Irrlicht Engine

The point (and problem) with p2p is: you can't trust the client. I think many people in this thread tried to explain that. It's not about earning money but avoiding cheaters.

But i guess this discussion can be deferred until there is some release to talk about. Good luck anyway...

iZOTOPE wrote:why you guys always have to say bad things over a good idea?

I'm pretty sure that I've never said bad things about a good idea on these boards.

Do you have a timeframe for the demo? Or a feature list?

[quote="iZOTOPE"]not to say "bah.. this p2p isn't possible, this and that can't be done..etc", you must have faith in what you are doing
quote]

Well, that's not much reallistic, there are things that are too hard or even impossible to do, regardless of how much faith you put in it.
For example, I have faith in building a game where 30 millions of people will connect on my home computer with a simple ADSL connection, and I will render trillions of polygons in every single frame making the most awesome graphics in the industry. I have faith that I will develop nice algorithms that will allow me to do that.

Sometimes faith is not enough...

monkeycracks wrote: ...you guys are being really harsh on him.

Haha your right. Why is it that people always get so flustered when someone mentions the word MMORPG?

BlindSide wrote:

monkeycracks wrote: ...you guys are being really harsh on him.

Haha your right. Why is it that people always get so flustered when someone mentions the word MMORPG?

They're scared someone MIGHT defy what they've always said and do something they thought couldn't happen and prove them wrong.
And no one likes being proved wrong.


Good luck Izotope

Well, that's not much reallistic, there are things that are too hard or even impossible to do, regardless of how much faith you put in it.
For example, I have faith in building a game where 30 millions of people will connect on my home computer with a simple ADSL connection, and I will render trillions of polygons in every single frame making the most awesome graphics in the industry. I have faith that I will develop nice algorithms that will allow me to do that.

Sometimes faith is not enough...

yes but there is a difference from spouting out random things that are obviously not going to work, from thinking about new and inventive ways of doing something, not saying p2p is going to work, but kudos to him if he is trying to figure a new way to do something, you like the guy 1000 years ago telling everyone that the world i obviously flat.

i dont think that its possible with no dedicated server. clients have to know what peers are playing, where to get all info about world and current actions. at least such game need something like torrent tracker. another problem - game rules. if everything is implemented in client, then i can hex the exe and play with different rules. thats bad too. so again we need something like server to get all rules for counting and so on... even starforce3 was cracked so you wont be able to create such a strong protections to protect all data. anyway idea is realy interesting

I've toyed with the idea of a p2p massive multi-player rpg game for a while, and I can think of several ways of doing it.. Of course, the gameplay would have to be unique (not like a traditional mmo at all) more of a 1-player game where other players can be invited to join in.
like rogerborg said, you can't have people on p2p actually playing the same game, but that doesn't mean you can't fake it and still make it a unique and interesting experience.. you see a ghost, I see a goblin, I can't pick up your loot, but we can still chat and do the same missions together.
having said all that, i'm also in the "it will never happen" camp!

doing same missinos with friend online is realy interesting. it would be like splinter cell chaos theory(i think this part) cooperation mode

monkeycracks wrote:They're scared someone MIGHT defy what they've always said and do something they thought couldn't happen and prove them wrong.
And no one likes being proved wrong.

Thats nonsense.

It's just that i only once would like to read a post like "here is my playable demo of a MMORPG with network support" instead of those "I am going to make a super duper MMORPG... blablabla" postings.

If you make the MMMORPG with P2P you still need a master P2P client for setting the AI, see the example from bitplane, thats why the commercial MMORPG today don't use P2P, they want to have full control over everything.

I've also toyed with P2P for mythorgia and i always get stucked on controlling the primaire jobs such as the AI.

If you have a master p2p host for AI and let the players only send player positions, and let the servers think you get realy performence boosts because the server only recive playerdata and send AI data, so you only need an AI server instead of a Database server and data transaction server.

Maybe you have to make a image with your netwerk idea

lostclimategames wrote: yes but there is a difference from spouting out random things that are obviously not going to work, from thinking about new and inventive ways of doing something, not saying p2p is going to work, but kudos to him if he is trying to figure a new way to do something, you like the guy 1000 years ago telling everyone that the world i obviously flat.

IMHO i think it is easier to make 30 million people play on my home adsl than doing a p2p mmorpg, so who is spouting random things.
I can do a 30 million people mmorpg with an adsl server, I'm sure of that,
by the same rule you have to say i'm an inventive guy that is trying to inovate, if not, you are like the guy 1000 years ago telling everyone that the world is obviously flat.

heres a p2p idea, it wont be completely p2p but a hybrid, you need several mods right? lets say you hire mods, and they have certain times they have to be on, they can all split the sever work, lets say 20 mods, that would allow huge worlds, and when a mod logs off, the servers switches peers like in heroes of might and magic, you would need a hierarchy of peer status's and have to allow several server setups on a peer that way if a computer goes out, an already existing one could shift to it.

Employing moderators/GMs is an admission that your design or implementation is flawed. At best, they can amelerioate the worst effects of your design/implementation until you fix it. At worst, well, "frenzied witchunts "would be a polite way of describing how bad it can get. Spend a fraction of your future projected moderator budget up-front during the design phase, and avoid the issue altogether.

I rather like bitplane's suggestion that you needn't necessarily be playing with consistent game states to get the impression of playing with others. A game needn't (by design) be as secure or consistent as a financial database.

That would certainly be a bold design decision, but it would save you an awful lot of work during design, implementation and maintenance. It would only be practical for co-op or a sort of player-instanced PvE though; PvP (including selfish PvE) simply can't be entrusted to clients, because of the "No you've taken 10,000,000,000 damage" problem, and you might as well admit that within days, everybody is going to be packing a +10 Orb of GODMODEness.

I dont understand what the issue would be with using many mods pcs for the servers would be, can someone explain?