[TERA PC & Console] En Masse is closing, but TERA lives on! We will continue to support TERA PC (NA) and TERA Console until service is transferred. Stay tuned for more information.
[TERA Console] The Grotto of Lost Souls update (v85) is now live! Read the patch notes here: https://bit.ly/TERACon_v85
[TERA PC] The 64-bit update (v97) is now live. Check out all the changes delivered on August 11 here: https://bit.ly/tera64_patchnotes
[TERA PC & CONSOLE] Summerfest Part 2: The Beach Bash is on from August 11 until September 1! Participate in event activities to earn tokens redeemable for costumes, consumables, mounts, and more! Details: https://bit.ly/tera_sf20
Tera Development Presentation Slides (from 2010)
In case people haven't seen this China Game Developers Conference presentation on Tera in 2010.
I'm bored, so some highlights:
I have much respect for the developers.
I'm bored, so some highlights:
- Initial development cost about $30 million USD (using today's currency conversion, though the Chinese currency is pretty stable over time) and took 3.5 years to make. It still cost less than an average Hollywood movie to make.
- The goal was to do something special, to make a true action-oriented MMORPG, without the crappy boring targeting.
- The action-oriented nature means high CPU usage on the server-side, and greater cost to run servers than with simple targetting MMORPGs.
- The game actually goes into over-drive (on the server and TERA.exe end) when doing heavy combat, broadcasting packets more often (in an optimised fashion) when there's lots of action.
- The server doesn't use a physics engine, collision detection, etc is done on your PC.
- The world is broken down into zones, but those zones do not have a dedicated CPU thread, instead something more complicated is implemented (with some stream of non-blocking threads) so that an area doesn't lag or crash when there are lots of players in it.
- In prototyping each server (like CH) was actually composed of multiple servers for the action, and one server (backend) which is the database of all your items and progress, etc.
- Another prototype had an intermediary broadcast server for communicating with the two (in addition to the original connection between the two). The many connections made it too much of a pain in the butt to implement.
- What they decided on was having you connect to a single server, no matter what you're doing (somehow inspired by Starcraft?), this is the arbiter server and it communicates with the pool of servers which take care of the actual game mechanics.
- The way the game works is that threads are non-blocking, and things happen in an ad-hoc manner (hence the timing on ghosts in FINM is sometimes off, and things go out of order in places like KDNM/HM when the bolt stuff happens).
- The arbiter server actually does packet optimisation with spare CPU cycles. There is more concern about bandwidth than the server becoming overloaded (for better or worse).
- A main (e.g. CH) server can be kept alive even if a particular dungeon instance or continent server craps out.
- The game has been tested with some dummy clients (roughly speaking bots) running around in the game world, with an average latency of 200ms (this is probably why the game is playable even at higher ping, which is insane for an action game, quite an achievement)
- Worker servers are optimised according to their roles
- The developers believed in "impossible is nothing", to make an action-based MMORPG a reality.
I have much respect for the developers.
2
Comments
Or give us the talent tree which they use to make the new classes...until that happens, we'll see serious imbalance here.