Let's create a World Server

mindeject · #1 04-15-2002, 04:00 AM

I have read quite a few post, and though new to the emu scene, think there should be a world server run by someone other than developers. It would allow more focus on the server rather server/emu/etc. If someone is running a server or would like to get one going and needs a GM for quest and support, just let me know. It would be great to make a less crowded world to play in and not under control of the lame VI GMs, some are cool but for the most part don't amount to a hill of beans. Just let me know, I'm at school atm, but post any ideas anyone has, maybe we could get an epic server going, like have ONE BIG ASS quest for everyone to work on.

Mind

ibleet · #2 04-15-2002, 04:05 AM

Umm....anyone can run their own server.

TheClaus · #3 04-15-2002, 05:17 AM

I am sorry to say this but it is needed for this. RTFM!!!!!

Did you even look at the files you can get?

Did you read any of the forums?

Did you open your eyes and look?

Okay I am done.

Azaken · #4 04-15-2002, 09:09 AM

Clause is right. You can do all that and MUCH more. You gotta look at the goods.

Read the readme

malbolger · #5 04-17-2002, 05:31 PM

He means set up a massive server like some sphere shards has for UO where there is constantly 200 logged in clients.... instead of lots of small shards that we all can run.. But that will take server space and donations to support the broadband.. and someone with some sence to remotely run the server.

TheClaus · #6 04-18-2002, 01:56 AM

Well last I heard the network code is not quite there yet. In theory what is going for is to have many people running zone servers and 1 world server. That way the bandwidth is distributed more across the board. However the network code needs to be reworked in order to do this. Basically it lags hard when it is like this. Besides why pay for all the bandwidth and having servers when the goal is to run your own server. I am not sure how UO Emu shards do it but are they all running on T1 or something higher. If so then I think the goal of the project was wrong cause no one(except that 21yr old girl in FL that won $108 million in the lottery) has the money to put a T1 somewhere for a EQ shard. Though I promise if I win the lottery or Powerball and have $108 million I would gladly get a T1 for my house. Ahhhhhh wouldn't that be sweet.

*added the whole bandwidth thing

kiff · #7 04-18-2002, 02:47 AM

I spent a while working on uo emu's and I think that the only thing that they have over this project was tools, which come as a project ages. as far as networking code goes it is much simpler with uo cuz it's one big zone with a 2d client. one thing that they did have that was cool was the popup gm menu that gave you almost every option available to the server without having to alt-tab out to a different proggie everytime.

Lurker_005 · #8 04-18-2002, 08:55 AM

The problem with spreading an EQEmu server across the internet is the database. ALL the zone servers have to constantly talk to the mysql database as well as the clients. Now if the server machines are all at the same location on a local network, then the lag issue should be isolated to the server locations bandwith to the internet. The fatter the pipe, the more users it can handle.

stormgod · #9 04-18-2002, 09:31 AM

maybe the database should be handle in another maner like
players in one central database located on the same comp than the world server , and each zone server with his own database

and each zone sending only issues of combat or exp won only things like that to the world server database

I dont know very much about networking and this method definitely is asking for too big modification but maybe in the future they should be able to obtain something like that

Lurker_005 · #10 04-18-2002, 09:58 AM

Spliting/duplicating the data to seperate databases for each zoneserver (or cluster of em) would create.... MANY posibilities for error/wierd behavior. Think about what is in the database, items, characters, spawns, guilds, trade skill recipies. Some things would just have to be passed around between databases on zoning. And could work ok. The rest would have to be maintained and duplicated across all the zoneserver databases.

If it isn't maintained between all databases corectly, you could ennd up losing everything that happened in a zone. What happens if you log out in a zone, and the next day that zone isn't running. Or someone runs a zone with out of date item lists, all of a sudden your missing someitems...

I am not saying it can't be done, the project just dosen't lend itself well to running servers in differeent locals.

Oldwulf · #11 04-21-2002, 07:27 PM

from what little i know of this emu project the intent was to create a server that could be used for private play on a lan. and that is all you will be able to do with this emu in it's current state. to host a truely mmog you would have to rework the code for this emu from the ground up. working toward an mmog not a lan server.

if you really wanted to run an mmog you would need some major funding. and if you had that why in the world would you be working on an eq emu. why not make your life much easier and write your own mmog. if image and the other devs desided to write their own mmog it would be complete and making them money in a year, two years tops. it isn't easy trying to reverse engineer a server.

well now that i've ranted a bit i will leave you to your thoughts on the matter. oh and i hope i made some sense... i tend to rambe when i type. never understood that.

Oldwulf

Trumpcard · #12 04-22-2002, 12:21 AM

Well, think about it though.. The server side code is alot less than half of the battle.. The client is alot harder. The server side just needs to define how it responds to the client and interfaces with the data, the client has alot of additional concerns, especially the 3d graphics which are a major pain in the butt if you ask me.. I hate graphics programming.. You spend all day doing matrix multiplies and see how well you like it ! Tranforms on 3 x 3 matricies are not a fun thing. Plus theres a world of graphics that have to be created, sounds, music, etc..

Image and the dev team have done an incredible job, and Im not lessening what theyve done, but in a year the server side code might be at a release point for a unique MMORPG, but there is a whole world of client side programming that would need to be done, and that would be the tougher of the 2 pieces. I agree though, to truly produce a true MMORPG, it would require major funding. There have been some interesting attempts at open source MMORPG's, but they have been very primitive so far (though some are great fun).