Server architecture: not for fun

It’s cool, wasn’t taking a dig at your grammar its understandable :smile:

Yeah it was that and a few other things lol… but back on topic

do you think if the stored location data in nosql database? or simply as the server memory is faster and better? file, mysql, nosql, memory (loaded classes)

I think use redis or mongo for geodata and c incudes for to complex calculations and sphinx for search

R-tree.

And here we see why soooo many games have such bad network code.

seriously. Just. Get. A. Game. Working. then worry about it.

1 Like

Am kinda conflict about this. Creating a multiplayer game from a single player game must be a nightmare. Am working on a multiplayer game right now and really the packets are the spine of the application. You would need a really clean architecture to get the transformation to work properly or it would become a bug nest infested by zergs and ladybugs.

So why not go the other way around: Getting the network to work before building the game around it. That being said, server building is not the funniest part of game development and pure synchronous game must be really hard to build. An asynchronous game is fairly easy to get tho, but really this mode cannot be used for every game type.

Am no expert at all however, and really seeing this thread make me wonder about so many things x) I should probably rethink my architecture x)

I a server programmer, i’ts simple.
It is simply individually. difficult to relate:
synchronization + network, grid, map editor, editor of quests, monsters ai, make a bunch of quests and items. write the updater.

Well, it’s a lot of work, but if you get the server part right it gets pretty easy. I have been working for 2 years now, only part time and whit no starting knowledge, but now the grid, map editor, and general combat is pretty much done.

After yes the ia and balancing can get pretty big, but this is why you usually try not to work alone. I don’t know how many people you plan to have in but if you just want to do a one man game whit-in a short period of time, networking might not be the best plan :chimpanzee_closedlaugh:

But am wondering, how far are you in the project? Cause if you would consider joining one, I could use some help. I have been crasy enouf to start an MMo and still not sane enouf to stop, and am pretty sure a second view on the code could help me, and probably you too :stuck_out_tongue:

I count on 2-5 years of development.
the worst thing to make a mistake with the architecture of the beginning.
and the need to do a lot of tools that will help simplify and speed up the work: how wysavig - for ordinary people.
think:
I have:

  1. Monsters: parameters, drop sets, range of attack, time and place of occurrence. attack range, the model animation.

  2. from the drop sets, and there are things drop rate (uslviya quests).

  3. things have: the model, requirements, modifiers

  4. quests - scripts: the ability to use the monsters, players, grid, chain quests. NPC dialogues. game events.

  5. global gaming events: the siege, opening event

  6. chat: clan, group, raid

  7. Grid: a graph on which you can set objects (map editor), calculation of distances between objects, used in AI and NPC monsters

  8. Map Editor: ability to set the scenery, set the monsters, set the NPC.
    Editor: quests - the opportunity to write the script and bind to each other all that is in the game

  9. Editor of things: the creation of descriptions of things. creating drop sets.

  10. the table of characters experience

  11. skill Editor



    All this should quickly communicate without errors

i write pdf sampler for beginers: basics, thread, net, db : https://yadi.sk/i/vzIZj67Fqqy4N
books + internet + docs

I start a develop framework for game server (java):
1)start stop restart daemon
2)mvc (Model-View-Controller)
3)orm (dao/active record)
4)sockets (udp)
5)messages
7)user sessions
8)system
9)localization
10)codegenerator
11) virtual objects + memory control

http://t-machine.org/index.php/2007/09/03/entity-systems-are-the-future-of-mmog-development-part-1/

sketch

udp
http://cppjavabahslinux.blogspot.ru/search?q=udp
tcp
http://cppjavabahslinux.blogspot.ru/search?q=tcp
client
http://cppjavabahslinux.blogspot.ru/search?q=client
server
http://cppjavabahslinux.blogspot.ru/search?q=server
uml diagram editors dessign patterns srchitect
http://cppjavabahslinux.blogspot.ru/2016/04/diagramm-editors.html?q=uml
pattern
http://cppjavabahslinux.blogspot.ru/search?q=pattern

Create multi thread framework tcp/udp
http://jsockframework.blogspot.ru/

Hello some books p2p multeplayer
https://vk.com/2bizdotnet?w=wall-86263143_1472

Are we supposed to be banned in google? :no_mouth:

? What does that have to do with this thread?

I mentioned most recent post. I didn’t see question or comments, just curious what was expected?

I wan’t help those who also makes the network for games