Friday Facts #189 - Specifying the 1.0

[doxsroxs]

I dont know if this has been mentioned yet, but Factorio is really missing one important feature!

• The ability to automate trains!
• Unique station IDs given to all stations when placed
• Ability to read station ID from station using combinator
• Ability to send train to station ID using combinator connected to station (without the need for a trainschedule etc.)

Please help us automate one of the biggest things in the game, it will be glorious!
Most of this is already available in Smart Trains, but the above additions would make vanilla game soo much better!

[Sharkking]

I really like the cyberpunk like style that came with the new items and machines. I hope all of the game content will be redone with this style.

keep it up

[Nemoricus]

My one big wish for Factorio 1.0 is more ability to automate trains. In an ideal world, I'd like to have networks detect that there's a shortage in transportation resources (too many resources being stockpiled and not enough consumed at other locations would do), and then build additional trains to transport those resources.

More modestly, I'd like trains to be able to decide their destinations based on conditions like iron exceeded a certain threshold at a mining outpost. Then they'd look for a destination with a low amount of iron. Something like this would help train networks to scale with the size of a factory, and reduce the need to change a train's destinations when adding new outposts and removing depleted ones.

[RobertTerwilliger]

Every single word about nerfing science and crafting times is absolutely fair. However I imagine very little number of players needed those nerfs, because, as I remember "ye olde" blue science: after you had automated it - nothing was hard past it : )

[cid0rz]

I dont know if this has been mentioned yet, but Factorio is really missing one important feature!

• The ability to automate trains!
• Unique station IDs given to all stations when placed
• Ability to read station ID from station using combinator
• Ability to send train to station ID using combinator connected to station (without the need for a trainschedule etc.)

Please help us automate one of the biggest things in the game, it will be glorious!
Most of this is already available in Smart Trains, but the above additions would make vanilla game soo much better!

Agreed, i'd love to have Smarttrains-like functionality in the game, I understand it is and additional feature but I think it really adds value

[gmliquidmedia]

Greetings, I like the idea of the Special mode where you start with power armor MK2, personal roboports and construction robots. last week I did a speed run of 0.15.5 with blue prints that I had planned out before starting the run, and went for the construction bot as soon as i could get them. I did the rail world start with a seeded map that I had picked before the run.

Here is the video.
https://youtu.be/JPxiu4F3IWY

Thanks for making an awesome game.

[LuxSublima]

"...train and (probably) the spidertron."

What does this mean? I get images of this thing in my head...

https://media.giphy.com/media/3oEduVeIg ... /giphy.gif

Please let it be this.

Pretty much, yes.



https://www.factorio.com/blog/post/fff-120

WOW that's... terrifying, and awesome!

[basementjack]

@klonan My thoughts for the 1.0 release:

The game is actually perfect the way it is right now (once bugs and maybe a few last minute ratio tweaks are ironed out)
The selling price is entirely too low ($20) and I worry that once you release 1.0, everyone who wanted the game has already paid $20 for it. I remember minecraft in it's early days and the price of the beta went up steadily over time.


For a final product, it needs more Campaign missions.
The sandbox environment is great, but for a v1.0 game, I think it'd be good to have something like 10-20 hours of structured gameplay tied to specific missions.

I'm a relative newcomer (0.13 was my first) and I started with the campaign - it guided me and taught me some basics, but it always felt incomplete. Those few missions were fantastic*, and I wanted more before I delved into a 'free play' mode.

In closing, when I bought .13 I was surprised it wasn't already a 1.0 product. Aside from feeling like the product lacked missions, I had no complaints about the sandbox, and the product felt 'finished' to me.

I would also really like to see the finished product on the Mac App Store - if for no other reason than the extra publicity you would get from that (they feature games almost weekly, and I have to imagine that you'd get fantastic reviews there.)

If you release it there, I'll buy another copy just so I can leave a 5 star review.

- Jack



*Level-04 in new hope was not fantastic - it was WAY too hard for level 4

[Albrat]

Alot of text removed.

In your system the bots would still clog. You can't guarantee they won't in all cases. Your condition is 60% used, so you could have the other 40% sitting in that exact roboport you are trying to insert more into.

I also don't play the way you play. I don't want to increase bots without bounds when 60% or more are in use. I want to set a fixed number of bots and make sure there are always that many in the network. If extra get released by players then I want to remove them. I want to increase the bots only when I feel my production is adequate or I need more bots (my decision based on various factors). If I do a large tree deconstruction and it uses all 3000 of my bots for 5-30 seconds, I don't want to pump in another 200 or whaterver just for the 30 seconds of overuse once in a blue moon.

Ahhh. Then ignore the available and use the max bots feature and remove construction bots directly from your supplier.
I have played the same style before, so I understand what you mean. Place a single filter inserer with a circuit condition that is Construction robots > 3000 activate. Filter it to construction robots only. Stack size limited to 1. (newest feature). This way it will slowly remove the extra bots down to your max value. Since it is filtered it will only remove constuction bots, though over many hours of gameplay it could get stuck eventually (you could use the same system on other roboports to remove excess into a Active provider chest then have a active requester chest beside your inserter port... This way excess bots are taken back to the source.

[Albrat]

My one big wish for Factorio 1.0 is more ability to automate trains. In an ideal world, I'd like to have networks detect that there's a shortage in transportation resources (too many resources being stockpiled and not enough consumed at other locations would do), and then build additional trains to transport those resources.

More modestly, I'd like trains to be able to decide their destinations based on conditions like iron exceeded a certain threshold at a mining outpost. Then they'd look for a destination with a low amount of iron. Something like this would help train networks to scale with the size of a factory, and reduce the need to change a train's destinations when adding new outposts and removing depleted ones.

I agree that we need more control over trains... But i think having trains send and recieve circuit conditions would be better. Maybe a track section that can read a trains data as the locomotive runs over it. (like real railways have). This way we could have a system that adjusts signalling and other data to make the train go where we want.
Eg a coal train runs over the sensor track, the signal is set so it can only go into the coal train stacker, Instead of it trying to route around the stacker and get to the coal station.

I tend to stack full trains ready to run into the delivery point, there are "hacky" ways of making your trains go to stations with high load and avoid low load stations. (circuits and station averages / signals) But they are not 100% effective and only work in the later game. I have successfully setup a loading system before that took the average of all my outposts, calculated the average and then anything below the average closes. Anything above the average opens for loading. Trains were held in a "outbound" stacker untill the outpost had enough ore to offload onto a train. To make the stacker work correctly you needed a train stop at the end of the siding, signals along it and all trains had to enter inbound stacker / station / outbound stacker / outpost . Where Every outpost had the same name eg "Iron post" (note that CaPitaliSation in the name is important. iron outpost and Iron Outpost are not the same. ) As a outpost closed... the signal on entry would be set to red. A chain signal on the "main line" would stop the train entering the outpost, this would block the main line for 30 seconds or so... but then it would re-path and go to the next open outpost of the same name.

The chain signal would only stop a train trying to go into the outpost and not normal traffic trying to bypass the outpost. (stoppages are usually rare and occur when two trains depart to the same outpost.)

Setup and maintainance is a steep learning curve (pictures and videos may be handy to learn the system.) but once it is setup the system runs well. (you can close the system if you have too much Iron for example by setting the stacker exit signal to danger if iron plates are greater than x - Your upper limit for storage )

[factoriouzr]

Alot of text removed.

In your system the bots would still clog. You can't guarantee they won't in all cases. Your condition is 60% used, so you could have the other 40% sitting in that exact roboport you are trying to insert more into.

I also don't play the way you play. I don't want to increase bots without bounds when 60% or more are in use. I want to set a fixed number of bots and make sure there are always that many in the network. If extra get released by players then I want to remove them. I want to increase the bots only when I feel my production is adequate or I need more bots (my decision based on various factors). If I do a large tree deconstruction and it uses all 3000 of my bots for 5-30 seconds, I don't want to pump in another 200 or whaterver just for the 30 seconds of overuse once in a blue moon.

Ahhh. Then ignore the available and use the max bots feature and remove construction bots directly from your supplier.
I have played the same style before, so I understand what you mean. Place a single filter inserer with a circuit condition that is Construction robots > 3000 activate. Filter it to construction robots only. Stack size limited to 1. (newest feature). This way it will slowly remove the extra bots down to your max value. Since it is filtered it will only remove constuction bots, though over many hours of gameplay it could get stuck eventually (you could use the same system on other roboports to remove excess into a Active provider chest then have a active requester chest beside your inserter port... This way excess bots are taken back to the source.

I already do this. Our discussion isn't about how to do it. I know how to keep a fixed number of bots in my networks. The issue is that the game doesn't allow a full proof system to keep the desired number of bots because the roboport is the only way to add or remove bots from the network and it suffers from two issues. First, it can clog with inactive robots, and thus leave no room to insert more, and two, there is no way to request robots of a certain type to land there so they can be removed. For point two, all that would be needed is to request 1 construction and one logistics robot to land at a selected roboport from the active robots and stay stationed there (even forever would be fine). For point one, all that would be needed is an auto release slot on roboports where if an inserter inserts a robot into the port, it will always go into the auto-release slot (same if a player places robots there manually). These robots would immediately exit the roboport and then find one to land at if they are not needed. Landed robots would never land in the auto-release slots.

The only other way in the base game to try to make a system that doesn't clog is to put the circuit logic (with 2 requester and 2 active providers) at every single roboport. This is obviously not feasible because there can be hundreds of roboports in one network. My bases get pretty big. On top of that the inserters would remove and insert too many robots when a change is needed (at minimum 1*number of roboports) and thus it would constantly add and remove bots in a never ending loop. The way to work around this is to connect every single roboport by circuit wires. This is obviously a logistics nightmare and would require taking red wire across many power poles to reach all the roboports. This would also make one of the wires basically impossible to use for any other circuitry as your entire factory would be a red network, so you couldn't do isolated stuff. This could also cause performance issues and at the very least wastes cpu on doing unnecessary updates to the circuit network.

As a summary, there is no way in the current game to reliably automate inserting and removing robots from the network as there will always be an issue with clogging or not being able to remove the robots reliably.

[wwdragon]

After finally getting to try them, I gotta say these new personal laser defences are perfect!

The map is really nice too.

Only thing I don't like is the so called nightvision; I can't tell weather it's day or night when wearing them because of that no-green tint thing.

[wwdragon]

I notice in .15.9 that the enemy creeps are building bases RIGHT next to my walls.
Isn't there supposed to be a minimum build distance of 3 world-tiles away for them?

[kinnom]

I notice in .15.9 that the enemy creeps are building bases RIGHT next to my walls.
Isn't there supposed to be a minimum build distance of 3 world-tiles away for them?

No longer

[Nova]

Only thing I don't like is the so called nightvision; I can't tell weather it's day or night when wearing them because of that no-green tint thing.

Well, for most people the night vision with the green tint was pretty much useless.

Um, do you need to know if it is day or night? I thought that the basic feature of the night vision was to make it irrelevent if it is night or day. Then again I think it's still easy to spot the time of the day even with night vision.

[Ohz]

I don't like all of those balances, I always automate everything and don't carry much basic ressources. When I want to make a new oil oupost, I actually prepare carefully myself. For the science, I accepted everything as it is, I don't complain. But that's my opinion.

Talking about balance, I found it a bit weird you can research and make portable robotport without a single red circuit, and few other nonsense like that, I can't remember, I will make a post later on balance forum.

[trolltok]

@klonan My thoughts for the 1.0 release:

The game is actually perfect the way it is right now (once bugs and maybe a few last minute ratio tweaks are ironed out)
The selling price is entirely too low ($20) and I worry that once you release 1.0, everyone who wanted the game has already paid $20 for it. I remember minecraft in it's early days and the price of the beta went up steadily over time.


For a final product, it needs more Campaign missions.
The sandbox environment is great, but for a v1.0 game, I think it'd be good to have something like 10-20 hours of structured gameplay tied to specific missions.

I'm a relative newcomer (0.13 was my first) and I started with the campaign - it guided me and taught me some basics, but it always felt incomplete. Those few missions were fantastic*, and I wanted more before I delved into a 'free play' mode.

In closing, when I bought .13 I was surprised it wasn't already a 1.0 product. Aside from feeling like the product lacked missions, I had no complaints about the sandbox, and the product felt 'finished' to me.

I would also really like to see the finished product on the Mac App Store - if for no other reason than the extra publicity you would get from that (they feature games almost weekly, and I have to imagine that you'd get fantastic reviews there.)

If you release it there, I'll buy another copy just so I can leave a 5 star review.

- Jack



*Level-04 in new hope was not fantastic - it was WAY too hard for level 4

This comment makes so much sense to me. Content-wise it's really past the 100% mark. I find some of the comments here a bit disrespectful, when people who already put hundreds of hours in the game say that feature X or Y is missing and it's absolutely critically crucial for calling it a finished game, and the devs have to do it. I mean, I'd like to see automated mining, programmable train schedules and space platforms as well, but I also think that Wube has made an absolutely outstanding game already. They did make it very well modable so people can tweak it and add what they think is missing, and make it 100% perfect for themselves. It's not reasonable to expect the base game to be 100% perfect for everybody.

It makes me a bit sad to see that Factorio development might be approaching an end. But at the same time, thinking about Wube taking on a new project, with all the lessons gained and resources earned from Factorio, is maybe even more exciting than thinking about further Factorio development.

[vanatteveldt]

Like gaslamp games with clockwork empire?

[wwdragon]

I notice in .15.9 that the enemy creeps are building bases RIGHT next to my walls.
Isn't there supposed to be a minimum build distance of 3 world-tiles away for them?

No longer

Well that sucks!

Only thing I don't like is the so called nightvision; I can't tell weather it's day or night when wearing them because of that no-green tint thing.

Well, for most people the night vision with the green tint was pretty much useless.

Um, do you need to know if it is day or night? I thought that the basic feature of the night vision was to make it irrelevent if it is night or day. Then again I think it's still easy to spot the time of the day even with night vision.

The green tint was perfecly fine; it was easy to see everything and it was a colour based way to tell the day/night cycle.

And yeah, I do. I made a solar based setup and need to test it, taking out the old steam system at dawn.
I'm aware I can just take off the nightvision, but it highlights it's lack of visual uniqueness.

[5thHorseman]

The green tint was perfecly fine; it was easy to see everything and it was a colour based way to tell the day/night cycle.

I (and many others, hence the change) disagree with you on that one. I personally just didn't bother with night vision before, as it was so green.

Maybe a setting in options, ranging from "none green" to "munchkin land" would be nice, so we can choose for ourselves.