Constructing large areas with eg. refined concrete, having 2 roboports mk2 & 50 construction robots, results in a weird behaviour regarding recharge operation.
Robots pick up tiles from my inventory, fly to their destination, place them down, come back to me to pick up the next tiles and so on. This draines their internal battery. At some point, carrying tiles and flying to the destination, they do a turnaround, not far away from their destination, to recharge, without having performed the job. Then, they fly the same way a second time to (finally) place the tile.
Isn't it possible to calculate if the remaining battery energy is sufficient to successfully complete the task without recharge? If the result is "no", then they should recharge before they fly out again to place something, even if there's (insufficient) energy remaining. This would speed up processes & provides against flying the same way nearly twice.
The same behaviour should occour with robots from stational roboports.
I think that's to prevent switching to "emergency fly mode" with massively reduced speed (no matter I can't imagine that something with a completely drained battery can still fly, but useless robots on the ground is surely not an intention).
[1.1.74] (Personal) robot energy calculation?
Re: [v1.1.74] (Personal) robot energy calculation?
This has been discussed before, and robots are UPS-lite precisely because they're stupid like this. Even more so with roboport-based robots; a robot that doesn't have enough energy to reach its destination would have to spend a lot (relatively speaking) of CPU to find a path through the roboports, and then do something sensible if a new roboport is added or an existing one is removed.
Re: [1.1.74] (Personal) robot energy calculation?
This is not a bug, make a feature request.
Re: [v1.1.74] (Personal) robot energy calculation?
I've searched but nothing found within the last 4 years. I apologize if I missed that thread. They can be merged if desired.
So the robots have this behaviour to keep UPS as high as possible? I think a short check if the energy is sufficient to place the desired tile shouldn't consume too much UPS, taking into consideration that a robot calculates its energy requirement "on the fly" anyway. So my thought is to bring exactly this calculation while the robot is over the character (or a stationary roboport) before it continues.[...] robots are UPS-lite precisely because they're stupid like this. Even more so with roboport-based robots; a robot that doesn't have enough energy to reach its destination would have to spend a lot (relatively speaking) of CPU to find a path through the roboports, and then do something sensible if a new roboport is added or an existing one is removed.
Is there an explanation why this isn't a bug? You mave move this to the desired forum.
Re: [v1.1.74] (Personal) robot energy calculation?
Everything you describe is working as intended.
-
- Smart Inserter
- Posts: 2768
- Joined: Tue Apr 25, 2017 2:01 pm
- Contact:
Re: [v1.1.74] (Personal) robot energy calculation?
I think you misunderstood the statement. Just because it's working as intended doesn't mean there isn't an idea that might be better. It just means it's not a bug.
viewtopic.php?f=6&t=103400&hilit=robot+charge
My Mods: Classic Factorio Basic Oil Processing | Sulfur Production from Oils | Wood to Oil Processing | Infinite Resources - Normal Yield | Tree Saplings (Redux) | Alien Biomes Tweaked | Restrictions on Artificial Tiles | New Gear Girl & HR Graphics
Re: [v1.1.74] (Personal) robot energy calculation?
The statement was clear: it is "intended".FuryoftheStars wrote: ↑Sat Feb 25, 2023 12:25 amI think you misunderstood the statement. Just because it's working as intended doesn't mean there isn't an idea that might be better. It just means it's not a bug.
Intended functions have the status to me "leave it as it is" so I don't want to bother.
If bots retire half of their job without success, then this is a bug to me, especially to an ambitious & interesting game with an eye for the finer details like this. It's, of course, impossible to turn this to a perfection because perfection never exists in real life, but some basic points should be consistent, IMHO.
Oh, wow, thank you for this impressive thread. I'll investigate this another time in detail, but a short "skim through" seems to be exactly my request.