Version 1.1.43

[FactorioBot]

Changes

• Reduced multiplayer latency on very good connections by up to 66ms(4 ticks). (71901)
• Changed multiplayer latency calculation to be smoother. This should reduce multiplayer stutter on bad connections.
• Changed default network send rate from 30 to 60 packets per second and added an option to configure the rate in server-settings.json.

Bugfixes

• Add a 70% damage bonus to discharge defenses with Energy Weapons Damage 6 and 7. (100149)
• Fixed flamethrower turrets not reporting the correct damage bonuses
• Fixed gun turret tooltips showing the ammo in the middle of the other stats
• Fixed artillery wagon and artillery turret tooltips not showing the damage of the loaded ammo
• Fixed drawing wall filler when hovering mouse over ghosts. (100243)
• Fixed crash when trying to connect wire after wire type in cursor stack was changed by script. (100220)
• Fixed roboport would not output signals for 1 tick when electric pole was mined while having circuit wire from same circuit network. (100221)
• Fixed that the pole drag building skip trigger didn't work, so the tip was always shown even when the player used drag building.
• Fixed LuaItemStack::get_tile_filter would incorrectly return entity filters. (100288)
• Fixed mining drill GUI not being openable if the drill didn't have module slots or energy source with GUI. (100282)
• Fixed a crash when using 0-length debug names in Lua bytecode. (100321)

Modding

• Added TurretPrototype::energy_glow_animation_flicker_strength.
• Removed obsolete pollution-visualization sprite, which was replaced by pollution_color utility constant.

Scripting

• Added LuaEntity::connected_rail_direction read.
• Added TrainScheduleRecord::rail_direction.

Use the automatic updater if you can (check experimental updates in other settings) or download full installation at http://www.factorio.com/download/experimental.

[Vxsote]

Reduced multiplayer latency on very good connections by up to 66ms(4 ticks). (71901)
Changed multiplayer latency calculation to be smoother. This should reduce multiplayer stutter on bad connections.
Changed default network send rate from 30 to 60 packets per second and added an option to configure the rate in server-settings.json.

This should be nice! There is another thing that I've been tossing around in the back of my mind that is pretty closely related - I have made a post in the suggestions section.

[mrvn]

Thanks. It's nice to see old bugs getting attention too and aren't just swept under the carpet.

And thanks for the Rail improvements. Will try them out this week in LTN.

[ptx0]

Thanks. It's nice to see old bugs getting attention too and aren't just swept under the carpet.

And thanks for the Rail improvements. Will try them out this week in LTN.

how are they related to LTN?

[mrvn]

Thanks. It's nice to see old bugs getting attention too and aren't just swept under the carpet.

And thanks for the Rail improvements. Will try them out this week in LTN.

how are they related to LTN?

LTN wants trains to go to specific train stops even if there are multiple stops with the same name. So it adds a temporary stop to the schedule using the rail tile the stop is connected to. The the schedule switches to the real stop and, oh my, the train is already at the stop, lets use that one.

Before 1.1.43 schedule entries using rails can approach said rail from either direction. So the train can arrive where the stop is from the wrong way. Its front is exactly where the stop is but facing the wrong way. So when it switches to the real stop it picks the nearest train stop with the same name and often making a loop to approach the train stop LTN was aiming for isn't the closest. Trains end up at the wrong train stop.

With 1.1.43 you can no ask the train stop what direction it has (simpler than figuring it out yourself) and specify that in the train schedule. Then the train will approach the rail from the right direction and the train stop will always have distance 0.

[Jon8RFC]

Looking forward to this potentially improving multiplayer performance! If computational wrongs were righted, clients running the same map significantly worse on multiplayer (versus locally) might not be as big of a deal now.