HTML5 Canvas as it stands now

[Jackolantern]

(I was going to post this in my Flash vs. HTML5 thread, but it became so long and really stood as its own topic)

This (my HTML5 Canvas endeavors) is not looking good so far. Canvas is sooo primitive, I can't imagine making the kind of games I want to make with it. Starting from scratch with nothing makes it look like I could get lost in trying to make my own engine for years. Animation is painful, although thorough. I am starting to understand why pretty much all you can find canvas-wise out there are tech demos. And I am also starting to see why the only serious game engine completed up to this point in HTML5 was made by John Resig, the author of jQuery and all-around Javascript super-genius, and then was immediately bought and made proprietary by Zynga, the creators of Farmville. HTML5 Canvas is so primitive I think they were worried it would be years until they saw an engine that complex again.

Not all is bad. There is a commercial game engine that has a couple of nice NES+ looking demos on it called ImpactJS. There is another engine out there, called Rocket Engine, but it looks like it may be quite expensive since it doesn't volunteer how much it costs and says it is "for professional developers". The site also does not really lend a ton of confidence considering that it is basically a single-page brochureware site with no forums and really no community services at all. However, probably the most promising engine right now is Isogenic Engine, an engine I am interested in knowing more about. However, it is not currently complete, is only selling its "premium" license to be part of the beta (which, at 65% off, is still $1000), and is under heavy NDA. If it makes it to 1.0 gold release, there will be a $200 basic license that would be more in my realm of purchasing. I say "if it makes it to 1.0" because they say on their site that the beta testers are helping to pay for development, which is kind of scary from an investment position. That same model was one of the reasons why RC got so bogged down and took years to get to a nearly usable state.

All in all, there are a few things I would love to see in the coming year for HTML5 Canvas-based game development:

1. More engines released in varying price ranges.
2. More frameworks released that handle the low-level plumbing of functions that are one-command in Flash (there isn't even a simple command to draw a circle; you have to make an arc and feed it a pi-based formula).
3. More hardware-accelerated Canvas support in browsers, and more people using them (animation or sprite-heavy games really start chugging along, while Flash just blasts straight by).
4. A Flash-like editor and IDE would be great. There are a couple of promising up-starts in the open-source and commercial spheres, but they are light-years away from the functionality of Flash. It would be nice if they could at least get in the same galaxy. All eyes have been on Adobe for some time on this one, but that won't happen until Flash starts going down, which doesn't look to be happening for at least a couple more years.
5. More books covering more complex game examples. I say books because most engine-less, complex games are likely to be far too long and involved for most free online tutorials. Skilled people want to be compensated for writing hundreds of pages of instructions, and that is understandable. There is a book I am looking forward to coming out in October called "Learning HTML5 Game Programming", which appears to be the first 100% game-related HTML5 book out there. While some write their books with a game slant, this will be the first total game-related book as far as I can tell.

I don't think this has been an unfair critique of the current state of HTML5 Canvas development. I want Canvas. I want open standards and total adoption of it. However, it isn't there yet, and may not be there for a while. But in almost every negative I mention, I have tried to show how things further up the pipe could drastically improve the landscape.

The future looks bright for HTML5, but I think we are still a few minutes before dawn for most game developers.

[dreamer]

I have been looking into HTML5 game programming for a while now and what I can see is that most of the engines support Canvas as a backup and DOM as a primary such as Crafty

Rocket Engine seems like it is a no-go since Disney purchased the rights from the developers stated on their own feedback website.

They seem to be aiming Isogenic Engine toward companies rather than indie developers from the look of things.

I do have my eye on a 3D web engine Copperlicht. The engine itself is free, the world building software which they state is completely optional costs 99 EUR (143 Dollars).

[Jackolantern]

Isogenic Engine is actually not that badly priced. 200 - 1000 GBP, depending on the license. Although I don't know for sure, I actually think Rocket Engine was the one aimed at professional studios. The fact that they have been sold and the engine is becoming proprietary would make sense from the hands-off, info-only website they give. It almost struck me as more of an info page that the engine exists rather than trying to sell a product.

[hallsofvallhalla]

I use the Javascript Game Library and love it. http://www.sean.co.uk/a/webdesign/javas ... melib.shtm

There are very few examples and the documentation is ok but the library works quite well.

As far as HTML5 I am going to wait. I have to many other things I am trying to do first. I am looking to switch careers from Dialer Telecommunications to a Programmer. Problem is a I need C#.net skills. I do have some and have toyed with XNA and others but nowhere near what I need to do it for a living so while want to seriously learn HTML5 it will be some time.

[dreamer]

Javascript with HTML5 will only get better with time and I really believe that we will start seeing more and more browser based games being released and client based games going to browser based. The company Frogster has announced that they are bringing Runes of Magic to the browser.

Our stories are almost alike Halls. Let me guess, you work in a call center/tech center and would love nothing more to get out to go to something else like me? I am trying to make the switch from Technical Analyst to Web Developer.

[Jackolantern]

I definitely agree, Dreamer. HTML5 is on the rise! I just don't feel that I have the stomach to be an early adopter; at least not at this point in my programming career. I want a jQuery that is updated from the ground up to work with HTML5, pretty editors that do more than just know the new tag names, better engines, etc. Basically after looking at those wants, I realized I wanted it and the community to be more mature. And there is nothing I can do for that but wait.

As a side note, I have no idea how Frogster plans on making a plugin-less Runes of Magic. WebGL is nowhere near the point of making 3D games (all the demos look closer to the 3D capabilities of the SNES, or maaaaybe the Saturn). This is of course due to the very low amount of resources the browser is given to run it. Most of the major developers are really pushing for the option to ask the browser user to OK temporary origin-based increases in resources so games could be run.

[dreamer]

Waiting and learning seem to be the only option at this point. I was looking at youtube late last night and noticed that many developers are developing SNES type games with HTML5/Javascript which do look really impressive, but unfortunately no full-on plugin-less 3D.

As far as Frogster, I am seeing that companies are still utilizing the tried and true method of flash-based engine for their games, or what I have seen for games like Dofus and Freerealms where the player has to download a plugin/mini-client that launches from the browser.

[hallsofvallhalla]

There is no need for HTML5 for SNES style games. My top down shooter and RPG engine proves that as well as hundreds of javascript games around the net. I am sure HTML5 will make it much better and easier to get it done however.

[Jackolantern]

Even as HTML5 stands now, Canvas does give better performance than old-style JS hacks for games. We don't need to wait for that.. HTML5 also allows for background music and sound effects, something <= XHTML cannot offer without a plugin.

However, this is my bottomline advice to people:

If you want to play around with HTML5, go for it. It can be a lot of fun. If you have a planned game you want completed, you should wait for at least a year for some more of the bullet-points I listed above to be hit and delivered.

[hallsofvallhalla]

javascript can do sounds effects and music though.