The threshold is still 35% of max power capacity (but per chain now).įixed an issue where destroyed blocks would re-appear (client side only) in some cases.Shadar73 : awesome advice! It works perfectly. Turret power sharing (for AI) now looks at every chain before it stops firing. T1025 Joystick and Gamepad support brokenįixed a layout issue for the Pixel texture pack. T1830 Block information for Russian/Traditional Chinese cause freezes This combined with the physics optimization should make ship battles a lot more enjoyable. From testing results, weapons that used to freeze your game for several seconds per volley, barely affect your FPS now. Volleys from capital ships will no longer turn the game into a slide show. This is now optimized to run several times faster (from around 1fps on two big ships with 200+ entities colliding to around 40-60, coming from 80 fps).īlock modifications in general, but especially block damage and removal caused by cannons, damage and salvage beams have been sped up considerably. However, sector changes still cause your ships to jitter, that’s a different issue.Īlthough physics performance on single entity ships is in a much better place now, stuff gets to a crawling halt when two bigger ships with docked entities meet. Speeds up to 300m/s had no noticeable jitter on the official test server. The new lag correction algorithm does a much better job at smoothing client and server differences out. This will improve bandwidth usage on bigger servers by a lot since the positional data is at least 50% of the total data sent.Īs you all know, when going fast and trying to hit an equally fast ship, stuff starts to jitter around. The further away an object is from you, the fewer position updates you will receive. Positional data is now sent dynamically based on where the players are. It also improves general loading speed by about 30%-40% (reported by our testing team). It doesn’t eliminate all the short freezes on startup, but it makes it considerably better. The data that needs to be in sync with the main thread is prepared and applied in one step per chunk. This archives much faster chunk loading as it utilizes multicore processors. Most of the work has now been put into the generator threads that load the chunks. This was mostly visible when starting up. Since this step was done in the main thread, it often caused fps drops, and/or stalling (small freezes). The step of putting the blocks in the game physically and updating all logical dependencies as well as center of mass, tensors, and other metadata. One major bottleneck of the old system (even more so with the new chunk32 system) was the final step of loading or generating chunks for the game. The latest optimizations make ship battles, including capital ship fights a lot more fluent. ![]() However, this doesn’t mean that the updates won’t have any impact. ![]() Since mainly we are working on the NPC factions feature at the moment, which is a big update with a lot of planning, design, and architecture, the updates during this period will contain smaller features, fixes, and optimizations.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |