can create applications & games using React Native (or equivalent) then export a native project file to create a truly native game or application, as well as use native iOS libraries and APIs?

Just a thought I had today that I thought would be great to share.

There are very few things that Construct 2 cannot do. The only considerations are performance, but honestly I think it is possible if you have the time to do so.

The best place to get time via an AJAX request is this: http://time.jsontest.com/ - used it before and works well.