Ashley's Forum Posts

  • The C2 editor cannot run on the Raspberry Pi - Windows 10 IoT is a very basic stripped-down version of Windows which only supports UWP apps, not traditional desktop Win32 apps like C2 is, so it can't be made to run. Also RPi uses ARM chips and we currently only build C2 for x86/x64. It's similar to why C2 can't run on Windows Phone.

  • This is really weird because firstly there is no technical reason CPU usage should affect touch input at all, secondly there really ought to be nothing special at all about the 50% mark, and thirdly because C2's measurement of CPU is a *very* approximate estimate based on timers instead of any real CPU measurement so is probably only accurate to +/-10% at best. So it's unlikely to actually be at 50% when it says it is. I would think it's likely something else is the cause and you've falsely correlated it with CPU.

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads

  • No, sorry, we can't give an ETA on any features.

  • Closing, please see the bug report guidelines and only post in English.

  • We have to balance requests like this with literally hundreds of other requests (many of them "just little things you could do easily, right?" - usually a big underestimation ), plus Construct 3 development, plus regular bug fixes and maintenance on Construct 2, plus all the other details of the day-to-day running of Scirra. Maybe we could do it, but you could end up waiting a long time. On the other hand if Kongregate were developing their own plugin they could update it right away. So I think it would work better for everyone that way.

  • It turns out the Xbox Live stuff is kept behind NDA by Microsoft. We're working on getting access to it, this is the only reason for the holdup.

  • The point of the blog I linked to is these numbers don't necessarily mean anything.

  • These measurements are surprisingly difficult to interpret correctly. See my blog on understanding CPU and memory measurements. Here's a relevant passage:

    [quote:2gx0cnl0]Javascript does not provide direct access to the CPU usage. Instead Construct 2 estimates it using timers. The key word here is estimates: it is not a real CPU usage measurement.

  • I would much prefer for Kongregate to develop and publish their own C2 plugin, since there are hundreds of third-party services out there, it's a constant burden trying to maintain them, and it's not our API/service (they obviously know their own platform better). Our plugin was developed a few years ago when we were just trying to get off the ground and offer some extra options to developers. I'd encourage you to go to Kongregate about further support.

  • I wouldn't bother with throttling - the multiplayer engine's approach is to assume if the data rate is too high the network will drop packets, and it can compensate pretty well for packet loss by interpolating between the packets it does receive. However to achieve that it has to impose an artificial delay on the incoming stream, so it has something to interpolate to. All in all it's a pretty complex system - it would probably be better for us to fix that bug...

  • It shouldn't matter that angle() returns something like -90 instead of 270. If your math correctly takes in to account the cyclical nature of rotation, they are identical. Usually people get this wrong because they treat angles as linear numbers, e.g. trying to use lerp between angles, where lerping 50% of the way from 30 to 350 degrees will go the long way round to 190 degrees instead of the shorter way round to 10 degrees. In other words, the normal lerp expression doesn't take in to account angles wrap back around. (This is why the anglelerp expression exists - it uses different math to solve this problem.)

    While solutions using things like (x + 360) % 360 can work for some cases, that does not usually solve wrapping problems, since it's still linear math.

  • You can ignore it. It is trying to use a service worker, which only works on HTTPS, not HTTP. There are lots of reasons to use HTTPS though.

  • if you load up that link on an iOS device you'll see that the 'NEW' button can open up a new window just fine but the sprite 'NEW' trigger does nothing, is this a bug in the way the sprite works as a trigger?

    It depends on precisely which events you used, but I can't see which you used without the .capx.

  • C2 now defaults physics to asm.js mode, which in theory is about as fast as a native physics engine.

  • No. Event sheets are just for organisation and the number you use doesn't affect performance. In general you can check things like this yourself, although often it is a waste of time.