Friday Facts #129 - The late game

[tehroach]

Please don't remove the barrels from the game!

It would completely remove the need for interesting designs like this one

[...]

It is not that I am against the adding of the fluid tanker, BUT in a way adding the oil tanker goes against some of your own philosophies about Factorio!; The same philosophies that you have used to discredit the introduction of other suggested modules, as it trivializes a function that can already be achieved by other components in the game.

The problem is people want to have the tankers not only because they want to transport crude oil but also different fluids.

OK this is NOT really a problem with Barrels, it is simply lacking from the game in general.

The tankers could do that perfectly.

Again Barrels could do the task equally as well if NOT better, remember there is currently no tankers so we don't know how the tankers will connect to the network, so we will have to wait and see;
As for the tankers being able to do this perfectly, unless you know something that I don't know, I can't see how you can come to this conclusion

Barrels on the other hand would require them to add 7 barrel types, one for each fluid type, severely cluttering the inventory/crafting menus. Nobody would like that.

Barrels have their own line in the crafting menu so adding 7 more would not affect the placing of any other type of item, so sorry your point here is moot,
However adding the 7 extra Barrels would solve the liquid transport problem perfectly fine

And that's why barrels should be removed because barrels are much more like a temporary solution until we have tankers and after they are implemented we don't really need 3 options to transport fluids

Don't get me wrong, I can't wait to see how they implement tankers, because being able to run massive banks of steam engines in the middle of deserts that are supplied water by rail really appeals to me.
But why remove an item that already solves the problem practically and elegantly, IMO having the extra option is better.

especially if barrels would be limited only to crude oil. Tankers and Pipes are already enough to deal with fluid transportation. Just get rid of the barrels.

Remember
*Pipes flow slows down over distance
*Tankers haven't been implemented yet (So we really can't say if your above statement is True or NOT)
*Without barrels you completely lose the option to transport fluids as an item.

It would simply be a poor game design choice to limit barrels to crude oil only and allow tankers to just top up with any fluid they liked, just think of the last time you put Coke in a glass that someone has just used for milk.
At least you could imagine the barrel being cleaned first when it went into the barreling factory

If anything I would rather see that they limit the tankers to only accept the fluid that they were first filled with. ie once you start to fill a tanker with crude, that tanker can only accept crude until the player physically removes and replaces it. Just imagine one of the many times someone has accidentally filled their vehicle with the wrong type of fuel, the liquid left in the tank becomes useless for either of the original fuels purposes (or at very best a very low grade) and often just becomes waste (or if services are available, requires re-refining).

TL:DR
Tankers would be a good option to Barrels, but can't replace barrels entirely as Barrels posses properties that simply can't be replicated via tankers.

PS.
To the developers if you want to make trains great we need options for how we can construct our train stations!
Your idea for the loaders is great, but this viewtopic.php?f=93&t=17295&hilit=warehouse is better

[Fatmice]

The barrels should be expanded to contain other liquids and not just crude, imo. Without them, liquid processing is bottlenecked by pipe throughputs. Late game, I always barrel the oil up and move them by robots if I want high throughput, without which this would be impossible.

[DuhDavid]

I've had the game a few days now, so I'm not a veteran player. I have watched some of the older videos though where the game had a zerg event during completion. I thought that was an interesting concept, but I think it would be cool if you reimplemented the drop zone, but made it give new technologies. For instance, you have to accumulate the huge amount of resoursces in order to place the drop or contact "home" to get a drop and then the zerg happens everytime a new tech is incoming and it gets more difficulet with every drop. Of course, one would have to first shoot your rocket and get a satelite in orbit.

I also think that once the satelite is up, we should get a huge clearing of map cover, due to the satelite, replace those radars or whatever and maybe have ot launch another form a different loc to get an even greater clearing.

I like the mining idea, but I think they should be nearly infinite and more rare, but only provide a certain amount of space ofr x number of drils which only allow x amount of resource. The endgame really needs something to push it on and make us build more, because I want to build more.. a lot more.

[DaveKap]

I just wanted to give my input as someone who has beaten the game within version 12 and did NOT build a megafactory.

Honestly, I want more incentive to build a megafactory. I do think the ideas you've outlined are very good ones and will definitely make the building of a megafactory a more attractive proposition. However, we definitely need more rewards at the end of the tech tree to make building a megafactory necessary. You mention the endless technologies but I don't think that's enough. I think a good idea would be an expendable object (kinda like the attack bots) that has some oomf behind it. Imagine, for instance, being able to nuke a large section of the map. Or a large, expensive teleporter building that could both teleport the player or take in objects to teleport, except each "teleport" would cost a "charge" of electricity that's based on distance between teleporters. With both of these ideas, you can "expand" your megafactory appropriately; trains would still be preferable long-distance method unless you build a gigantic power plant for teleportation and nukes would clear out massive regions of space for expansion. Perhaps a "map blueprint" which lets you blueprint from the map instead of the game so you could copy/paste massive chunk for megafactory expansion. Of course, filling in lakes/rivers/oceans would have to be a part of that as well.

Basically, I think you've got good ideas, but I want a fun incentive and helpful reward to have the fun of building the megafactory.

[Fentus]

I had some fun fixing the issue I've been having... I kept setting up and destroying trains endlessly for ore and that became the end game... So I made this mod in response to fff:
viewtopic.php?f=179&t=21354

[MeduSalem]

[huge post]

You forget that the Factorio dev's try to follow a certain rule: They don't like to add items that are either extremely specialized or would clutter the menus.

That's why they will probably NEVER add 7 barrel types, one for each fluid. It's an insanely specialized solution due to a special item being required for each individual fluid. It doesn't matter if there is currently enough room to place 7 items in the menus, it DOESN'T scale well if they ever decide to add more fluid types because it eventually clutters the menu. So it breaks above rule in both points.

The tanker is a more general solution. It removes 2 existing items (empty barrels/crude oil barrels) and adds 1 tanker item which is able to be used with every existing fluid and any fluid the devs might come up with in the future and it also helps mods that also rely on additional fluids.

I'll quote a design rule here:

General Solution > Specialized Solution

The barrels should be expanded to contain other liquids and not just crude, imo. Without them, liquid processing is bottlenecked by pipe throughputs. Late game, I always barrel the oil up and move them by robots if I want high throughput, without which this would be impossible.

In my opinion we shouldn't remove certain throughput bottlenecks from the game. Some of them are there to put the player in front of design challenges. Without those design challenges the game would become even more boring towards the end than it is already due to lacking endgame.

It's a reason why I am even arguing that robots should be so expensive to use that it is only reasonable to do it for certain applications and not as one-solution-fits-all. The problem with them is that they break the game mechanics on a fundamental level because they basically bypass any balancing that is in place. Design challenges are basically the "difficulty" of Factorio. If there is almost no difficulty it eventually becomes a meaningless affair.

Let me ellaborate with an example based on an example game:

Imagine Factorio were a space 4x strategy game instead. At first everything is challenging because you aren't far advanced in the tech tree. The forces are equally distributed between the empires. But as you progress through the game and gain technological superiority you eventually come to your doomsday tech and once you researched your Death Star you wander off in the galaxy and start to annihilate everything with a single blast of the Death Star. It may be fun the first time around but after a while you realize that it is boring and unbalanced as hell because basically you are squashing ants with a railgun.

That's why most of these games never implement Doomsday tech that causes a major powershift and difficulty relief. It renders gameplay imbalanced and boring. And if they decide to implement such a thing these kind of super-weapons have an upkeep attached to them that outclasses anything else found in the game by magnitudes, often meaning a player is only able to power one such entity in the entire game.

That said I don't think we should only have one robot but that there should be an element to robots that makes scaling the number of robots increasingly more difficult.

Difficulty Scaling

Difficulty Scaling.png (12.73 KiB) Viewed 9314 times

May be a bit exaggerated but it hits the nail on the head. The game starts out more difficult than the average game, but at the point of reaching robots/solar power/laser turrets you basically won and there is no real challenge.

[tehroach]

[huge post]

You forget that the Factorio dev's try to follow a certain rule: They don't like to add items that are either extremely specialized or would clutter the menus.

That's why they will probably NEVER add 7 barrel types, one for each fluid. It's an insanely specialized solution due to a special item being required for each individual fluid. It doesn't matter if there is currently enough room to place 7 items in the menus, it DOESN'T scale well if they ever decide to add more fluid types because it eventually clutters the menu. So it breaks above rule in both points.

The tanker is a more general solution. It removes 2 existing items (empty barrels/crude oil barrels) and adds 1 tanker item which is able to be used with every existing fluid and any fluid the devs might come up with in the future and it also helps mods that also rely on additional fluids.

I'll quote a design rule here:

General Solution > Specialized Solution

Yer I don't think that rule really applies here, ie what if they decided to add a different ore?
would the rule apply then? No.

Fluids are a general item in the game, same as ores etc

[MeduSalem]

Yer I don't think that rule really applies here, ie what if they decided to add a different ore?
would the rule apply then? No.

Fluids are a general item in the game, same as ores etc

Adding a new resource that adds completely new technology/production cycles/gameplay that don't exist yet OR adding redundant items (7 types of barreled fluid even if 1 tanker could do them all) are two different kind of things.

[funktapus]

People like "end games" that produce good stories. WoW is a perfect example of an end game with awesome stories: not necessarily because of WoW's lore, but because players who did difficult end-game raids always had amusing stories to tell about them. Other games have really chaotic and exciting end-game PVP that makes for good stories, or really difficult challenges/missions, or entertaining "end scenarios" that result from the sum of a player's actions. That last one is evident in lots of RPGs where players sometimes compete to see who can be the most evil and leave the in-game world in the most sorry state possible.

What can Factorio players sit around and boast about? Right now we have "rockets per minute", which is a really one-dimensional (boring) metric. Players can share images of their enormous factories, but that also is just a testament to how much time a player can spend in game. I think the level of dedication that a player has can make for a good story, but there should be some additional layer of in-game feedback. Maybe the game should acknowledge when players have gone overboard, and push back against them harder and harder the larger they build. We already have that with biter evolution, but there could be more larger and more disruptive responses (such as the entire planet "getting sick") that players must come up with creative solutions to overcome.

Or maybe players could elect to research a number of "doomsday technologies" that lead potentially lead to crazy in-game consequences (almost like disasters in SimCity). They could research a giant robot that eradicates huge numbers of biters, but may turn on the player and start to destroy the factory. Or they could research a terraforming device that turns everything green, but then might lead to the evolution of a hyper-invasive plant that starts to clog up the factory. Or maybe the factory becomes self aware, spits you our on the other side of the map, and then becomes an enemy faction that sends robots after you.

I like to thing of my factory as a living thing that I created. Watching it go awry and turn on my could make for a really cool story.

[Rainhate]

Instead requiring players to hunker down in the late game with longer lasting mining sites(which i am excited about), why not also give the player the choice to embrace these fleeting ore sites with a less permanent mining gear.

Right now you have to run utilities out to the site, which takes up time and resources on a temporary site and the power poles are another possible point of failure in the system. Lets remove that requirement with Explosive powered mining equipment!

This mining equipment would use explosives to extract ore at amazing rates at the cost of lower total yield. i.e site has 500 copper ore after 2 blasts you get 250 ore and exhaust the site. This temp set up would be a great tie-in for the Loader in FF#128. Explosive Miner ->chest->unloaded-> belts-> loader->cargo train. This would allow easy set up along a rail line to grab some extra resources.

Pros:
Extreme mining rate
No Power required
Letting you clean up that 1 pesky stone deposit in you base in a timely manner so you can build/pave over it.

Cons:
Not realizing your full extraction potential.
No Power means, inserters and laser turrets have no backbone
irregular supply of ore/materials over time, due to the fast extraction nature.

This is about giving a few options to the player when setting up mining camps. Sometimes you stumble on a large rich deposit and you want to set up for the long haul, other times you find a small deposit next to some other part of your base that you would like to capture some of the resources but the small size of the deposit cannot justify the time investment of a full mining set up.

Of course the numbers and rates are not set in stone and i would appreciate people fleshing out this idea.

[diego-fm]

About the mining improvement.

I like the big excavator idea (viewtopic.php?f=93&t=12898), if it's more than a simple big mining drill. I've read all the posts, and thought about it for a few days, and i think a good improvement can be made with little effort (i'm algo a programmer, and i really think it can be done easily enough), combining it with the "dirty ore".

So, how about a BIG mining drill (like the mod one, or bigger), with a big mining area, huge power drain (multiple MW), but instead of "precision" like de normal mining drills, uses a more "brute force" approach, meaning that it only outputs dirty ore, but at a much higher pace. It also produces a lot more ore for the same field in the end.

The dirty ore would also require especial processing to turn it into plates, in a more efficient way, but given the drill output rate it encourages the player to design the infrastructure to support that throughput.

To put some numbers (none of them final, of course): A big drill, produces ten times more "dirty ore" than a normal drill for the same ore field, and for processing you could turn 10 dirty ores into 5 plates, effectively making it 5 times more resources, but at the cost you would have to process 10 times the input throughput.

But i think this requires a few things, in order to work in a sane manner.
1: the previous FFF loader, in some way, to allow the load/unload of trains with this throughput.
2: The big miner should be able to output to BOTH sides of a belt, if facing away from it (could integrate the belts in the miner sprite itself, they are basically the wheel and belts, plus all the things needed for this two things to work), and i'm thinking fully compressed yellow belt output rate.

The idea is that you only have 1, to 4 miners in each patch, so the logistics of each one being able to fill a yellow belt wouldn't get too messy. With the loader you could have a full mining outpost with no inserters (miner->belt->splitters->belt->loader->wagon), but the miners would only work when the train is at the station loading. You should put some sort of buffer to allow them to work fulltime (which in itself is no joke, given the thoughput involved, unless robots, which leads to maybe this miner can't output directly to a chest, only belt??)

To sum up, this idea would require: new entity(the miner), new items (the dirty ores), and new recipes (could be done in the chem plant) to process the dirty ores into plates. The miner needs to have "special" behaviour, in the sense it's not just a normal miner, it need to output not the ore, but the dirty ore (could be trivial, or not, depending on the code, my guess is fairly simple). If you want to make the process of the dirty ore a bit more complicated, you can make a few more recipes, make the chem plants "wash" the ore, and from 10 dirty ores produce 5 dusts, and each dust can them put into the furnaces to make one plate, that way our furnace setups are not useless at that point.

This would allow the resource patches to last longer, while not making them infinite, but if the player wants that, it has to design a system that can support that high volume of products logistic problem).

Also, talking about the loader idea. I would like to see 2 separate entities, the loader, to belt->wagon(maybe also chest) loading, being a simple upward sloped belt (1x3?), and the unloader, as a hopper pouring into a belt underneath, but with the need of insertes putting stuff into it. Low "internal inventory" like 2 or 3 stacks. Realistically the hopper could be 1x1, but if you want to fully saturate a blue belt i guess it needs to be 1x2, even with the inserter stack bonus. It would be awesome if it could be placed inline with other belts, as it's simply a hopper over a belt, if there is already something in the belt it can't place its thing (like an inserter)

Obviously the miner idea is not excluding with the mining drones idea, those being the lat-ish-game replacement for the normal mining drills. Easier to set up, place a robominer control station next to a ore field, give it some robots, and the robot roam around the ore field, mining, and when it's internal inventory is full, go back to the control station and dump it there, and the control station has an internal inventory that can be accesed by inserters (maybe even and output location like normal miners, although the output would be in batches) and acts as a active/storage chest, so the logistic robots can move the new ore around. Obviously big power drain.

I would love to see the two systems, robominers for ease of use, bigass miner for throughput (but requires better infrastructure), but if a have to choose, i pick the bigass miner (because com'on, those things are big and awesome (i love engineering ))

Just my two cents.

kinda offtopic: With all the bucket wheel excavator discussion nobody, really nobody, points THIS https://www.youtube.com/watch?v=9Qlrq9ETjzc out ?? I had my hopes in you, internet..

edit: mistakes and stuff

[Yaua]

FFF #129 doesn't exist in Factorio's page, and the link in the OP doesn't seems to work (or is it only me ?).

Well, I totally agree with what you say and honestly, this "step routine" is one of the things who needed some works. You're on it, that's perfect !
And all your ideas and propositions are great !

Concerning the new smoke visuals : don't forget about optimization and performances of course (I personally don't have a big computer, like so many of us )

Edit : And before I forget : I replayed the campaign to see how it evolved since last time I did... May be there should be more explanations concerning flying robots, oil and circuit networks [storage/requester chests, combinators...] (even if it's only the bases). Not a top priority, but it could be a very good thing for new players.

[Hecter94]

Since the game seems to switch from a focus on belts and items to a focus on liquid production and manipulation, why not do something similar with late-game mining? Which can also alleviate the issue of being forced to continually find new ore deposits as the old ones run out.

Instead of the regular smelters that we currently use that turn ore directly into plates, have a large (4x4 or larger) liquid smelter that draws a lot of power but can hold literal /thousands/ of ore and constantly pumps out molten metal which then has to be sent to a secondary facility and combined with water to make the actual plates that we all know and love.
Of course, by it's very nature, it would be considerably more efficient than simply smelting individual pieces of rock at a time and the secondary facility could be tweaked by the developers in updates to produce as many plates per second as is deemed necessary.

It would make endgame mining differ from early game mining in a way that isn't simply "bigger numbers" as well as fulfill many of the same goals that "dirty ore" would, such as requiring a secondary facility to process the ore before it can be used.

Finally, in my opinion, atleast, it fits within the logical flow of Factorio progression, you start with individual physical pieces moving along a conveyor belt and being smelted one at a time, to piping networks with massive smelters capable of holding literal tons of ore.

--Edit--
Also, if temperature ever gets updated to a more "realistic" system(not that I have a problem with the current system), it would be interesting to have to find ways to keep the metal liquid over long-distance piping, lest you run into the risk of it solidifying in the pipe and blocking it. Though that's probably too complex for a relatively simple change.

[tehroach]

Yer I don't think that rule really applies here, ie what if they decided to add a different ore?
would the rule apply then? No.

Fluids are a general item in the game, same as ores etc

Adding a new resource that adds completely new technology/production cycles/gameplay that don't exist yet OR adding redundant items (7 types of barreled fluid even if 1 tanker could do them all) are two different kind of things.

Lets separate this comment up a bit, because I agree with most of it
Adding a new resource that adds completely new technology/production cycles/gameplay that don't exist yet OR adding 7 types of barreled fluid are two different kind of things.
Yes I can agree that they can be viewed as two different type of things, different types of interface tools.
it doesn't make adding barrels "an insanely specialized solution" because the way different types of players will use them can vary greatly.

Barrels interface between pipes and belts etc. Something that tankers simply can't do.

[waduk]

Why the blog entry was gone ?

[RobertTerwilliger]

Barrels interface between pipes and belts etc. Something that tankers simply can't do.

Do pipes & belts really have to have interface between? I mean, you will never be able to send, say, gears though pipes - why should you send liquids though belts then?
Or another example: you can't send electricity through belt or pipe - it has it's own network. It's okay to have different networks for different resource types.

I literally never used barrels longer than from filling assembler to train and from train to pouring assembler. And I don't see any reason why I possibly should - 'cause pipes do it faster than any belt, just plonk some small pumps if distance is too long.

The only problem with tanker that appears to me is, like a loader from previous FFF, it is highly specialized entity and doesn't fit perfectly into Factorio concept. However this one at least has sense))

[RobertTerwilliger]

FFF #129 doesn't exist

the blog entry was gone

Confirming the same!

[Koub]

FFF #129 doesn't exist

the blog entry was gone

Confirming the same!

I reported it here : viewtopic.php?f=7&t=21684

[Qon]

I'm playing with train outposts mod and infinite ore mod (Didn't know that train outposts also did that, I have no idea if having both is a good idea). Because I don't want my factory building time to be spent 100% replacing mines. Higher tier mining drills with bigger area of effect would also be nice. These small miners are tedious to place. We have electric poles in different sizes for a reason.

This was a more minor point of the FFF, but I don't like the idea of increasing cargo wagon capacities. One of the coolest things I've built in my Railworld series is the Megatrain, a 10+ wagon train with at least 2 lead engines and one rear engine, which essentially acts as a portable fortress-builder around the small sea my factory is built beside. Only a select few custom-built stations can handle it and it's arrival in an outpost is a major event. I feel like the impact of the thing would be greatly diminished if you could fit that sort of capacity in something half the size.
Not only that, but it would mean that my entire factory would run of trains with only single cargo wagons, which just doesn't seem right.
!

10 wagons isn't really a mega train though after seeing a real mega train with hundred wagons that I'm not allowed to link to because of post count (NinjaEdit: https://www.youtube.com/watch?v=CxUxIsa ... Jhf95x2LKs, lol I can just edit my post to insert the link!). In my first playthough 3 loco 6 wagons was my small train and 12 wagons was the larger one.

I would like trains to not lose so much acceleration and top speed from having lots of wagons. Trains are supposed to be long. If someone says trains you think of a long chain of wagons pulled by a locomotive. When someone says factorio trains most people seem to think of 1 or 2 cargo wagon sadness vehicles.

Ore fields must not be infinite like oil wells, because it is desirable to mine them out completely and expand the base over them.

Eh what? You want to harm the end game because of aesthetics? Even if ore could be hidden by laying concrete over it or something?

If you didn't know it already: you can build your base on top of ore patches. Problem solved. I don't want to spend all my time in factorio building and removing mining outposts. And I don't want the reason to be a non-problem like this.

[canidae]

People like "end games" that produce good stories.
...
I like to thing of my factory as a living thing that I created. Watching it go awry and turn on my could make for a really cool story.

I'll echo in on the essence of funktapus' post. The current endgame doesn't produce enthralling stories, the enemies are too predictable, there aren't really any "new" things that happen. While I don't agree with all the ideas funktapus mentioned, I find the idea of the factory going awry interesting. Not so much fan of disasters Sim City-style, those you can't really protect yourself against, the challenges that appear should be possible to avoid or overcome (although it may very well be exceptionally hard to do so).
Another "endless" game like Factorio I've found to have a quite interesting late game is Dwarf Fortress. A giant two-headed, flame spitting steel moth eradicating your entire fortress? That's a good story. Losing is fun, but in Factorio you pretty much only lose if you choose to.