Why HTML5, and the future of exporters

Discussion and feedback on Construct 2

Post » Sat Feb 11, 2012 1:17 am

Hhehe :)
B
58
S
13
G
10
Posts: 632
Reputation: 12,515

Post » Sat Feb 11, 2012 2:21 am

Yeah, the browser object already has 'Request fullscreen' and 'Exit fullscreen' modes, which should work in FF10 and latest Chrome.
Scirra Founder
B
359
S
214
G
72
Posts: 22,952
Reputation: 178,630

Post » Sat Feb 11, 2012 2:54 am

Yes, it surely is easier to click a URL and play the game while you are online. Also, the security warning argument, I never find it as an excuse whatsoever. How hard is it to click on yes or no, keep or discard? And if you know the game you are downloading, how can anybody be scared of warnings? Warnings are there to make sure you know that you are downloading an EXE that could contain a virus, if EXEs were so evil, then I guess most of us wouldn't even on the forums discussing how evil EXEs are... Oh wait... I wouldn't be able to use C2 either or any of the very best game making tools. And don't you need an EXE (or any other native format) to run an HTML5 game/app. anyway?

Nothing is fully fool proof and I realize there's lots of those around the net, but that's not yer, browsers devs., game devs. or my problem. Is it?

My drivers are ALL up to date, most drivers aren't updated anymore though, so there's no further updates possible. I'm sure this is the case with most people around the world as well.

I never intended my words to be taken into the context of a heated argument. I was just trying to point out the problems of HTML5 now and who knows if they will ever be fixed. It is clear that at the current state, HTML5 isn't older hardware friendly, and that's quite tacky. HMTL5 performance issues might get fixed on new and powerful hardware, what about the other guys? Will offline play be easier?. We can blame the card makers and such for some of this issues, but let's be clear on something, native formats do not have a problem running on older hardware. I have ran native games on hardware that was less than satisfactory for the game requirements, still a very decent speed. Will HTML5 accomplish such a thing on older computers? Somehow I really doubt this. But, Who knows? And even if it did, how long would it take it? 5 or 10 years?

Also, I never said anything about making exporters now, I never said that at all. In fact, I said I understood yer position to want to finish something first before starting anything. I however, did say that having a back up plan, is a good idea, and I still think that and won't change.

And the whole running in a browser thing, it seems backwards, for reasons I already stated.

More dependencies = more convenience? Better technology?

Only when HMTL5 can rid of ALL its shortcomings it will be able to stand up against native formats. Nobody can say what the future holds for HTML5, but it surely cannot stay like this.

There's nothing I want right more, than to be proven wrong. Sadly, it is all speculation.VampyricalCurse2012-02-11 02:55:29
B
47
S
10
G
6
Posts: 455
Reputation: 8,326

Post » Sat Feb 11, 2012 4:07 am

[quote=Ashley]When we started the HTML5 exporter no browsers hardware-accelerated it so it was really slow everywhere. Now, with WebGL, it absolutely flies, it's just mobile is a little behind desktop.[/quote]

In page 2 of this very discussion.

And what Ashley means by "when we started" is about a year ago (give or take a few months).
So your "5-10 years" range is quite pessimistic imo.
Give it a couple of years to have became more accepted as "selling-hits" games will have appeared.

Also in 5-10 years you intend to still have your actual old gear as main computer ?
Let me hope that by then you'll hopefully have upgraded for better hardware, with better OS all designed in a world where HTML5 is a reality.

In the end it all falls back to this idea that "games in browsers aren't true games", or that "casual" games are inferior to "exe games".

Let's hope that most of the entries to the rotary comp will tend to prove you wrong.
New to Construct ? Where to start

Image Image
Image Image

Please attach a capx to any help request or bug report !
Moderator
B
247
S
85
G
40
Posts: 7,000
Reputation: 57,805

Post » Sat Feb 11, 2012 5:48 am

I can for sure upgrade, however I wouldn't do it just for HTML5, that's crazy. But this is not the case for most people around the world, is it? This is not about my own isolated issue, it goes beyond.

I also understand this is relatively new tech. and it means it has ways to go, but some of these issues I just don't see them resolved easily (or ever). Performance on new powerful machines will increase and this is obvious. How much will it increase a year from now? Who knows, but I sure hope it's enough to run things at 60FPS. What will the increase on older software be? Probably, still unplayable. My 5 years mark might seem pessimistic for some stuff, but for others, it really isn't.

Yes, EXE stands superior to HTML5 now and as long as it supports older machines, it will stay that way. This goes for any native formats, not just EXE. The topic of casual gaming vs gaming is a different animal and it wasn't my point or mentality at all. This is about the future of gaming in general, and as I see it right now, it is going backwards in performance and dependency in exchange for ease of use an cross-platforming. I just don't think it should be this way. I am sure it is not only myself who does not like to depend on even more people ( who by the way, cannot seem to be able to agree on anything when it comes to standards ) to have my games or apps. run well.

HTML5 needs independence from browsers, a huge performance boost, compatibility, hassle free offline gaming, game-pad support ( I have already seen this, but this just started ), much better full screen ( also know this is being worked on ), audio needs to be worked on badly. How long does this seem it's going to take?
B
47
S
10
G
6
Posts: 455
Reputation: 8,326

Post » Sat Feb 11, 2012 10:45 am

[QUOTE=VampyricalCurse]HTML5 needs... gaming, game-pad support ( I have already seen this, but this just started ), much better full screen ( also know this is being worked on )[/quote]
it sounds like most of the problems you have are indeed being worked on, and this is just what you've heard of. you're not the only one who thinks the platform is incomplete. html5 isn't even done yet. I would say everything's moving amazingly quickly, as everything always does in the computer world. I mean, I didn't even hear about webgl until fairly recently, and now it's already supported on the chrome beta for android. adobe abandoned flash because of html5 and js(there are flash games on steam). There are plenty of big players who are looking to make it faster, better, etc.

as far as very old machines? even construct classic requires dx9, and though there were a few people complaining about that, almost anyone can run a construct classic game. the best supporters of older machines in the industry as far as AAA titles go I'd say would be valve. and you'll notice some of their newer games, like portal 2 and left for dead 2(which was released a little over 2 years ago) don't run on older machines, at least not at any playable framerate. valve is known for their games having reasonable system requirements.

[quote]How long does this seem it's going to take?[/QUOTE]
I don't think any of us know, probably a little while. my own personal pessimistic prediction is that within two years, some combination of these web technologies will be comparable to flash speeds, but I think 1 year is closer to the mark. also, I can't speak for scirra, but I'd think within 5 years, work on the next exporter would be begun, or the edk, or the web platform will be approaching the needed maturity.   if you need exe's support now, honestly, use construct classic. It's got your working audio, game pad support, full screen, exe, etc. especially if you're planning to develop a game so quickly that you're concerned about waiting a year.   it's already been stated that native exporters and/or the edk are in the cards. C2, js, and html5 are being worked on hard by people that are probably much more diligent than ourselves. it's not going to arrive faster by being upset it's not here yet. Honestly, cc is probably your best bet, or if you want to sacrifice alot of ease of use for more crossplatform compatibility, Unity is your best bet.lucid2012-02-11 10:51:45
Spriter Dev
B
88
S
21
G
12
Posts: 3,240
Reputation: 16,486

Post » Sat Feb 11, 2012 11:16 am

[QUOTE=eyeliner] [QUOTE=lucid] ............... [/QUOTE]
If there was a possibility of exporting a capx to Classic, there would be no problem and all problems would cease to exist.[/QUOTE]
oh yeah, forgot to respond to this. classic already exists, and it has the exact same type of event system as c2, it would make no sense to do this. C2 is a vastly superior, albeit less complete(at this time) product. just messing around in the ide, everything works much more smoothly, and c2 improves on everything that was wrong with CC. Things like the file format, ide responsiveness, and ide stability, and greater extendability.

However, from your statement, and many others like it, I think Construct Classic's brokenness is hugely overestimated in the C2 community(the part that doesn't also use CC). Obviously it had enough problems at it's core that Ashley thought it better to begin anew, rather than continue to iterate a flawed codebase. This is something to keep in mind, because it does have it's quirks, and bugs. Transitions don't work well(not sure if C2 has transitions), so you don't use transitions. Sometimes the IDE starts up with a blank layout, and you have to doubleclick your event sheet in the properties pane to bring it back into view before you begin. You have to save frequently, because the IDE is known to crash now and then.

However, that being said, it's not so broken it's unusable, and the runtime(which is what really matters in the end) is stable. You can make games with it, that work on everyone's PC, run ridiculously fast(seriously, it's fast), and look as good as is possible in 2d on a pc.   I personally have some plans for webbased projects in the future, and Construct Classic, even in it's quirky state gives me enough confidence in Scirra that I have no doubt c2 is the best way to make web games(and eventually - any games). But right now, I'm working on projects that require that exe speed, so I use construct classic.

The skills for both are interchangeable, and we're one big happy community, I really don't see why the "when will there be an exe?" crowd don't just buy the C2 license now, so they have it when they need it, and to support scirra since Construct Classic is free, and use CC in the meantime.lucid2012-02-11 11:19:06
Spriter Dev
B
88
S
21
G
12
Posts: 3,240
Reputation: 16,486

Post » Sat Feb 11, 2012 11:37 am

Well, no, HTML5 is not by any means my preferred platform, but I do not need EXE right now. My game will be fine on HTML5 from what I gather. May have to sacrifice a little eye candy ( for example particles which for some reason it's eating a lot of resources ) for it but it will be fine. Sadly I do not think it will be playable on older machines. I have machines to test this on. These machines are from around 2004 I believe.

Like you I have high hopes also, that C2 becomes a multi-platform exporter, 5 years it's a long time though and I would keep an eye out for competitors. I think a good move for Scirra is to release the EDK first, this way they can let users work on the exporters and they can worry about what they think it's more important.

I was told months ago to stick to C2 and that is what I am doing and I never doubt C2 at all, in fact, I think it is the best there is right now, I just dislike what it exports to. Let's hope in a year some light can be shed on HTML5 issues.
B
47
S
10
G
6
Posts: 455
Reputation: 8,326

Post » Sat Feb 11, 2012 12:51 pm

[QUOTE=VampyricalCurse]I think a good move for Scirra is to release the EDK first, this way they can let users work on the exporters and they can worry about what they think it's more important.[/QUOTE] I'm not sure how close what Ashley's currently working on is to an EDK. For all we know, writing the edk may be as much work as writing an exporter. But if it is more like the combined time it takes to add 2 or 3 major features like local variables, grouping, and families - then I don't about whether it's a good move for scirra, but personally that'd be the feature I'd most look forward to.

Ashley is probably weighing the potential userbase for the EDK into how much of a priority it is. I know of at least a few people in the forum who are experienced enough with coding they could make a runtime, and there's probably well over 200,000 construct downloads between classic and 2, so there's bound to be many more out there, but there aren't many users who've mentioned in the forum they'd like to use the edk, just the desire for the runtimes people would make with it.

I personally would like to start such a project with android, and hopefully recruit a team. I've worked with android development before, and I enjoyed working with it, and it'd be an awesome platform to be able to make native apps for with c2. I'd be curious to know if there was anyone else who would want to contribute to such a project?lucid2012-02-11 13:27:59
Spriter Dev
B
88
S
21
G
12
Posts: 3,240
Reputation: 16,486

Post » Sat Feb 11, 2012 1:56 pm

I think I have heard a few commentaries here and there about people who would like to code with the EDK. I think Kiyoshi was one them I think?

Anyway, I'm sure the EDK is no light work at all, but releasing that instead of making an exporter might prove more effective for Scirra. And once it is out, more likely than not, people will start to turn up. I imagine quite a bit of people are following C2 from the outside.
B
47
S
10
G
6
Posts: 455
Reputation: 8,326

PreviousNext

Return to Construct 2 General

Who is online

Users browsing this forum: Yahoo [Bot] and 7 guests