Looks like it’s not going to appear any time soon…
I like the way Penny Arcade says it.
Console VMs come down tothis…
Some fo us at Sun think it would be a veru good thing for Java. SOme of us ven believe their are workable business models for Sun to make revenue out of it.
BUT they arent the business models Sun is used to, so getting management to understand them is hard. Some of us in Sun take a run at it every so often but we also have to chose our battles. Project Darkstar, being server technologies, is something that is muche asier to make Sun understand (and even that took years of effort fromChris and myself.)
There really is no compelling business reason for Sony to care. I dont thin kthey’d stop us, but they dont have a reason to pay us or help us either.
As I say, one killer Java game that was a PS3 must-have would chnage the world.
Til then, you guys who are writing successful and noticed games are helping, even if they arent the killer title to push Sony over the edge.
Can you comment on one of these proposed models? I’m curious to know.
[quote]BUT they arent the business models Sun is used to, so getting management to understand them is hard. Some of us in Sun take a run at it every so often but we also have to chose our battles. Project Darkstar, being server technologies, is something that is muche asier to make Sun understand (and even that took years of effort fromChris and myself.)
[/quote]
I’m in a similiar situation on my job. Convincing management to try something new? Winning the lottery has better odds. ;D
[quote]As I say, one killer Java game that was a PS3 must-have would chnage the world.
[/quote]
Is there a java game engine that can produce a AAA game at the level of engines like Renderware or Unreal Technology? It would take that kind of game to get Sony’s attention. This isn’t a put down, I would just like to know.
there are two market-places today: business market-place (servers, programs like office, everything that companys like sun or microsoft can provide other companies ) and second : game-market-place. Both are strong and both make huge revenus, but sun is really intrested in one of them- business (from that what said Jeff management is intrested only to those projects which are sure to earn something on it). Game servers are finally step forward to good direction, but i think that jvm for consoles is that what sun can realy benefit, why? Becouse it is inwestment in future of modern platforms which every year gets more popular, consoles gaming are getting stronger every year , it is huge market (sony plans to sell 100 milions of ps3). Well it is my opinion but i think that pc gaming every year will get smaller, true money are today in consoles market.
Sun with success has promote java technology for cellphones but sun has very lucky situation in the first years of mobile cell phones. phones VENDORS demended technology for launching programs on diffrent configurations of their phones so they have inconporated java . Generaly sun sold licenses and made standard jvm how it could jvm for phones made and it is everything. Today sun control only jsr program and thets all. I mean that other companies has made lots of job for sun , phone vendors promoted java etc on their own. TODAY the situation is reversed sun must make FIRST STEP to gaming market and not wait like in early 90’s that someone will implement java technology (if netscape would not inconporated their java tech in 1995 today probably NO ONE heard about java ). So sun first time must make step on their own not wait that sony will make their jvm … I like sun and i have nothing against that company but it should at least try to make another step forward…
SUn traditioanlly is a business to business company.
Sun gets the cell phone market because there is someonesun undrstands who is willing to pay a lot of money for a Java license (the phone makers) anda single monsterous customer to sell the benefits of Java to (the carriers who select the phones to be sold to their customers.)
COnsoles are different. Sony is perfectly happy with the etchnologies it has and has no reason to spend money on a console-wide Java license,
A model that might work is to license it as middleware to trhe game developer, (You want to release a Java game on PS3? Sure, itll cost you a quarter per package sold.) That model actually has the potential to make a lot mroe moneyh then what we make on handsets. On handsets were just getting paid per box, but this way we get pauid per peice of software sold ON each of those boxes.
But it would take vision and bravery for Sun to go after. Its not like anything they have ever done before. And anyone who works ina big company Im sure has heard the “thats not our ‘core competance’” refrain…
Well, I’m new here, but here’s my take on the subject.
There are currently three major game platforms: PC, PS and XBOX. Two of them are controlled by one company - the big bad giant from Redmond. Fortunately, PS is still a more successful platform than XBOX, and will probably remain so for the next couple of years. However, if at some point the tide will turn slightly in favor of the XBOX, developers will have this to consider: They will most probably be able to use the same code for PC and XBOX, and use the same tools Microsoft will develop for them. In addition, it is more likely than not that Microsoft will start pushing using C# for game development for both Windows and XBOX, and they will make it easy to use for game development and they will make it work. In fact this could be one factor to push game developers to Microsoft’s arms.
Now once that happens (and it will happen) C# will become a standard development language for games, perhaps even THE standard. Since .NET is also used for enterprise applications and now even for mobile phones, C# might become the language of choice for teaching in schools. Once that happens (and it will happen), Java will be in serious trouble.
Therefore, business model or not, Java MUST be on the PlayStation even as a defensive strategy, because if a language is to become the leading language it must be everywhere. This is what Microsoft does. It knows that even if it’s technology is not in the lead everywhere it must at least BE there or be gone.
This is what Sun must consider. If you can do everything with .NET but only some things with Java, young programmers will only learn .NET. And as much as we all like Java, we must admit that .NET is not that much different. These are two quite similar competing technologies. So if we want Java to thrive, it must have a presence EVERYWHERE and particularly in the game console market which is both huge and sexy.
You forgot MacOS, which for some of us, is a major platform too.
Cas 
I think my point remains.
Back in the early nineties, project managers in the defense industries were faced with a choice between two languages: C++ and Ada. The American DoD bet on Ada and was followed by many defense contractors world wide. Well, they bet wrong. And not because C++ is any better than Ada. They chose Ada because it was a safer, more readable and better organized language. They believed it allowed better collaboration. They were right on all counts but they still bet wrong. Why? because C++ became ubiquitous. So by the end of the nineties, it became more and more difficult to find Ada programmers or nifty Ada development tools.
So, if I were an IT officer for a big firm who bets his chips on Java, I would be very much concerned if I saw that Sun was abandoning the gaming industry, because I realize the importance of that industry and its attraction to programmers. Now, you can say that this site proves that Sun is paying attention to Java games, but in the current market conditions, abandoning the game consoles means abandoning serious games altogether.
So if I were an IT officer who’s wise enough to learn from past mistakes, I’d think twice before I bet on Java.
Because, you see, when you find yourself in direct competition with Microsoft you have to keep running just to stay in one place. Sad, maybe. Unnerving, certainly, but that’s the truth.
So I hope Sun finds the right business model pretty soon, or Java will start losing ground.
The difference is that Java itself already has gained so much momentum in the business space it’s basically unstoppable in that area. Gaming is just a different area which Sun may or may not be bothered with competing in. Our purpose here is to convince them to put a wee bit more effort into it 
As for Sony and a JVM on PS3, the business case has conveniently been proved quite nicely by the success of indie developers on XBox - but in truth the story’s only half Sun’s issue there: they can only push it so much, Sony have to agree to implement it.
Cas
Unstoppable like COBOL or like FORTRAN? These two languages were very popular in their niche, but never made it out of there. Or maybe like Ada which was a general purpose language like Java but was used primarily in the defense industry?
As for Sony, they might have some interest, albeit a technical one. Programming for the PS3 is going to be very difficult because of the groundbreaking yet unfamiliar concepts used in the CELL processor. The CELL is composed of one PowerPC “driver” processor, and eight Symmetrical Processing Units (SPUs) each with its own on-chip memory. The SPU’s memory is not a cache of the main memory, and data has to be written to those memories explicitely. This provides extremely high performance while completely avoiding the difficulties of cache locking. However, it’s a bitch to program.
Now the JVM with its smart memory model, and the newly added escape analysis and lock elision might enable running regular Java threads on the SPUs with the JVM worrying about all the tricky details. Another point is that the SPUs make use of a different instruction set from the one used by the PowerPC “driver” CPU, and so the code that should run on the SPU must be compiled to that instruction set. The JVM, on the other hand, might choose at runtime which threads to run on the SPU and which on the main processor.
This brings me to mention the third, and very important player, in the PlayStation JVM issue, and that’s IBM. IBM developed the CELL, and they are going to use it for their servers. IBM endorses Java and has tremendous expertise in JVM development (In my experience their J9 JVM is about twice as fast as HotSpot, and most of the papers on JVM technology are written by IBM researchers). So IBM will probably develop a JVM for CELL, the question is only how soon. If Sun and/or Sony want to collaborate, all they have to do is adjust the JVM which IBM will develop to whatever OS the PlayStation is using.
I think it is inevitable. IBM will surely develop a JVM for CELL. But would this push Sony into adding that JVM to the PS3. I doubt. But it would be a first necessary step. Sony wants to make money off there game console. Like Jeff said earlier, a great game will get their interest and a JVM for CELL will make it possible for someone to make that great game. Great meaning all of the visuals, high level physics and AI.
IMHO, the problem is that there is no such thing.
The only thing that would come close to being a “killer game” as far as PS3 is concerned is a game that makes tens of millions of dollars on PS2 or PS1. That’s it. They couldn’t give a rats arse about a game, no matter how great, otherwise.
How many times have you seen Bejewelled being sold on PS2? That’s a killer game from every single angle you look at, and one of the best selling games of all time (profit-wise). How much interest is there in a PS3-Bejewelled? Go figure.
So … let’s not talk about this mythical Java killer game that would make Sony sit up. There isn’t one, until there’s a JVM on a Sony console.
I won’t agree to that. Java gaming is just now slowly merging into the mainstream. So a great game may get their interest. But if what you say is comes to truth, then java gaming on a game console will never happen. But maybe a great game is not the way to go. Maybe a great demo on a high end machine and submit it to SCEA? ??? Just a thought?
I tend to disagree… the Quake Franchise. The GTA franchise (if they havent wrecked it). Heck even SIms4D would probably do it.
Your problem is you are defining killer from your perspective which is wrong. You need to define it from Sony’s perspective. And from Sony’s perspective it is “does the availability of this game potentially drive a ton more Playstations off the shelf…”
To convince Sony and other big players in the industry that Java is a very good thing on the consoles, I think that we need something equivalent to Renderware because we need APIs that accelerate the productivity of developpers not only for PCs, Macs and Linux but also for consoles. JME and other high level APIs of this level do not compete like Renderware because they do not have the same goals.
As I mentioned before, my concern is that Java has to convince at all, while the competition (Microsoft and C#) just makes things happen. I realize Sun does not have the same effect on the market as Microsoft does, but if Sun continues to wait for a “killer” game from Sony’s perspective or anyone else’s perspective, which might or might not happen, and if it does - might or might not convince Sony, then we’re in trouble.
I hate to sound pessimistic, but this is what will happen. If Sun (or anyone else in the Java community) passes over the Sony PlayStation, which is becoming one of the most successful "computer"s ever, then that means that Java is giving up on serious game development. Microsoft isn’t. And soon C# will become the convenient language you can do everything with, and Java will become the convenient language you can do ALMOST anything with, and almost is not enough.
And then, sooner or later, young programmers will learn only C#, and Java will be toast.
I’m just trying to point out the tremendous effect that abandoning serious gaming is going to have on other applications of Java. Maybe that will scare Sun into action.
[quote]And then, sooner or later, young programmers will learn only C#, and Java will be toast.
[/quote]
You seem to forget that Java has a huge marketplace in traditional business systems. Having said that, how can you presume that Java will die? The fact that in the future developpers will be able to do everything with C# doesn’t mean that everybody will say: Hey, I can do everything I want with C#, let’s give up Java! And the shools can’t ignore the trends of IT. That’s why Java still has a very strong place even after many years of .Net. Java developers and companies don’t want to switch to C# because you can do the same thing with it. C# will become dangerous when it will clearly surpass Java in terms of productivity and development costs.
First of all, .NET hasn’t been around that long, and it already took a bite out of Java.
With regard to schools - consider this: according to Gartner, the IT industry is split 50/50 between Java and .NET when it comes to new projects. Java still has a larger user base because it’s been around longer. But let’s assume that in time the industry will be split exactly in half. I can live with that. I like Java better, and I don’t have a problem with having another “super-power” around.
But if a school can teach only one language, even if the market is evenly divided, it will probably choose the language that you can really do everything with. That’s what the students will want to learn, and the schools wouldn’t care much about Microsoft’s dictatorial regime. In fact, I can bet you that Microsoft will start paying schools or giving them other incentives so that they favor .NET over Java. Now, I realize Sun can’t do everything, but passing over PS3 is a bit like simply giving up.
It pains me to see how Sun answers Microsoft’s agressive marketing with what must be described as sheepishness. C’mon guys, we want to see some teeth.
Sun doesn’t realize that there are certain things you must do even if they don’t generate tremendous revenue right away simply for the marketing benefit. We want to see Sun putting up a fight.
i think that today is the moment when sun should decide if he want to enter gaming market or not becouse consoles will became much more popular and in my opinion pc gaming will loose its position. PC’s will be only used in business . Today pc’s are popular becouse lots of people use them for everyday functions like internet / email etc i think that consoles in future will incorporate some of those functions. Note that the latest xbox 360 and ps3 have thier own hdd so its one step forward to incorporate pc features etc. Of course it is not something new becouse even to ps2 we were able to connect hdd, but the new generation of consloes will have them as standard. I think that consoles in future will play the role of today pc, really . Theoretically all things you make in pc you could be able to do with ps3/xbox like using net, typing documents and priniting them why not? ps3 has usb so only special drivers are required. I know that consoles are today only realated to gaming it is true , but in future i think not pc but consoles will be the standard in every modern home . Why becouse of economic reasons, they are much cheapier than pc’s which you must upgrade every one year to play latest games, but console you buy only once for 5 years. As a general cost they are much cheapier. And when they will be so popular, consoles user will demand some new features so maybe programs like office will appear for consoles etc. So when in future pc’s will be used only at work sun will not have java gaming clients (except emplotees who play games at work) and in that time it will be much harder to get to true console game industry…
Your argument about future school trends makes some sense but I’m not 100% convince of that. Anyway, what should we do now instead of just complaining? We know that compenies listen to clients to make more money. We, as indie developpers, can sign a petition or whatever the form of document needed to ask Sun (and maybe the JCP) to develop a JVM for PS3 or Revolution or XBOX 360 or all of them (ok XBOX 360 is unrealistic ;D). Our voice can have some weight but what we need most I think is the recognition of the need for Java in the gaming indistry (AAA titles) at large and this is far from reached. C/C++ developpers don’t think Java is an advancement over traditional game dev. The obvious advantages of Java in terms of productivity over C is kind of ignored maybe because they are used to magage with the evils of C and they have access to excellent kits like Renderware. And don’t forget that they manage with memory pointer crashes because of middlewares that take care most of the time of these situations at compile time.