You are viewing a single comment's thread from:
RE: Why we need blockchain-based MMOs
Actually I was thinking about this quite a time now.
The problem is that MMO game is so dynamic - building a blockchain protocol for the game is rather a complex task.
If each action of the player is a transaction - you have vast amount of data for each second. And above this you should have block confirmation time
~1scecond - 20ms (for technical time frame rate)
So it should be PoS or something.
Next thing is blockchain size - you cant keep it from the very beginning - too much data. So genesis will have to be moved to father blocknumber.
If you have ideas to discuss please contact me.
Yeah, that were the technical difficulties I was mentioning, but again these can be solved by hiring right people.
My solution is to use different distributed protocols for raw data storage like IPFS or zeronet. The blockchain itself would only facilitate financial transactions in game using MMOCoins. Also miners to mine MMOCoins would also need to dedicate part of HDD to data storage part.
The calculations done by miners as I was saying in the OP won't be abstract math problems and instead would be things like calculate AI in the region X, etc. Things that actually benefit the game and aren't as hard to calculate. Also there will be probably some "powerminers" - powerful machines both in storage and computation that run the miner software, some of them may be set up by game's company (since they'd get % of transactions at in-game $$$<->MMOCoin exchange) that would alleviate the problem.
Also to alleviate the problem of possible data manipulation (e.g. in response on aforementioned "calculate AI" sending falsified data that would freeze the AI or make them all die) the same "problem" would be sent to multiple miners and then any results too different from others (indicating fraud) would be thrown out. Miners would be chosen to be closest to the most players in specific regions to minimize lag.
Things like player card, eq, inventory and so on, basically any things that can't be sent in under a kilobyte would be sent using IPFS/zeronet which are good at this.
Now that I think of it, the hardest part would be probably marrying blockchain with IPFS/zeronet/some new protocol designed for it enough so it will work well.