Thoughts ?
I’m of the opinion that most people wont download a plugin to play your game if your trying to promote it via facebook or something .
And I’m back all , after being away for a long time( as in no unity dev for about 2 months and I’m back at it . )
Wasn’t there just a thread about this less than a week ago?
I think that Runescape is not really comparable as it would be running though an authoritative server. (Though HTML5 has some issues in runescape for the old rich ;)) Most of our games are not and thus easy to hack AFAIK(Might give HTML5 runescape a source look when I get home but ATM I’m on holidays, It will be interesting how things like collision, shadows and reflections are done)
I find it strange that it got a big graphics update as they went to HTML5. Performance is terrible though.
its possible, as html 5 is gaining more traction, though i think its still not an official spec yet.
with flash and native client, i don’t see why they would spend the resources working on it just yet though.
If it’s like EA’s html5 game, “lord of ultima”, all the calculations are done serverside and it simply returns a view for the html5 client.
So it’s server side rendering just shown in a HTML5 page?
People keep talking about using “HTML5” for games, but is that actually the case? Aren’t they using external technologies that sit on top of HTML5 which are each their own thing, like WebGL? People talk about “HTML5” like it’s some kind of holy grail, but I’m under the impression it’s just another document format with a new wild west of techs sitting on top of it. Which is cool and all, but it makes talking about “moving to HTML5” pretty meaningless.
The html5 is basically a web standard that adds some new elements and more interactive experience where you can do what used to require flash, you can now do with javascript and html5. For the big html5 games, any input is sent to the server, calculations made, db updated, and the client just constantly polls the server for the different variables and populates them into the html front end, using ajax or some php scripts for example.
You can use plain javascript for games with 2d graphics, or with html5 it adds the canvas element but logic is still done in javascript. You can also combine a game in the canvas with normal html dom divs.
I made a little rpg a couple years ago… when I was learning javascript, spawn system, item system, magic system, buy/sell system, etc in plain javascript written in notepad:
https://www.youtube.com/watch?v=_fLODEM_0ls
Html5 is just a newer standard that gives you more control to do what you’d normally need flash to do…
Not bad.
There seems to still be “some” demand for more browser games too. I am interested in something like CthulhuNation, Urban Dead or Travian myself. Maybe even a Master of Orion style online browser game using unity.
Ok I’m going to lay down the prediction, there will be a unity to html 5 exporter for Unity 5. It will be the main feature draw.
My suspicion for this? UT is smart and nimble, they could get a stripped down Unity to run on HTML 5 if they wanted to, it’s becoming obvious this is the best route for html 5.
My second suspicion, in a forum post a while back concerning html5 like a year ago, a unity developer said that html5 just isn’t really viable because of performance and NaCl is better. The javascript performance is really the limiting factor. This to me insinuates that UT has been playing with HTML5 and WebGL already, they probably have functioning prototypes. Just waiting for javascript to become viable. Well the next Safari release claims like a 5x increase in JS performance. Browsers are now going to get into a JS performance battle. In a year we will probably see every major browser radically increase JS speed. When this happens UT will probably move towards making Unity for JS viable.
Also I haven’t seen any unity devs come into these threads and dissmiss HTML5 like they have in the past… so I suspect they are working on it.
From just a few days ago. Not a dismissal, but saying they aren’t sure if it is possible is a bit of clue that it isn’t right around the corner.
I don’t doubt that someday it may happen, but I wouldn’t bet on it for a couple of years. I don’t think that it’s so much a technical issue. You are right, the parts of HTML5 required to pull this off will only get better. But right now, compatibility is a moving target. Implementation differences, device capabilities and so on. It’s almost like there would have to be multiple exporters, at least HTML5 mobile and desktop. Hopefully those differences will become less. Also, there really isn’t a commercial demand for it right now. The money is in mobile and desktop. To justify the development costs, it would need to lead to sales (of either the exporter or pro). It will probably be viable eventually, but not for a while. They dropped flash, and that was probably less of a tech challenge (because just one target) and certainly a wider base currently. I just wouldn’t hold my breath.
I would take the 5x increase in JS performance claim with a gain of salt and would guess that the value that Apple is touting is in a very synthetic benchmark that shows a positive for them.
After the introduction of JIT compilers for JS into browsers, which was the latest major increase in JS speed, JS speed hasn’t improved in a significant way in the last 3-5 years and I don’t see a huge jump coming any time soon. The software is very well tuned at the moment and any improvements in speed usually come from improvements in hardware instead.
This article is an incredibly good read that is about why web apps are so slow on mobile devices. It talks a lot about JS and why/how it is so slow which relates very closely to this discussion. Here are some comparisons from that article in speed between Chrome from 2010 and Chrome from 2013:
The first image shows very little improvement or actually a negitive improvement in certain tests. The second benchmark shows some improvement but this is what he has to say:
There are some very important reasons why Epic had to use something like asm.js to get the Unreal running in HTML5/WebGL at a decent speed and I haven’t seen anyone but Mozilla say they will implement something like asm.js. This technology is still a long way off but there is something to say about big companies getting in there early and trying to push the technology and the adoption of that technology forward. Does it make business sense to Unity though? Only Unity and their business analysts who have undoubtedly worked out the business case for HTML5/WebGL know this.
here’s a speed comparison of C, NaCL, Asm.js, Java and some other technologies running the Box2D engine.
http://j15r.com/blog/2013/07/05/Box2d_Addendum
Asm.js runs about half the speed of C and is in the same ballpark as Java.
They test Asm.js running on Firefox and Chrome. I have also heard that Microsoft is interested in putting it into IE - it would fit well with their Typescript project.
The first post in the comments is by a familiar name 
Which suggests that Unity is definitely looking into this type of thing…
What’s so special about html5??