Chests work.

[zggz]

viewtopic.php?f=18&t=60344#p362465
If man in this post rights, can we make chest checking not so often? 1 time at 1 second will be enought for everyone.

It will add a lot of performance.

[bobingabout]

If you're going to make a sugestion, please write the whole sugestion in the topic. Posting a link for reference is okay, but not when the entire sugestion is "Do what it says if you follow this link." you have no idea how much of the topic you're supposed to be reading, and you drop us right in the middle of a conversation.

I have no idea what you're actually trying to sugest... so I'm going to take your sugestion as "Read this link", of which I will decline. (yes I loaded the page and had a quick scan.)

[zggz]

blue chests check recourses in them too often. Can we make 1 time in 1 second?

[quyxkh]

I'm almost certain the engine doesn't poll chests, only when something gets put in them. The post you're referring to even says it's making "wild guess"es. Testing says inserter swings are expensive, sleeping inserters are dirt cheap, chests are fine.

Reduce the number of inserter swings your map generates. You do have to unload assemblers fast enough that they don't spend too much time with 3 or more cycles' worth of product in their output buffer -- max-boost A3's running high-speed recipes like green chips want a pair of staggered fast inserters doing their unloading to keep ahead of the game‚ but get as close to direct insertion as you can get without multiplying the number of trains you need, and clock any inserter that would otherwise be swinging more often than about once per second where you can, is my rule of thumb.

[Rseding91]

blue chests check recourses in them too often. Can we make 1 time in 1 second?

That's not how any of this works. The logistic network doesn't poll chests to know what's in them.