Q: Stop teasing us, give us a clue!
We finished up for the day here just now. It was a good day, the focus has been:
Finalizing the flow into game
Not holding up release
All the basic stuff like, creating your character, joining a server, etc... Then some steam integration so you know which servers your friends are on, and such. We're using steams API, which while is it great being free of gamespy we're not sure whether external apps (like six or dayzcommander) are able to poll steam's servers. So we want to make sure our browser provides some improved functionality.
Optimizing/Bugfixing Dedicated server.
This is why we are not out. We need performance
We think we need a minimum of 15 FPS will 50 players, 2000 zombies, and 25000 loot items. Our latest tests have all shown some runaway systems in the code we have to tidy up. The variable synchronization system that was developed for work with the network bubble, is checking the variables very often. We're optimizing this. Also, there are many string comparisons. These have been refactored so references are used (lookup numbers) to speed up the process. things like animations etc... are generally recorded as strings - ArmA wasn't build to handle very large numbers of things so this has been a large area of optimization.
We also have a bug where sounds (which are temporary vehicles) are being queued up and sent to all JIP players. This causes us a steady loss of performance on the server.
The synchronized variables are also checked for each player, this is inefficient and we are refactoring that. It is our biggest obstacle to releasing the alpha right now. We know what needs to be done with it so we're working on that, then we will again reassess the performance.
Why not just say its not out on xxx day?
If I start saying "oh it won't be out then" people start asking me about the day after, and the day after. So it just encourages people to keep asking me when it is, and the "announced date" would be when I go "yeah, it could be out then".
What we have now
We have now something that provides basic functionality for 10-20 players. The new zombies are in, they provide excellent pathfinding outdoors, and improved pathing indoors. They are capable of breadcrumb navigation or line of sight. At low server FPS they will start to rubberband and glitch through walls much more often. They are very much a work in progress.
Client performance out of the cities is very, very smooth. We still have a bug in the cities that occurs (and shouldn't) that causes lower framerates. This will eventually be solved, but for now it is noticable when looking at center of cities. Overall, the feedback from the testers is that performance is much smoother than with the mod.
The inventory is a bit of a mixed bag. There are some mistakes we have made that, unless we delayed the standalone, we can't fix until later. But overall, I think it is a huge step up. Stacked items, wetness, damage, crafting, containers, clothing, weapons, pistol holsters, bags, melee weapons, chainsaws, masks, gloves, boots... you name it.
So, what next?
We keep optimizing the build. In the meantime, I would encourage everyone to checkout Project Zomboid which is now available on steam. This has been a massive design inspiration and I recommend it to everyone interested in DayZ. [source]
Gâteau