Project Darkstar -- potential technology platform for Spring

[polaris_silvertree]

Hey guys, I am new to this forum -- but not new to TA or TA-spring. I'm also linked into the gaming industry from a technology perspective (software/hosting). I recently posted on the AI forums on a topic about an open source lounge and community effort (dated from way back to 2006).

I had recommended that you adopt Sun's open source Gaming Server for this task. Its code-named Project Darkstar and it is being evaluated by a number of top developer/publisher houses. It made its debut in 2005 at GDC. I believe its now at a stage, where the product is comprehensive and polished enough that TA-Spring may benefit (I say may because I haven't had time to review the source).

It does much more than that though. For example (and btw I don't work for Sun) -- I saw Never Ending Nights played on the PSP against a guy on the PC (same instance -- so persistence is preserved). You can actually play across phones, PSPs, PCs, even Gamecubes... Granted Spring games will probably be a formidable task for say portable gameplay -- but it may address any future desires for cross-platform functionality. If you have been following the technology, it has been circulating around the developer houses. I know Chris Taylor (Gas Powered Games and author of TA and SupCom) has been at many of these venues.

Bottom-line is that its free and it basically handles the "socket layer" (or basically the network stack) as well as the cross platform issues right out of the box. If you are interested, I know the team and the architect of that technology personally -- and could provide more info.

http://www.projectdarkstar.com

I am wondering if this would ease the burden of having to write code for these areas -- and bring focus back onto the game IP (i.e. art, engine, gameplay). Just throwing my $.02 in hopes that its useful for this effort.

Regards,
Polaris

[AF]

www.darkstars.co.uk Stop nabbing my name!!!

Spring is written in C/C++, this would require us to port spring to java

[lurker]

A toolkit for making persistent virtual worlds? What would we use it for? And the socket-level stuff is cake compared to floating point sync and save/load.

[polaris_silvertree]

First of all, while the Project Darkstar platform is Java -- there are C++/C APIs for it. From what I understand, Sun is actively trying to hammer out the docs and sample code for this to help people understand how it interfaces with Java. So, we really aren't talking about re-writes into Java (if that was your concern) -- that is strictly for the server back-end side. The game engine that sits on top of Project Darkstar should remain C++.

Persistent worlds is but one of many types of applications that you could use it for. The point of the matter is that it optimizes that piece for the game developer without them having to work through that also...

Getting back to the comment about Persistent Worlds -- I was thinking with regards to large scale tournaments going on at the same time (probably regionally due to time zones) -- where you can have campaigns now handled by allied teams of players at the scale of SupCom. And as the respective battles converged geographically -- you have seemless entry/departure. Just a thought.

Regards,
Polaris

[AF]

We can already have games with more players and unts than supcom can handle. The newest spring can handle 250 players in a single game and 32,000 units.

Basically I dont see the justification behind taking an extraodinary amount of effort to add a dependency that does something that what we already have, already does, with no benefit.

[polaris_silvertree]

Fair enough -- thought the highly scalable cross-platform optimized back-end that Darkstar presented might have some future value to the effort.

[lurker]

The backend doesn't have to do anything but relay traffic.

[imbaczek]

i like the idea, but this comes way too late for us - we've invested quite a lot of time in something different and working on new things like this in the light of Oracle buying Sun isn't an encouraging thought.

[polaris_silvertree]

The news of Oracle buying Sun hopefully shouldn't hinder Sun's open source directives other than promoting future communities. Oracle has no real line of sight into the gaming space -- so I doubt they will touch this one.

I understand that a heavy amount of investment in time and effort have already been made. I didn't know the state of affairs with development -- so I just threw it out there.

[el_matarife]

Oracle will probably just stop sponsoring most of SUN's open source projects. They can't do anything to touch the GPL code that's already been released and I think several projects are already quietly discussing forks if Oracle starts trying to run things differently.