Time to move forward?

> I think all this slagging off of C2 and in the process promoting a competing product on Scirra's own homepage is pretty tasteless frankly.

>

Not at all, these are serious concerns.

People should better understand the limitations of C2, despite it being a great engine for intuitive coding.

Critique is fine, it's just when it spins into hyperbole it projects a skewed image of the whole situation, which I feel really isn't helping.

But yeah I was being an ass about it. Sorry 'bout that neverk and others.

—

You have really good points.

I haven't compared Construct to bikes, but I could compare it to a car that looks like a sports car, but has an engine from Toyota Prius. The Prius engine is very modern and engineered very well, but it's not meant for racing, it's meant for commuting. It's really hard to win a racing competition with that kind of a car, even though it might look like a sport car. You have to be very clever to compete against real sports cars.

HTML5 isn't primarily meant for games, but that's what we have under the hood.

I wish C2 had something underneath that is meant for games and HTML5 could be just one export option. Now the thing is upsides down. We have something that isn't meant for games and we turn it into games. It should be other way round.

ummm... people say html5 engine, when i thought it was javascript engine

(html 5 is only one line of code, a placeholder for all the javascript that runs the engine and stuff (c2runtime.js + data.js = gameengine + stuff (~10k or more lines), index.html ~10 lines )

i'd love to see if there are ways to optimize the js code, but noone has the time or will to try/test it

saiyadjin the technology is called HTML5. You code it's functionality with JS.

how is something called html5 but then JS is used to code it. that doesn't make sense.

html is just a markup language which utilizes JS, but since the difference in amount of really used code, i'd rather call it a javascript engine, then html engine.

Because JS is just a text which is read/parsed by the browser or other interpreter to execute commands. JS is not a language which gets compiled in the end - it's being interpreted. It's just a remote for the technology.

There is a new standard for the markup language as well and it's cool. It's more semantic than XHTML 1.0 strict or any other previous version... but there is also a canvas and other features which make HTML5 a technology.

BTW with JS you can do more then web control.

how is something called html5 but then JS is used to code it. that doesn't make sense.

html is just a markup language which utilizes JS, but since the difference in amount of really used code, i'd rather call it a javascript engine, then html engine.

HTML5 is widely considered to be the combination of HTML, CSS and JS. The actual HTML5 is simply referred to as HTML, in the same way that no one ever said "I know how HTML4".

HTML5 is widely considered to be the combination of HTML, CSS and JS.

Indeed.

— Nice and objective post you wrote there. +1

I urge people to go and try out other engines as well. As I (and others) have said before: A good developer knows more then one engine. However, learning to script in new languages, despite that the overall computational thinking required to make software is the same, is not suitable for all. There are a lot developers (myself included) who develop in their spare time alongside their work. So for us the speed and ease the tools and engines provide to make a software is a primary concern. Construct 2's flexible and very capable event sheet system is the best one I've seen. The 2nd would be the blueprint system in Unreal Engine 4.

I must echo what zenox98 said - we don't mind discussions on the forums about other tools, or even any perceived shortcomings of C2, but it cannot resort to personal attacks or aggressive language.

FWIW, I'm pretty sure the issue of the wrong canvas size in Safari is also an iOS/Safari bug, and similarly I've reported it to Apple and put in a workaround for the next build. However these workarounds really are hacks - to give you an idea how flakey they are, another issue with the iPad Pro actually ended up being a years-old performance hack for when the iPad 3 first came out - exactly the same kind of issue where it's a problem with iOS or Safari but we find a way to circumvent it - which now was causing a problem on newer hardware. Hopefully this demonstrates my scepticism of our own workarounds for bugs in other software - when the problem is not with Construct 2, the best we can do is paper over the cracks.

Other than that, I've seen long rambly threads like this crop up from time to time over the years running Scirra: someone is annoyed by some various issue (which in this case I am pretty sure is ultimately not our fault), then everyone piles in with their concerns and inevitably the spectre of other products is raised. We know Construct 2 isn't perfect, and nothing ever will be, but we do work super hard to make things work and make the best of the technology we use. For us the main point of Construct 2 is the event system, so remember if you're looking at programming-based tools it's really a whole different kind of product in my opinion.

Also FWIW, Classic had some cool advanced features like mesh distortion, and WebGL can do that but not canvas2d. This is the main reason we've been holding off more advanced rendering features. According to webglstats WebGL is closing in on ubiquity. To give you an idea how far things have come, when C2 first launched, WebGL did not exist. Only Chrome and Firefox ran the games at all, and with a software-rendered canvas2d that could only do 30 FPS on a high-end gaming machine. We got a lot of flak for not choosing Flash, and there were a bunch of threads like this over it The fact we're now talking about ubiquitous WebGL is pretty amazing. I think 8-10% of users still don't get WebGL (higher on certain OSs/browsers). Inevitably, it will reach some threshold, I don't know where to set that but I guess 98% is a reasonable one, where we can just entirely drop the canvas2d renderer and go full steam ahead on WebGL with more advanced features including all the stuff our old native-based engine did. So no shortcoming for being HTML5 there, especially with WebGL 2 coming. Also I must throw in yet again: WebGL gives us the same access to the GPU as a native app, and many apps are bottlenecked on GPU rendering, so no loss for being a web-based engine in that regard either.

Construct 3 also changes a lot. Not so much in the runtime, at least at first, but the editor is being completely rebuilt from the ground up.

Ashley

Is the architecture for other exporters still in place?

Ashley I think that people need these kind of posts from time to time. To keep them updated with a little fresh information, ideas and to ease the tempers. What you said sounds promising. I personally looking forward to it. Just don't overwork yourself, have some rest sometimes.

Ashley, glerikud these kind o posts is what we need plus some screenshots or a quick video with some of the changes which will keep us think more positive. anyway i dont like this topic i respect the thoughts of anyone for c2 either positive or negative but i dont find what a topic like this could help the topic starter?if you feel that you are at a dead end with c2 posting this nobody will change your experiences with c2. Or again its an opportunity to throw our problems and thoughts for c2. which i respect but we can do it without a new post. there are many of them already

Many of these bugs isn't the fault of C2 , I've said as much many times.

It is often the fault of Apple, they break things when they do iOS updates.

It was the fault of Google, who broke Chromium v-sync. It was also their fault for introducing memory leaks into Chromium.

The problem is because C2 is a HTML5/JS engine, it relies on these tech and so, despite not being the fault of C2, we ultimately suffer due to these issues.

I would kill for a C3 that exports to native, bypassing these browsers.

Just 3 export options most widely used: DirectX for PC, Java for Android, Objective C for iOS. No need to go nuts with all the platforms. Go where the audience are, where the money is at.

So I guess until someone makes such an engine with the intuitive event system that exports native, C2/C3 is still the best we got for 2D.

So I guess until someone makes such an engine with the intuitive event system that exports native, C2/C3 is still the best we got for 2D.

I'm keeping an eye out for Clickteam's Fusion 3, for this reason.

Lets hope they have a forum where people can either ask for help, offer support, or tell everyone that they are leaving.