Advanced Ore Processing

[HarryE]

So I was playing Mekanism the other day. Mekanism is another one of those Industrial Mods for Minecraft™, and like any other Industrial Mod for Minecraft™, it includes an ore scaling mechanic. If you're not familiar with what that means, it means setting up a process by which you get more ingots per piece of ore. Most of the mods go with 2x ore processing (1xOre -> 2xDust. 1xDust -> 1xIngot)

Mekanism is different from most of the others. It allows you to scale your ore processing from anywhere up to 5x. The numbers are not the interesting part, though. The interesting part is the sheer amount of processing involved. A whole chain of chemical extraction and purification is required to set this all up. Ores must be infused with Hydrogen Chloride and purified with Oxygen, which can be collected from the atmosphere or separated from Water, etc, etc, etc- it's complicated! Secondly, you can scale it as you progress. Mekanism allows you to scale your ore -> ingot production at first 2x, but as your tech increases and you can perform those chemical processes, you can scale your ingot production up to 5x.

There's more info on the website, but this flowchart gets the idea across;



I think a similar system would work in Factorio's favour. I'm not suggesting to rip off Mekanism entirely, but I think an additional production line of ore would play into Factorio's hands quite nicely. I don't know about you, but my metals production/demand tends to fluctuate as I progress through the game. Generally speaking, I have metals coming out the wazoo towards the start, but as I progress my factories begin to choke from lack of metals. The only way to counter this is to add more factories and expand outwards in search of more ore to freight in. It would be great if Factorio could capitalise on it's greatest strength as a game and allow you to build a grand ore refinery into your base! And if you still prefer doing trains, you can still do that.

As a side note, whacking ore into a furnace to get an ingot isn't the most realistic thing in the world. A quick look at the Wikipedia Page on Metal Processing will show you that. It's not the best point, I know, but if you're going to replicate Oil Refining to a degree, then why not Metal Processing?

But back to a more gameplay-centred argument; think about Factorio's future. It's still in early development. There's going to be more and more content to be added. Which means for the player, there's going to be more stuff to be built. More production lines to be added. More demand for metals. This will mean that the players will be forced outwards into the map find more metals sooner. Not only might they not be prepared for biters that early, but it will also mean that the player's control of the map will be forced to expand, and they will have to deal with the problems that brings, and more and more (more than it is now, anyway) of the focus, therefore, becomes having to deal with biters attacking your outposts, and that is a very repetitive process. So why not replace that repetitive process with one that isn't so repetitive. One that takes arguably more thought and care to perform?

Ore processing. S'the future. It's one of those things that's bound to be added to Factorio anyway. What do you guys think?

[CreeperDaReeper]

I like it!

[pyrolytic_tungsten]

We already have a modest ore scaling system. It's called productivity modules but whether they are worth the energy/resources is a debatable question. I personally think they are worth it.

I do think there is room for ore processing. In real life one of the things that ore processing like ore flotation allows is the economical extraction of less concentrated deposits. In some places like the upper peninsula of Michigan old tailings were reprocessed a long time ago because there were improvements in the ability to extract copper from copper ore. It may be interesting to have lower grade ore deposits away from your base.

Personally I think the current endgame scale for the amount of ore needed and metals throughput required is too small currently. If you can get to the endgame without needing to move a significant distance for materials in late game it seems a little easy to me. I also use the resource spawner overhaul mod which helps bring resources more to what I want out of the game. It's also the only game of Factorio I played where I felt a strong need to get off of coal power as the closest significant non-starting deposit was ~750-1k tiles away.

[ssilk]

I like the idea of overcomplicated second way (but then super efficient) to produce stuff.
As that I would like to stay this suggestion here.

If the suggestion was thought more or less to implement exactly the above description, I would like to move this to mod requests board.

Opinions?

[HarryE]

If the suggestion was thought more or less to implement exactly the above description, I would like to move this to mod requests board.

Opinions?

I was thinking of making a mod of this. The configs that Factorio lays out make it rather easy. But I realised that if I wanted to make this in a way that I wanted to I'd have to make my own buildings and graphics, and I don't have the skill to make them match the style of Factorio. I'd have to make buildings for crushers (Imagine putting one of these over your conveyors. You could replicate this scene.), atmospheric pumps for oxygen, etc.

I'd also have to come up with a more thought-out design than above. Think about this in a more game-play centred way, making sure it's all balanced.

We already have a modest ore scaling system. It's called productivity modules but whether they are worth the energy/resources is a debatable question. I personally think they are worth it.

That's true, but I think it's a bit of a cop-out when all you need to do is slap them into your machine and that's the end of it. I think it's a missed opportunity when Factorio has the potential for so much more.

I do think there is room for ore processing. In real life one of the things that ore processing like ore flotation allows is the economical extraction of less concentrated deposits. In some places like the upper peninsula of Michigan old tailings were reprocessed a long time ago because there were improvements in the ability to extract copper from copper ore. It may be interesting to have lower grade ore deposits away from your base.

Personally I think the current endgame scale for the amount of ore needed and metals throughput required is too small currently. If you can get to the endgame without needing to move a significant distance for materials in late game it seems a little easy to me.

Agreed. That's why it may need better scaling. Like I say, when you start you normally have way more metals than you need compared to the demand in late game. Perhaps the answer is to have less ore when you start so that your default scaling at the start. Or maybe even throughout, so that you have to branch out as well as set up ore processing.

[Akronymus]

I hope they add it. Cause it seems fitting for the endgame.

[HarryE]

Hmm. I found this article on Iron Ore Processing that could be useful to learn from. You can kind of imagine how it would all slot into Factorio, too.

What are your thoughts on having different processes for both Iron and Copper? In Mekanism, you can apply the same process to each type of ore, but then again there are Five of them (IIRC), so handling five, different, separate processes would be crazy. But In Factorio, we only have two metal ores (for the time being, anyway), and it's generally a good idea to handle them differently anyway because Factorio's sorting system is so underpowered compared to the Industrial Mods for Minecraft™ (and that's not a bad thing. In fact it's probably a good thing). Therefore, having two different processes wouldn't be that hard for the player to implement.

[ssilk]

In Factorio you have the Logistic bots, which can sort.

[n9103]

There's no real need to adjust the availability of resources for those that would have the tech and resources to implement such a process.
A player that doesn't research in a very bad order, will have reached railroads and it's related researched before they run out of resources in the starting area, and therefore they should be expanding to new resource fields.

If you're simply looking to add extra steps that don't create more ore, then it would have a significant negative effect on the beginning of a game, even if you gave the player a free full smelting setup.
If you'd reduce the recipe efficiency of primitive smelting to combat this problem, then you simply move the worst impact to a bit after the beginning, when they would end up running out before they could reasonably expand.

A big difference between Minecraft and Factorio, is the scale of the production.
Minecraft has you dealing with dozens to hundreds. Factorio is thousands to tens of thousands and beyond.
The resource density and distribution of both games are so incredibly different that it's not very reasonable to take ideas from one and put them in the other.

[HarryE]

In Factorio you have the Logistic bots, which can sort.

Oh yeah, of course, but there's still lag there. It's not instantaneous like the sorting systems in Minecraft. In Factorio, sorting systems are less of a crutch, so you have to be clever how you lay out your base.

There's no real need to adjust the availability of resources for those that would have the tech and resources to implement such a process.
A player that doesn't research in a very bad order, will have reached railroads and it's related researched before they run out of resources in the starting area, and therefore they should be expanding to new resource fields.

That's why it would be necessary for the system to scale, to have sequential tiers. The first step would have to be balanced so that the resources to make that step would offer the sufficient yield to get to the next step.

If you're simply looking to add extra steps that don't create more ore, then it would have a significant negative effect on the beginning of a game, even if you gave the player a free full smelting setup.
If you'd reduce the recipe efficiency of primitive smelting to combat this problem, then you simply move the worst impact to a bit after the beginning, when they would end up running out before they could reasonably expand.

Are you saying that the problem would be that the player would be choking for resources? The solution would be to anticipate the amount of resources the player would need at that stage, and then provide them with the means to replicate the resources required for that stage of the game using the resources available at that time.

A big difference between Minecraft and Factorio, is the scale of the production.
Minecraft has you dealing with dozens to hundreds. Factorio is thousands to tens of thousands and beyond.
The resource density and distribution of both games are so incredibly different that it's not very reasonable to take ideas from one and put them in the other.

True, but I'm not suggesting that you simply just lift the mechanic as-is. I just think that there may be some merit to the idea when applied in Factorio. You're right that the current solution to resource choking is to spread out and use trains, but doing that gets really dull and repetitive after you've set up so many tracks, signals, mines, loading systems and defences the same way just in different places again and again and again. Maybe that's due in part to the way the overworld is designed. All I'm suggesting is another solution that works with Facotrio's greatest strength.

[ssilk]

In Factorio you have the Logistic bots, which can sort.

Oh yeah, of course, but there's still lag there. It's not instantaneous like the sorting systems in Minecraft. In Factorio, sorting systems are less of a crutch, so you have to be clever how you lay out your base.

That's part of the game: Be clever. About 10% of all new players in the forum say something into that direction (see discussions about splitter for example), but after playing a long while more (30 hours +) they say it's good like so.

True, but I'm not suggesting that you simply just lift the mechanic as-is. I just think that there may be some merit to the idea when applied in Factorio. You're right that the current solution to resource choking is to spread out and use trains, but doing that gets really dull and repetitive after you've set up so many tracks, signals, mines, loading systems and defences the same way just in different places again and again and again. Maybe that's due in part to the way the overworld is designed. All I'm suggesting is another solution that works with Facotrio's greatest strength.

I'm sure this is so, cause of the alpha-state game. I think we will get at some time tools to lay out tracks fast (see the automatic tracklayer mod if you need something like that yet https://forums.factorio.com/forum/vie ... =14&t=4287). For mines and loading systems: Use blueprints. Use the construction bots, they can built things in about 1/100th of the time you need. I cannot play without them. There is the pocket bots mod https://forums.factorio.com/forum/vie ... =14&t=4441.

All of that is currently a bit "crappy", it works more or less well. Again: This is alpha. We will get such stuff per sure, there is just no way to play the game without tools in the way the devs think it should work in the end.

[Blackence]

I like this.

It's for endgame, obviously. So you start out like you do now. Eventually you build mining outposts. Then you add more outposts. And more. And more.

I'd love to replace the 5th or 6th mining outpost with an "advanced ore processing factory", because setting up new mining outposts all the time is relatively boring and repetetive, so not very factorio-style. When using advanced ore processing, I'd expect that advanced mining outposts produce (much?) more resources at the cost of a large up-front investment for the required machines/buildings.

So.. Just like productivity modules, but much more expensive, difficult to set up and I guess more effective (less pollution, more freebies).

[ssilk]

I'd love to replace the 5th or 6th mining outpost with an "advanced ore processing factory", because setting up new mining outposts all the time is relatively boring and repetetive, so not very factorio-style.

Yes, that looks like a good reason! I searched for arguments to this and I this this one is a valuable argument in my eyes.

[Artman40]

A system like or similar to this would work well with tier 1 being mid-late game thing while the 4th tier being in the very endgame process.

One interesting thing would be if different ores (something that would be found only further away from the center of the map) could also produce useful by-products (like rarer metals) which would only be extractable through ore processing.

[Artman40]

Thinking about this further...
Current limit on production expansion happens when the amount of mining sites that can be established starts to equate to the amount of mining sites where resources get depleted.

In order for the ore processing be viable to be made on-spot, I think it would be better if processed ore is the one which produces more plates rather if raw ores produce more processed ores. Else it would be more economical to transport raw materials off to the main factory area due to lack of need for longer trains.

I think the ore processing method would be good to be at least 3-tiered, maybe 4 tiered. The first tier would be nice to be something like auxiliary building for mining drills. The second tier being something that, aside helping to produce more ore for that type, would also produce useful by-products like I said before (e.g. in real life, molybdenum is produced as a by-product of copper production). The same would go to tier 3.

I think the end product growth would be better to be exonential instead of linear 2× 3× and 4×. Maybe using tier 1 ore processing would yield 50% more products, tier 1 and 2 combined would yield 150% more products and tierl 1, 2 and 3 would yield 300% of more stuff. This would be to compensate for increased effort on setting the building network up.

Sorry if I did not make myself clear.

[CreeperDaReeper]

Thinking about this further...
I think the end product growth would be better to be exonential instead of linear 2× 3× and 4×. Maybe using tier 1 ore processing would yield 50% more products, tier 1 and 2 combined would yield 150% more products and tierl 1, 2 and 3 would yield 300% of more stuff. This would be to compensate for increased effort on setting the building network up.

(Going off of the flowchart in OP.)
Without Refining: 1:1
--1 Ore = 1 Plate

With Tier 1 Refining: 1:2
--1 Ore = 2 Dust
--1 Dust = 1 Plate

With Tier 2 Refining: 1:4
--1 Ore + Oxygen = 2 Clumps
--1 Clump = 2 Dust
--1 Dust = 1 Plate

I won't bother with writing them all out, but unless we used a ratio of like 2 in to 3 out, or 4 in to 5 out, per refining tier, the overall refining rate is ^2. Exponential, not liner.
Ore=Plates
Without: 1=1
Tier 1: 1=2
Tier 2: 1=4
Tier 3: 1=8
Tier 4: 1=16
Tier 5: 1=32

And then there's one other thing, assume for a second that each building in the 'refining' process has module slots.

[Artman40]

Ok. That explains things. Ore processing should definitely be reserved for game stages where exponential resource growth is no longer possible in other methods.

[The Phoenixian]

[snip]

I think the end product growth would be better to be exonential instead of linear 2× 3× and 4×. Maybe using tier 1 ore processing would yield 50% more products, tier 1 and 2 combined would yield 150% more products and tierl 1, 2 and 3 would yield 300% of more stuff. This would be to compensate for increased effort on setting the building network up.

Sorry if I did not make myself clear.

I'd like to make a case for the opposite. As Factorio is very much about doing things on a massive scale, in my eyes at least, I feel it would be better to make productivity upgrades a project of diminishing returns so as to add a bit of complexity while still promoting massive builds. The first part of the structure should be fairly complex, though not overwhemingly so, and give the either the single best returns of any upgrade or close to it to provoke players to take that first step. After players have gotten used to that you can draw them in further with more complex upgrades being a little more worthwhile and less complex upgrade being less so. (Modified slightly for how early or late they are in the cycle.)

In my experience people can and will work hard for modest 10-30% gains, especially if they're cumulative, and those little gains really do make a meaningful difference, especially if it switches you from one big problem spot, or in this case production bottleneck, to show another you can improve on. The most important part is not how much progress is made, but that a player feels like they're making progress in the moment.

[HarryE]

In Factorio you have the Logistic bots, which can sort.

Oh yeah, of course, but there's still lag there. It's not instantaneous like the sorting systems in Minecraft. In Factorio, sorting systems are less of a crutch, so you have to be clever how you lay out your base.

That's part of the game: Be clever. About 10% of all new players in the forum say something into that direction (see discussions about splitter for example), but after playing a long while more (30 hours +) they say it's good like so.

Aye! I never said that wasn't so. I think the current sorting mechanic is very balanced as it is!

I'd love to replace the 5th or 6th mining outpost with an "advanced ore processing factory", because setting up new mining outposts all the time is relatively boring and repetetive, so not very factorio-style.

Yes, that looks like a good reason! I searched for arguments to this and I this this one is a valuable argument in my eyes.

This is my problem exactly. Even with the track layer mod and using construction bots, the process of setting up outposts is still arduous but more importantly; repetitive.

In my experience people can and will work hard for modest 10-30% gains, especially if they're cumulative, and those little gains really do make a meaningful difference, especially if it switches you from one big problem spot, or in this case production bottleneck, to show another you can improve on. The most important part is not how much progress is made, but that a player feels like they're making progress in the moment.

That's true. I think people are trying to come up with a solution that makes this an advantage over using trains, but I think it would be nice to give people the choice over whether they use ore processing or trains rather than forcing a metagame of working out which is categorically better. But then again, whichever way you cut it, it would be best to use both, but at least you get to choose which you prefer until you get to that point. Course, if you don't want to force a metagame you have to make sure both choices are balanced in some way.

[SuperSandro2000]

Did you mind it if I maybe make a mod out of this?