As is common on Reddit, threads often take tangential directions. In a thread yesterday Grum threw some light on the development of Minecraft 1.13 and possible future improvements that would make sense in the Java edition. First and foremost, the goal of 1.13 is *not* a performance update, confirmed by Grum himself stating:
Where do people get performance from, that is not the goal other than at minimum keeping it the same!
The 1.13 update is focused around technical tweaks to the Minecraft core code which should enable future improvements to game. In the yesterday’s posts, Grum confirmed the Java edition of Minecraft has just five full time developers at Mojang, which contrasts with the Bedrock edition seeing more than forty full time developers – nearly ten fold more resources. Most interestingly of all to me though was talk of improving Minecrafts use of multiple threads with Grum opining:
Things we should definitely thread are the ticking of worlds, chunk generation and things like assetloading.
I’d imagine larger multiplayer server owners would leap with joy at the mere possibility of chunk generation one day being able to use the full power of multi core/thread processors. On bigger servers, repeating lag events can be linked directly to new chunk generation – especially so when using accelerated means of travel like, boat, elytra, or ice boat.
Projects such as HOSE have attempted to address this to limited success in the past, but direct implementation by Mojang could be revolutionary for SMP. I’d imagine the same would apply to ticking different dimensions on unique threads too, both items if implemented could transform server performance.
Grum confirmed that the under the hood technical focus of 1.13 might allow for these threading items to be look at down the road – which means serious performance improvements could be coming down the line for server owners. Watch this space as they say.