I had been reading the forums for patch feedback and such and after responding to a couple of threads, I think it’s a good idea I gather everything into a blog, not only for posterity but simply because a lot of people don’t read the forums at all, let alone a thread with a dev answer deep inside it.
Why patch now when the game is having so many problems?
We’re patching the game exactly because of the many problems plaguing the network. MIKA is the biggest code-only fix, improvement and optimization patch we have done to Squidwolf Server in years.
As a result of the extensive work done in the code, (total code is now about a million lines), we know there are going to be deployment problems. Veterans of Squidwolf Minecraft know this as “Patch Day Syndrome”.
Many have asked (or will ask), why doesn’t Project: Hazel test their patches? This is the eternal problem of testing high loads. We can’t duplicate the code’s native environment (Squidwolf Minecraft), because we have no way to replicate the load of 28k entities in the game. We do a lot of testing before we deploy patches and the ALIVE code branch, our internal nickname for the code branch, has gone through these phases of testing:
- 2 Months of Alpha testing in Polaris
- 1 Month of closed Beta testing in Sirius
- 1 Month of open Beta testing on the Alive test server. Max concurrent load was about 38,000 entities.
- Deployed for Release this month
Since then, we have further fixed and improved the ALIVE code base, which then adds 2 months of internal and public testing on Alive.
Those additions are small in comparison with what was done before the august release, then again we must remember that in the Alive environment it was released on a clean database. Here, we have to migrate a lot of data and upgrade a lot of database tables. This has also been tested on Alive, but even multiple tests there have not prevented issues from popping up in the last minutes.
So for short, ALIVE is supposed to help Squidwolf Minecraft out the current state, but it will be bumpy getting there.
Why cease support for SpoutCraft? Security? Encryption?
Let’s start with the Spout project. A player on the forum stated that less than 3% of visitors to his websites had the SpoutCraft client that we are stopping support for. For us, it’s less than 10% of visitors to the game world.
This still doesn’t make this an easy thing to do for us, as we have invested a lot of time developing features for Squidwolf Minecraft that depend on SpoutCraft.
I stated in the news that not only will the SpoutCraft client no longer work with Squidwolf Minecraft, but also the Spout project has completely stopped developing it.
However, this is just half the story. The security and encryption added to the server is very necessary, not just for our safety but yours too. Most client modifications will no longer work, we have enhanced over-the-wire security and beefed up authentication to name a few.
This wasn’t a decision made lightly or made by me in a drunken stupor (which I’m highly likely to do) but with us crying in a corner in the fetal position.
You see, our strategy is to support more Minecraft clients, not fewer. That’s why we want to work with RAGE, that’s why we’re looking at Squidwolf Minecraft Mobile clients for PDAs and more.
To those users with SpoutCraft, I hope you can understand our reasons, although I’m pretty sure it doesn’t alleviate the pain of having to upgrade or worse, quit if you decide not to.
Faction Economy changes
On the Faction Economy changes, this is more to clear up some misunderstandings and provide answers to what’s going to happen next.
Please note: No changes are being made to the land claiming mechanics or SquidCredits you have accumulated, ot the mechanism itself. Only that Faction Vaults no longer have any use. You can simply manually remove the emeralds within those chests and put them in your own.
First of all, we have a very old economy system which is costly in resources, is limiting in what we can do and requires the plugin creator to have a rudimentary understanding of Java 7 and our own unique programming language (HazelCredit) to create a native economy add-on, and the associated chances of bugs cropping up on all levels of the authoring of missions. This is bad, mkay? 🙂
We have had for some time a new economy system, which doesn’t require any of that understanding enables us to do more than the old system in a shorter period of time.
The “new” system was actually introduced in Inherit Velocity and you mainly noticed it in the early days of 2013. This is because the new system enables us to do things with developing our storyline which will be, like, really cool!
What is happening in this patch is we are removing the old system, which means the entire Faction Economy will be removed. But I stress again, YOU WILL NOT LOSE YOUR EMERALDS!
What does the future behold?
Since the introduction of the new economy system, lag has dropped by 20%
We have taken the freed-up server resources and added Faction Mobs.
We’re very sorry the transition can’t be smoother than this, but we want to assure you that we are really improving the economy system and we have a lot of people working on creating storylines and missions.
Oh, and to reply to the eternal question I’m quite sure will be asked:
Why do you have people creating new stuff instead of fixing stuff?
The people creating story lines and missions are content developers. They can’t write code, they can’t fix bugs in code, in fact, one could say that they were able to create bugs and if anything, they have less chance to create them now in the new agent mission system. 🙂