Friday Facts #222 - Christmas avalanche

[Light]

What about nerfing standard train car storage too? It always struck me as odd that most trains in Factorio only have 1 or maybe 2 cars. Real life trains can have 100+ cars easily!

This is because you're not considering frequency and speed of material transportation.

It's not unusual to have several trains delivering material to the same destination at a rather rapid speed in Factorio. It's unlikely you'll run out of material before another train is already pulling into the station. Speed bonuses from fuel make large trains even more pointless when any size trains can supply outposts at a fraction of the time. It also helps with keeping stations uniformly the same length for easy blueprinting regardless of what's being supplied/delivered.

Compound that with actual trains which can take several days or weeks to get to its destination due to their slow speed, their costs to run and maintain, and material consumption speed or storage capacity at the destination location. Then trains being 100 cars long makes more business sense.

[Oktokolo]

Nobody would use fluid waggons just because they look nice.

I do so since they got into vanilla.
Barrels are superior to tankers in almost every aspect even before the removal of the split-tanks feature (that i never used). Barrels need more steel and some added boxing/unboxing at the stations.
But trains with barrel-filled cargo wagons are faster and can transport both, more liquid and more sorts of liquid, than trains with the same amount of tankers.
Limiting barrel production and filling is easy when using buffer chests (that exist at stations anyway) and circuit-connected inserters.

The reason to use tankers is that using them is more like you would do it in the real world. It just looks more right - and they add some variety to rail-based transportation.

I would appreciate even more different wagons for different types of cargo. We got box cars and tankers. And the box cars should keep beeing usable for any type of cargo (liquids after barreling) but should be the less efficient option when there is a more specialized wagon available.
What about flat cars for sheet metal, concrete slabs, tracks, lumber or poles.And maybe Hoppers for ores and stone.
The hopper may support special equippment for faster loading and/or unloading (special rail with a belt underneath for inserter-less unloading and special ramp for inserter-less direct loading).
That open cars with visible cargo would make looking at trains inside Factorio even more satisfying.

[leoch]

Agree with the smaller fluid wagon capacity. Barrel capacity should also be reduced. This mod has nice balance in my opinion:
Smaller Fluid Wagon & Barrel

I agree with others that trains wagons have ridiculously high capacities. But so do so many other things for their size: steel chests, cars and tanks, and the player inventory. Physically bigger chests (e.g. 2x2) might work well for balance. Changing stuff like this too much though would change difficulty a lot; e.g. at the moment it's easier (IMO) not to automate coal delivery to early smelters but just refuel 50+ smelters by hand from time to time. Maybe there should be a stack size adjustment difficulty setting?

[pleegwat]

I would appreciate even more different wagons for different types of cargo. We got box cars and tankers. And the box cars should keep beeing usable for any type of cargo (liquids after barreling) but should be the less efficient option when there is a more specialized wagon available.
What about flat cars for sheet metal, concrete slabs, tracks, lumber or poles.And maybe Hoppers for ores and stone.
The hopper may support special equippment for faster loading and/or unloading (special rail with a belt underneath for inserter-less unloading and special ramp for inserter-less direct loading).
That open cars with visible cargo would make looking at trains inside Factorio even more satisfying.

I so want to see this.

[GrumpyPhysicist]

Am I the only one who was manually compressing using arithmetic combinators?

I'm sad that having to manually compress belts is going away. I seem to be in the minority by pausing belts using combinators to compress them, but I found it an interesting challenge and much less hacky than using splitters or underground belts.

I would prefer a solution that simply set ticks per item to an integer, thus making simple pause-based belt compressors easier to perfect, but I suppose I'm too late to the discussion.

[sicklag]

.

[Vandroiy]

What about nerfing standard train car storage too? It always struck me as odd that most trains in Factorio only have 1 or maybe 2 cars. Real life trains can have 100+ cars easily! Making the player have longer trains matches better what a train should be, IMO. Right now they're sorta like semi-trailer trucks on tracks.

I use 8-wagon trains, but going higher has diminishing returns 'cause it gets more difficult. IMO, long trains are rare for three reasons:

• Factory section too small. (player can fix this )
• Out of good options for station/track design
• Penalty for using locomotives in both directions

Express belts can't tunnel through wagon-to-belt offloading stations in volume, and train tracks can't cross without blocking each other. While some balancing in belts' favor would be nice, simply nerfing trains now could cause unwieldy train stations and track networks.

It could work better when combined with some more pseudo-3D features, like train tunnels/bridges or an extra underground belt type, to help with building train stations.

Am I the only one who was manually compressing using arithmetic combinators?

There are some people who did that. But really, most players just want to build a big factory, and adding circuit network stuff everywhere won't be too great for performance, newcomers, or looks. In most cases, people would just skip compression and belts would look and perform worse. There should be better places for this kind of complexity than basic belt mechanics.

[uint]

It could work better when combined with some more pseudo-3D features, like train tunnels/bridges

This is so necessary. As for nerfing train car storage, that would be good, but I believe regular storage chests should also be nerfed in some way. Right now chest capacities are absurd for their size. Maybe cut them to like 1/5-1/10 and introduce some larger-size, bigger-capacity lategame storage unit/robotised warehouse. Factorio is about the flow of goods, not hoarding, anyway.

[DaemosDaen]

My only issue with your statements is that you're considering breaking fluid wagons when compared to barrels. If you rebalance wagons, you will NEED to rebalance barrels.

I can live easily with just a one fluid per wagon tank.

Just need to stress this: REBALANCE BOTH BARRELS AND THE WAGON.

[Triaxx2]

So, there's two things to consider: First, it's a lot faster to unload from a tanker, than from a cargo wagon. Not only do you have both inserters and belts moving to deal with, but then there's the time the factory takes to empty and eject the barrel. If you don't have enough space on the belt to go out, you're slowing the operation down while you wait on the belt. Second, you're intending on removing the tank splitting for because the UI improvement can't handle it. Sounds like you need to take a hard look to make sure it's actually an improvement if it's preventing you from keeping features. It might be prettier but if it's not functioning the way you want, something is wrong. Function should always trump form.

Two options for the fluid wagon: Two tiers. Mk1 has three 15k fluid tanks that are always separate, and an Mk2 which has one massive tank, that can't be split but hauls more than two of the Mk1's.

Alternately, modular flatbed wagons. Three slots, function changes depending on what's installed. Empty slots, carries vehicles. Steel Chests in the slots? Carries items. Fluid tanks, carries fluid. Accumulators? Charges off of a network and can discharge at a distant one. Solar panels turns it into a mobile power producer, so you can set up new outposts without bringing house power, say if you're just setting up a temporary one as a safe place to attack biter nests from.

[xng]

Selling over a million copies of a game makes you one of the Big Kids too, Wube Software! =)

[rldml]

So, there's two things to consider: First, it's a lot faster to unload from a tanker, than from a cargo wagon. Not only do you have both inserters and belts moving to deal with, but then there's the time the factory takes to empty and eject the barrel. If you don't have enough space on the belt to go out, you're slowing the operation down while you wait on the belt. Second, you're intending on removing the tank splitting for because the UI improvement can't handle it. Sounds like you need to take a hard look to make sure it's actually an improvement if it's preventing you from keeping features. It might be prettier but if it's not functioning the way you want, something is wrong. Function should always trump form.

You can make an unload station that way, you unload fluids faster as with a fluid wagon. It's just a mattor of your station design.

edit: after i thought about it, this comment is nonsense. Nothing is faster as unload from fluid station. You have to sum up the until the transported fluids into your pipe-network. If you`re using barrels, the barrels have to be emptied first, before you can use the fluids.

Long story short: you're simply right.

However, the fluid waggon has to be more effective with the higher fluid capacity. This is the only way it makes sence to me after all.

[greaman]

So, there's two things to consider: First, it's a lot faster to unload from a tanker, than from a cargo wagon. Not only do you have both inserters and belts moving to deal with, but then there's the time the factory takes to empty and eject the barrel.

Nonsense.

I can unload with 12 inserter per waggon, 12 barrels per load, meaing 144*250 liters per second practically (euqlling 36k/s versons 36k per sec for three pumps).

It just takes a littel more space.

[mathturtle]

Merry Christmas everyone! Devs, enjoy your time off with your families. No need to rush out the next bugfix release.

I'm excited that belt compression from sideloading is back, especially sideloading two yellow belts to make a red. I will echo everyone else to say that if you adjust fluid wagons you MUST nerf barrels. Barrels (at least trains of barrels) are too powerful compared to fluid wagons already. I'm disappointed that you are removing the split tanks of the fluid wagon, I used it occasionally when I only needed a little bit of a fluid or needed to transport multiple fluids somewhere. But I don't think it will break any of my main designs...

Again thanks for all the hard work and enjoy your holiday! You have my vote for the 'Haunts my Dreams' award.

[3trip]

I would ask you to take the time to allow seperate tanks, since I find it very convenient to load smaller amounts of fluids such as lubercant and sulfuric acid in a single car.


Why do we even have barrels now? Even if they're technically better, why bother with the unnecessaryly complex, superfluous and unintuitive method of transport via barrels? can we just keep the fluid car, (maybe lower its gross weight) and nix barrels all together?

[agmike]

Nonsense.

I can unload with 12 inserter per waggon, 12 barrels per load, meaing 144*250 liters per second practically (euqlling 36k/s versons 36k per sec for three pumps).

It just takes a littel more space.

Barrels only stack to 10, so stack inserters can only move 10 barrels per swing, which makes it 3.03s per full wagon or 33k/s.

[pleegwat]

Barrels only stack to 10, so stack inserters can only move 10 barrels per swing, which makes it 3.03s per full wagon or 33k/s.

And it is the full wagon that is key here. 40 stacks of barrels means 40 swings to move everything out, plus 40 swings to move the empty barrels back in. If you store fulls and empties in the same car, and you use the entire car capacity for both, you'll probably be using 6 inserters in and 6 out. 40/6=6.66 swings, round up makes 7 swings to unload with 6 inserters. Loading of empties can happen at the same time and should take another half-swing once the final slots in the wagon free up, so 7.5 swings to swap out the entire wagon contents. The wiki gives me 0.417 seconds for a full swing, or 3.125 seconds to swap out a wagon. You'll need fewer swings if you use fewer slots, of course. Notably using only 30 slots means you need 5 swings to unload, plus half a swing to finish loading, or 2.294 seconds.

For fluid wagons, on the other hand, the wiki gives me 200 per tick or 12k per second for a single pump. Using three pumps 25k/(12k/s)=2.08s and there are no empties to load in, so you're done there.

Honestly though, I think the time it takes to get that train out and the next in will be much more significant?


EDIT: I just realized if you use less than the max 40 slots, but not by blocking out the remaining slots but using circuit network or filter tricks, you should be able to make it so the final half-swing to put empty barrels in is not needed. For 30 slots, this means you can load in exactly 5 swings per inserter, which comes to 2.08s, so in the ideal case barrel wagon and fluid wagon take the same time to unload the same amount of liquid.

[agmike]

And it is the full wagon that is key here. 40 stacks of barrels means 40 swings to move everything out, plus 40 swings to move the empty barrels back in. If you store fulls and empties in the same car, and you use the entire car capacity for both, you'll probably be using 6 inserters in and 6 out. 40/6=6.66 swings, round up makes 7 swings to unload with 6 inserters. Loading of empties can happen at the same time and should take another half-swing once the final slots in the wagon free up, so 7.5 swings to swap out the entire wagon contents. The wiki gives me 0.417 seconds for a full swing, or 3.125 seconds to swap out a wagon. You'll need fewer swings if you use fewer slots, of course. Notably using only 30 slots means you need 5 swings to unload, plus half a swing to finish loading, or 2.294 seconds.

For fluid wagons, on the other hand, the wiki gives me 200 per tick or 12k per second for a single pump. Using three pumps 25k/(12k/s)=2.08s and there are no empties to load in, so you're done there.

Honestly though, I think the time it takes to get that train out and the next in will be much more significant?

You are right, I forgot that in any practical design you will need to load empty barrels in too. With unloading times like this station throughput is basically only limited by train movement indeed. I also doubt any reasonable design could use full throughput of such unloading station.

[Drury]

What about nerfing standard train car storage too? It always struck me as odd that most trains in Factorio only have 1 or maybe 2 cars. Real life trains can have 100+ cars easily! Making the player have longer trains matches better what a train should be, IMO. Right now they're sorta like semi-trailer trucks on tracks.

People don't build short trains just because they carry a lot of items. Network size is also a consideration. Even the biggest of megabases just don't sprawl so far apart as to make long trains feasible - there's literally no room to put them without gridlocking the rail system.

[golfmiketango]

If inserters can fully compress a belt it will probably no longer be sensible to build large belt balancers. Not sure they're really all that sensible, now, either so I guess I'm OK with that... actually, thinking about it, it's not hard to think of ways a lot of silly things people (including myself) sometimes do with belt balancers can already be done with the right combination of inserters, circuit networks, and splitters, and I suspect that, beyond a certain scale, those designs would tend to be more space-efficient.