Will trains receive wire signals in the future?

[split saddle]

I've trying to create modular train stations that do a lot of things: Train Merger, Train Paint Combiner, Train Painter and things like this, so that I can just use the blueprints and direct the correct trains to it.
my goal is to make a MAM using trains, but that will be a lot of work if trains don't have wire signals. So I wonder, will this feature be there? Like using signals to change the path of train.

[nimble ferry]

One thing i want to comment on is usually you want 0 internal trains in a mam, because each one adds at least 100 seconds of latency, which can really add up fast

But to answer the direct question, it's been requested a dozen times in a dozen forms, and I suspect the team is discussing ways to improve trains
We just don't know what form it will take, yet

It may be wires, it may be train recoloring, it may be something else entirely

[proven ravine]

yeah we have it in our list

We are not sure what the best approach is though

[split saddle]

Gotcha! And I had no idea trains would add delays in MAM (I never designed one), I was planning on preparing a lot things far away and just tweaking signals to choose which one to use. Like have one train track for each paint and then just choose a signal to choose which one should move forward.
I think some sort of line switcher would be the ideal. The real question is, how do we send signals on space if the number of global channels are limited?
My idea to solve this would be to either:

• Allow players to create channels on demand using either color code or a slug as the channel identification, this would allow more flexibility on setting up complex machinery.
• Allow to have wires on space, just like we have belts and tracks (IMO, this is too much of visual pollution)

[nimble ferry]

mam's are typically used for submitting the ROS (random operator shapes), and those need very low quantities of shapes

so production speed is often less important than the time delay between "new shape signal" and first delivery

so you could have a 12 belt mam that goes much faster than a 36 belt mam if the 12 belt is built for low latency

every train station you add needs at least 90 seconds to load and then however long to travel, per train, and thats on top of the time it takes for your mam to flush out the junk or previous shapes

so to reiterate, the reason that one might want to pay extra attention to latency in mams and only mams is because theyre constantly changing what theyre making

latency only ever matters for the "first shape produced", which doesnt matter if you have a factory that only produces 1 milestone shape forever

[buoyant palm]

Hmmm. My MAM starts with gathering ALL the basic shapes and colors and training them in to the front end of the MAM. The MAM just grabs whatever it needs (knowing it's already available) when the next random shape comes in. The 3 slots in the train loaders and unloaders serve as nice buffers in case the MAM ends up needing a lot of particular shape or color. Once the train route is filled I don't really care about this latency?

[nimble ferry]

Correct, those arent "internal"

That's continuous setup for the mam to run

[buoyant palm]

Balankin1 I noticed you said the ROS doesn't need a lot of shapes. I was kind of wondering about that. I'm striving to make a full 12 lane MAM... because I'm assuming the number of ROS shapes being delivered will continue to go up (like the milestones do). But if the ROS is always 1000 shapes, I might want to rethink my MAM. Space (in particular, reduced distance to the vortex) would be more important than assembly speed. Can you (or anyone else) give me a hint about what the future of ROS delivery counts look like? 🙂

[nimble ferry]

This is the total required shapes to get to the next level for each operator shape
You may notice that 2 of them (ros 0 and 1) are basically 0 for a long time by comparison

That flat line at the top is INT_MAX

[buoyant palm]

Thanks Balakin1, that's useful information!

[neat quartz]

( - 1000)

[nimble ferry]

yeah, vagues

[proven ravine]

for reference, thats to encourage exploring MaMs

(as they are more efficient to process operater levels)