More infinite research: Belt speed/Train speed

Post your ideas and suggestions how to improve the game.

Moderator: ickputzdirwech

Post Reply
theorderofthings
Burner Inserter
Burner Inserter
Posts: 15
Joined: Tue May 30, 2017 12:32 am
Contact:

More infinite research: Belt speed/Train speed

Post by theorderofthings »

Just as the title implies, I think it would be a nice feature if we could upgrade belt speed and train speed just like we can upgrade bot speed.

Currently in factorio, there are three main logistical methods, viz. belts, bots and trains. All with their own disadvantages and advantages. And right now we can upgrade bot speed, but we can't upgrade belt speed and train speed. This is somewhat of a problem in the endgame with infinite research. While bots in the early game are usually better than belts for low-throughput production, with enough speed they start to become better at belts for medium-throughput items.

Now the bots vs. belts debate aside, I think adding train speed and belt speed upgrades would also make the game more interesting in general, as mining productivity and bot speed is the only thing I really see people research currently. Adding belt speed and train speed would give players more options and make their choices feel more unique.

As far as the programming difficulty and ups usage goes, train speed probably isn't too much of an issue since it's already implemented with the differences in rocket fuel vs solid fuel vs coal. Belt speed may be a problem though, since belts in general are bad for ups and it may mess up inserters if items go too fast (although, I'm not sure.) However, if it isn't too hard to implement, I think this would make the endgame more fun and much more varied.

User avatar
bobingabout
Smart Inserter
Smart Inserter
Posts: 7351
Joined: Fri May 09, 2014 1:01 pm
Contact:

Re: More infinite research: Belt speed/Train speed

Post by bobingabout »

My initial thought was "No, because they would effect all belts everywhere", but then trains... well, you already have that braking power modifier, so why not just a speed modifier too?
and if trains of a specific force can be modified, why not belts too?

So... Yeah, it does kind of make sense that it could be an option.
Creator of Bob's mods. Expanding your gameplay since version 0.9.8.
I also have a Patreon.

JohnyDL
Filter Inserter
Filter Inserter
Posts: 533
Joined: Fri May 16, 2014 3:44 pm
Contact:

Re: More infinite research: Belt speed/Train speed

Post by JohnyDL »

like the trains idea but the belts I'm going to suggest the same thing when infinite modules idea came up, a new item that the boost effects and leave the rest alone, infinite belts like infinite modules would initially be the same benefit as their little brother but there's a substantial extra cost to them, the modules I suggested 4 module 3s and some blue circuits to make a module infinity with this it could be similar, 4 belts/4 splitters/8 undergrounds + 100 gears + 100 lube to make their infinite versions. making them only end game and not even immediately after rockets except for bragging rights.

dragontamer5788
Fast Inserter
Fast Inserter
Posts: 154
Joined: Fri Jul 15, 2016 1:44 am
Contact:

Re: More infinite research: Belt speed/Train speed

Post by dragontamer5788 »

theorderofthings wrote:Just as the title implies, I think it would be a nice feature if we could upgrade belt speed and train speed just like we can upgrade bot speed.

Currently in factorio, there are three main logistical methods, viz. belts, bots and trains. All with their own disadvantages and advantages. And right now we can upgrade bot speed, but we can't upgrade belt speed and train speed. This is somewhat of a problem in the endgame with infinite research. While bots in the early game are usually better than belts for low-throughput production, with enough speed they start to become better at belts for medium-throughput items.
Note: Bots are practically limited by Roboport Charging stations. The 4MW per Roboport means that even with infinite research, you get no additional bandwidth from bots. You'll need fewer bots to saturate a Roboport, but its really not a big deal to be using "only" 3000 bots eating up 100MW of power... instead of 4000 slower bots still eating up 100MW of power. The primary cost to bot power usage is their Energy-per-tile mechanic. Faster Bots use up more power, which requires them to stop by Roboports for more charging. 4-Bots per Roboport and 4MW means... no matter what, bandwidth is limited by Roboports. Basically: Infinite Research (Bot Speed) is a trap... well... almost one anyway. It improves latency characteristics so its still useful, but it cannot increase the bandwidth of Bot-based networks.

So from that perspective, both Bots and Belts are equal in the scope of infinite research. Neither really benefits. With that said, Bots have the most bandwidth in the game at the cost of significant power-draw. The power-problem is solved by nuclear however. If you really want to increase the bandwidth of bots, you'll need to increase Roboport Charging Speed.

-----------

Long story short: All methodologies in the game have a constant bandwidth that cannot be increased by Infinite Research. This means that no matter how much infinite research you get, the optimal design for any factory remains the same. I like that.

JohnyDL
Filter Inserter
Filter Inserter
Posts: 533
Joined: Fri May 16, 2014 3:44 pm
Contact:

Re: More infinite research: Belt speed/Train speed

Post by JohnyDL »

or use more roboports, which is the common solution, if you have 1 roboport for every 20 requester chests you kill the amount of charging slots necessary especially if you limit the network size to the botted system and 2 train stations any long distance common stopping point add multiple ports and the robotics bandwidth isn't limited by one roboport per 50x50 tiles

theorderofthings
Burner Inserter
Burner Inserter
Posts: 15
Joined: Tue May 30, 2017 12:32 am
Contact:

Re: More infinite research: Belt speed/Train speed

Post by theorderofthings »

dragontamer5788 wrote: Basically: Infinite Research (Bot Speed) is a trap... well... almost one anyway. It improves latency characteristics so its still useful, but it cannot increase the bandwidth of Bot-based networks.

So from that perspective, both Bots and Belts are equal in the scope of infinite research. Neither really benefits. With that said, Bots have the most bandwidth in the game at the cost of significant power-draw. The power-problem is solved by nuclear however. If you really want to increase the bandwidth of bots, you'll need to increase Roboport Charging Speed.

The belt speed may need to be rethought, but the train speed research would match the bot speed research. Because train speed doesn't increase the bandwidth of train networks, since they are more limited by loading and unloading regardless.

Of course, if we get a bot charge/roboport distance research line, and a belt speed research line, then the train research line may need to be rethought, although I doubt bots and belts would ever take the place of trains even with infinite research.

dragontamer5788
Fast Inserter
Fast Inserter
Posts: 154
Joined: Fri Jul 15, 2016 1:44 am
Contact:

Re: More infinite research: Belt speed/Train speed

Post by dragontamer5788 »

theorderofthings wrote:
dragontamer5788 wrote: Basically: Infinite Research (Bot Speed) is a trap... well... almost one anyway. It improves latency characteristics so its still useful, but it cannot increase the bandwidth of Bot-based networks.

So from that perspective, both Bots and Belts are equal in the scope of infinite research. Neither really benefits. With that said, Bots have the most bandwidth in the game at the cost of significant power-draw. The power-problem is solved by nuclear however. If you really want to increase the bandwidth of bots, you'll need to increase Roboport Charging Speed.

The belt speed may need to be rethought, but the train speed research would match the bot speed research. Because train speed doesn't increase the bandwidth of train networks, since they are more limited by loading and unloading regardless.
You are somewhat incorrect.

True, train-stations are the limit. But (just like Roboports), you can add more and more train stations until you max out the bandwidth of the rail system. Where you go wrong however is that due to speed-limits and acceleration limits of trains, there is a limit to the number of train-stations you can add to a rail-line.

Consider that Trains move at ~250km/hr, which is roughly 70 tiles per second. Also consider that an iron-ore cargo wagon contains 2000 ore, which will be unloaded in just 6-seconds. Assuming big-enough rail stations (so that the trains are all moving at 250km/hr as much as possible), 3x 1-2 trains can be serviced every second to keep the trainline moving at max speed. This gives a rough theoretical limit of ~18 stations per rail-line to cap out a standard rail station.

If trains moved faster (say at 300km/hr, or 84 tiles per second), now 24 stations is the theoretical maximum.

-------------

In practice, trains are so big and have such huge bandwidth that most people never run into train concerns. And furthermore, traffic issues means that acceleration is the primary "cost" of sub-optimal trains. Still, faster train speeds could increase bandwith of a train-network if the network were so perfectly designed that speed were the only thing holding it back.

Not that I've ever seen such a perfectly designed train network, but I argue that its theoretically possible.

theorderofthings
Burner Inserter
Burner Inserter
Posts: 15
Joined: Tue May 30, 2017 12:32 am
Contact:

Re: More infinite research: Belt speed/Train speed

Post by theorderofthings »

dragontamer5788 wrote: In practice, trains are so big and have such huge bandwidth that most people never run into train concerns. And furthermore, traffic issues means that acceleration is the primary "cost" of sub-optimal trains. Still, faster train speeds could increase bandwith of a train-network if the network were so perfectly designed that speed were the only thing holding it back.

Not that I've ever seen such a perfectly designed train network, but I argue that its theoretically possible.
That's pretty interesting. I think it would still pretty nice if we could have infinite research that would increase the throughput of bots, trains and belts. I see the argument that not having infinite research increase bandwidth allows there to be one specific optimal setup, but I feel like that itself hurts the endgame. Having multiple (locally) optimal setups depending on your choices in game seems much more interesting than having one best setup that everybody copies. I think it would encourage more creativity and dramatically increase the variety of design in endgame bases.

And honestly, I also like unique spaghetti-fied bases, and the fact that belts don't scale very well makes me sad. I wish we had more base variety and more unique bases. Spaghetti doesn't need to be optimal, but it would be nice if we had more optimal choices.
bobingabout wrote:My initial thought was "No, because they would effect all belts everywhere", but then trains... well, you already have that braking power modifier, so why not just a speed modifier too?
and if trains of a specific force can be modified, why not belts too?

So... Yeah, it does kind of make sense that it could be an option.

Obviously the choice is up to you, but it would be an interesting addition to bob's mods, with the even faster belts.

Post Reply

Return to “Ideas and Suggestions”