Lots of doors are being closed for Java

[kevglass]

FWIW I'm revisiting XMLVM after a year or so and it's improved greatly. Performance isn't terrible any more and much more of the JDK is available.

I'm going to sort out some performance tests if any one has any ideas for a suitable targets?

Kev

[DzzD]

FWIW I'm revisiting XMLVM after a year or so and it's improved greatly. Performance isn't terrible any more and much more of the JDK is available.

I'm going to sort out some performance tests if any one has any ideas for a suitable targets?

Kev

could you try this one plz http://demo.dzzd.net/SIMCA1000/ (soft one), I would love to see if it works

[longino]

One of the problems is that there are whole exosystems around existing languages for Triple A game development; libraries, programmers, engines, support, etc. Walk into a games store and most of the top 10 games for PC and consoles will be built with only a handful of engines and libraries. Microsoft has started building similar for C#; deployment to XBox, XNA framework, various communities, indie gamers channel and more. However for Java none of it really exists. There are some Java game forum sections dotted around the internet, some home made libraries and that's about it!

This is what I was saying. I have seen Java pushed only by amateurs with a few exceptions. I don't know about J2ME.

I don't think any corportation will solve this problem. If Java is to be used for gaming then the developers themselves must do it.

Games published by our own members! Check 'em out!

[BatKid]

This is what I was saying. I have seen Java pushed only by amateurs with a few exceptions. I don't know about J2ME.

Well with android, this could be changing.  The causal mobile games market is huge, so this could be good news for Java.  The way I see it, android is the last hope to see Java deployed on the client-side.  Hopefully Oracle would work out a deal with Google soon.

[kevglass]

Dzzd, could you make the source available. I'll have to write an integration wrapper to make it render to iPhone. Though tbh, it'd be a better test if it used OpenGL

Kev

[zammbi]

This is what I was saying. I have seen Java pushed only by amateurs with a few exceptions. I don't know about J2ME.

There is c&c red alert game for J2ME.

[DzzD]

Dzzd, could you make the source available.

ok, I ll put them on the server ASAP (just need to figure out where are the lastest source code )

Though tbh, it'd be a better test if it used OpenGL

maybe in a second time ? the OpenGL version use only one small class ( here is its source  ) to access OpenGL, I suppose this class need to be translate ? does it look hard to do ? (in Hardware mode this is the only 3DzzD class that perform OpenGL call, I am very interresting in your tests results and what it will potentialy enable for the futur if it work with OpenGL )


EDIT :  re-reading the opengl source I see that I let some old shaders tests in it, dont bother with them as they are not used

[BatKid]

Dzzd, could you make the source available. I'll have to write an integration wrapper to make it render to iPhone. Though tbh, it'd be a better test if it used OpenGL

Kev

Kev, can xmlvm cross-compile lwjgl calls?  Or maybe opengl es?  I have a small android library in pure opengl es, maybe that would be suitable?

[kevglass]

I xmlvm build some opengl calls from android already, so it's definitely possible. I'll take what I can get, just really want to see what sort of performance issue there is. For my 2D stuff I don't see any issues but then it's only a rogue like.

Kev

[kevglass]

DzzD, the class looks nice and simple. Let me know when you've got some source I can much about with.

Kev

Games published by our own members! Check 'em out!

[gouessej]

Well with android, this could be changing.  The causal mobile games market is huge, so this could be good news for Java. 

No it isn't as Gameloft and some other companies prefer using the Android NDK to write their games in C++

The way I see it, android is the last hope to see Java deployed on the client-side.  Hopefully Oracle would work out a deal with Google soon.

Android is the most serious successor of J2ME but don't forget that lots of desktop computers already have Java.

[JL235]

Oracle have also talked about getting the JVM running via pure JavaScript, which could make client-side development of Java more desirable.

[gouessej]

Oracle have also talked about getting the JVM running via pure JavaScript, which could make client-side development of Java more desirable.

As far as I know, that is only a rumor, such a JVM would not be as fast as the current J2SE JVM and the client-side development is not only about applets and browsers. I want to develop games outside the browsers and I'm not alone in this case.

[kappa]

Oracle have also talked about getting the JVM running via pure JavaScript, which could make client-side development of Java more desirable.

As far as I know, that is only a rumor

Its not a rumour, it was confirmed by Adam Messinger, VP of Java Development at Oracle.

Interview can be found here (MP3 at 21:18) or transcript here.

[BatKid]

Its not a rumour, it was confirmed by Adam Messinger, VP of Java Development at Oracle.

Interview can be found here (MP3 at 21:18) or transcript here.

Pretty cool idea!  The implication is that we have a compiled language running on top of an interpreted language running inside of web browsers... my mind is spinning just thinking about it.  I just don't see this working in the short term (3-5 years).  Although this could be a very good strategic direction in the long term.

[JL235]

Its not a rumour, it was confirmed by Adam Messinger, VP of Java Development at Oracle.

Interview can be found here (MP3 at 21:18) or transcript here.

Pretty cool idea!  The implication is that we have a compiled language running on top of an interpreted language running inside of web browsers... my mind is spinning just thinking about it.  I just don't see this working in the short term (3-5 years).  Although this could be a very good strategic direction in the long term.

JavaScript is very soon to be JIT compiled in all browsers, not interpreted.

[princec]

It'll take a few years before it's all faster. Even then it's still roughly an order of magnitude slower than Java which is a bit of a pain.

Cas

[JL235]

It'll take a few years before it's all faster. Even then it's still roughly an order of magnitude slower than Java which is a bit of a pain.

Cas

MS just released the release candidate of IE 9, and say there are very little reasons not to release the final version very soon. FireFox looks like it'll be delayed to March.

[princec]

Ah you must live in a world of early adopters and risk taking geeks.
The rest of the world is about 3-4 years behind you I'm afraid. I only just started using Java 5 about a month ago because of the number of OSX users still on JDK1.4 - now at an acceptably low 10% or so, so I can live without their money.

Cas

[gouessej]

Its not a rumour, it was confirmed by Adam Messinger, VP of Java Development at Oracle.

Interview can be found here (MP3 at 21:18) or transcript here.

He speaks about 2 ways (JavaFX & running Java without JVM) and sorry there is no plan concerning the second solution whereas JavaFX 2.0 has a real roadmap: http://javafx.com/roadmap/
However, I know the Orto project.

I find this about Java running in JavaScript:

[Roquen]

Running Java in JavaScript.  Well the first question comes to mind is: WFT! Why?  I mean if it was an student or just for fun project..okay.  If it was Google...okay maybe it would be good hype for V8, but for Oracle?  What problem is it suppose to solve?  Seems like a total waste of resources.  Tossing them at getting Zero/Shark usable and the dynamic language extensions.

[JL235]

Running Java in JavaScript.  Well the first question comes to mind is: WFT! Why?  I mean if it was an student or just for fun project..okay.  If it was Google...okay maybe it would be good hype for V8, but for Oracle?  What problem is it suppose to solve?  Seems like a total waste of resources.  Tossing them at getting Zero/Shark usable and the dynamic language extensions.

It's better and more widely supported, whilst Java's future is bleak. Look at Java penetration, still only just over 60% for Java 6. Look at Apple maybe pulling Java from future versions of MacOS. Look at ChromeOS missing Java.

Ah you must live in a world of early adopters and risk taking geeks.
The rest of the world is about 3-4 years behind you I'm afraid. I only just started using Java 5 about a month ago because of the number of OSX users still on JDK1.4 - now at an acceptably low 10% or so, so I can live without their money.

I work freelance on some sites that receive a lot of traffic per month. On some of these a higher percentage of FireFox users run FireFox 3.6 or newer then have Java installed (regardless of version). On some FireFox 4 is already the second most popular version of FireFox, and it's not even out yet! Almost all Chrome users run Chrome 8 or higher, which is only 3 months old, and a quarter of them run Chrome 9 which is only 15 days old! Why? Because FireFox and Chrome upgrade automatically (and do a good job of it). Once IE 9 is shipped via Windows update, it'll receive widespread usage too within a month.

Your simply wrong. For home users, the '3-4 years behind' rule just doesn't apply to browsers.

[princec]

No, I think I'm right. Some places are worse than others. The entire UK government for example is still forced to use IE6 and they can't even change. That's millions of desktops right there.

Cas

[JL235]

No, I think I'm right. Some places are worse than others. The entire UK government for example is still forced to use IE6 and they can't even change. That's millions of desktops right there.

Cas

As I said, home users.

[DzzD]

Your simply wrong. For home users, the '3-4 years behind' rule just doesn't apply to browsers.

No, I think I'm right. Some places are worse than others. The entire UK government for example is still forced to use IE6 and they can't even change. That's millions of desktops right there.

Cas

As I said, home users.

same in france and some big society I ve worked for, problem is that in such society it is very difficult to migrate a software and especially browser because there are a lot of dependencies on other software and it cost a lot of money each time a software is migrated in a big society, both related to technical probleme but also because of formation and/or user resistance, lot of user have very low knowledge in computer because they only use it for similar task every days, and they get lost when you change them something they have used for years.

dont know how to explain that but I guess that Chrome is used a lot for personal computer (and/or Geek) that's why it is updated so frequently you cant imagine how much money it can cost on non-computer enterprise to migrate software.

Here is an example, imagine an enterprise with 2000 employes using one or more web based application daily (Intranet), let say one day they decided to migrate and next day they found that the application they need to work crash because of something have changed in the browser (this is not forcibly related to a bug, this can even be due to the correction of a bug), until they managed to fix the web application problem their enterprise is freezed

[JL235]

I can equally give stories of big enterprise companies that do constantly update their software regularly (yes they do exist); but it's all just hear-say, speculation and opinion. The analytics I've been looking at simply disagree. They say the majority of users do upgrade their browsers within a matter of months.

[princec]

http://www.w3schools.com/browsers/browsers_explorer.asp

Cas

[DzzD]

....I can equally give stories of big enterprise companies that do constantly update their software regularly (yes they do exist)...

this is not something very fun to upgrade a software in a big society, not something you do just because you want to be in the fashion, what kind of society ?  computer related maybe ?

The analytics I've been looking at simply disagree

what kind of statistic sources ? becarefull lot of (nearly all) statistics are not objective

[DzzD]

[size=8pt]IMHO, it is really necessary "like journalist do" to look a cupple of statistic ressource before concluding anything, even if the source looks reliable
 (Google, W3C, Mozilla,etc...) http://gs.statcounter.com/ , propaganda is never far[/size]

[DzzD]

China statistics are interresting too, this represent a lot of people.... http://gs.statcounter.com/#browser-CN-monthly-201001-201101
but maybe some statistic grabber dont really care about china, after all they are honest, no ?