Bomberman massively multiplayer in HTML5

Re:Bomberman massively multiplayer in HTML5 by rartichoke on Wednesday February 27, 2013 @11:52AM (#5293303)

I don't know. When a professional 3D game developer (the lead guy who coded grits) tells me it's necessary that both ends use the same language to correctly calculate predictions and simulations in the correct timing then I believe him.

If your game is completely out of sync and the input feels horrible because it's delayed then your game is going to be instantly rejected. Having all of your hard work instantly rejected seems catastrophic to me.

Personally I don't understand why both ends need the same language because the server is definitely going to execute the code at a different pace than a random client, but I'm going to take his word for it because he's a professional in the industry with a really good track record.

They had every reason in the world to use Go but they didn't.

[ Parent | on HN ]

Re:Bomberman massively multiplayer in HTML5 by Jare on Wednesday February 27, 2013 @07:01PM (#5295975)

It's not necessary per se (much less catastrophic if you do otherwise), but it's certainly much more convenient. It's only necessary if you rely on simulations being identical, which would happen if you are using deterministic dead reckoning or lockstep networking. Client side prediction and server side cheat detection are normally not so strict.

[ Parent | on HN ]