Effect of quality for various items

[Khagan]

Thinking about how to prioritise quality upgrades for various items, and indeed, whether to bother with them at all, it seems to me that the effectiveness of quality for different end products can be grouped into several distinct classes. (The examples given for each class are not intended to be exhaustive. The value of quality for intermediate items is obviously determined by the end product they are used to make.)

1. Qualitative benefit: high-quality entities do things better than any quantity of low-quality ones can.
   Quality and productivity modules.
   Armour and vehicles with a grid, and most items that go in a grid.
   Radar and turrets.
   Bots.
   Fuel (when used in a vehicle).
2. Large quantitative benefit: the smaller required number of high-quality items also means a smaller required number of (expensive) associated items.
   Crafting machines.
   Speed and efficiency modules; beacons.
   Ammunition.
3. Small quantitative benefit: the smaller required number of high-quality items needs less space and perhaps less power, but otherwise is no better than using a larger number of lower-quality ones.
   Most entities used in electricity generation (steam engines, solar panels and accumulators, etc.).
   Power poles.
   Roboports.
   Walls.
4. No significant benefit: quality either provides no benefit at all, or just a health increase that is irrelevant for entities not exposed to combat.
   Belts, pipes, chests and tanks.
   Rails and all railway equipment.
   Circuit network items.
   Tiles.
   Fuel (when used in a building).

For class D items, there is no point complicating and slowing your production line with anything quality-related.
For class C, you might roll the dice for quality at the final stage of construction (it's more-or-less a free bonus), but not care all that much what quality you end up with.
For class B, it might be worth developing a production line with guaranteed quality intermediates by creaming off the best at each step, but time-consuming and wasteful recycling to maximise quality is probably uneconomic.
Only for class A items might repeated recycling to get the very best possible be a plausible choice.

EDIT: Deleted bots from A class: construction and logistic bots are C class; combat bots are technically A (because of their increased range) but as consumables would not justify repeated recycling.

[mmmPI]

Is this a tier list of quality ?

If so, i would like to argue for beacons compared to fuel and bots. Quality beacons are very strong because they cost less energy and provide more module effect which to me seem not only quantitative but also the same kind of quality that seem too generously given to bots and fuel regarding "unlocking possibilites" that could regarded as qualitative, as "it becomes usable on space platform" for beacons . Whereas bots or 'worse' imo fuel, it's more on the quantitative side. I wouldn't chain recycle fuel for higher quality for another purpose than making something go (needlessly) super fast.

I think the value of the intermediate of "quality" may also depend on how "easy" it is to get it, some quirks in the recipes allows to get some materials for free once you have enough research, or wether they come from infinite ressources or not in one way. And also in another way on how versatile it is how early in the prod chain. To me legendary copper is worth nothing, whereas legendary plastic is worth a bit more but legendary uranium is worth the most. Despite it being not versatile, it's the most difficult to get, copper is free if you recycle low density structure, or rather you can trade a little bit of plastic for a lot of copper and steel, and plastic is used for red circuits which is one of the most complicated/costly ressources that don't get productivity bonus. Unlike blue circuits.

I think another thing to consider regarding the conclusion of what to recycle is "how productive is the recipe". For example beacons are made in electromagnetic plant, which have a 50% productivity bonus in them, and they are recycled very fast despite requiring a lot of material to make. This allow for the first few quality modules that you have available to find a spot with a lot of efficency, because they churn through a lot of material wether making beacons or recycling them. ( after placing them in quality module making machine )

[Khagan]

Is this a tier list of quality ?

Sort of.

If so, i would like to argue for beacons compared to fuel and bots. Quality beacons are very strong because they cost less energy and provide more module effect which to me seem not only quantitative but also the same kind of quality that seem too generously given to bots and fuel regarding "unlocking possibilites" that could regarded as qualitative, as "it becomes usable on space platform" for beacons . Whereas bots or 'worse' imo fuel, it's more on the quantitative side. I wouldn't chain recycle fuel for higher quality for another purpose than making something go (needlessly) super fast.

Fuel being a consumable, I agree that chain recycling would not be sensible here. Nevertheless, I class it as 'A', since the extra speed and acceleration is not achievable any other way. Bots are not so clear: perhaps a 'B' classification would be better.

I also agree that quality beacons are strong, but I would still call them a strong 'B' rather than an 'A', since it is possible (if expensive) to achieve the same result simply by using more.

My intent was that the initial classification be fairly objective, but the suggested treatment of each class be just a starting point for personal judgement in each case.

(And this is exactly the sort of discussion I was hoping to provoke. )

I think the value of the intermediate of "quality" may also depend on how "easy" it is to get it,
[...]
I think another thing to consider regarding the conclusion of what to recycle is "how productive is the recipe".

Yes, these are both other things to consider in practice.

[mmmPI]

I also agree that quality beacons are strong, but I would still call them a strong 'B' rather than an 'A', since it is possible (if expensive) to achieve the same result simply by using more.
My intent was that the initial classification be fairly objective, but the suggested treatment of each class be just a starting point for personal judgement in each case.
(And this is exactly the sort of discussion I was hoping to provoke. )

Well then my personnal judgment is that beacons are underated, they totally deserve the rank "A" not even "A-" ! because on space platform the fact that quality beacons consume less energy cannot really be replicated by "more" of the lower quality !

I think pipes can be exposed to combat with flamethrower it's something that often gets damaged in my death worlds.

I think some items for energy production may be of higher tier for platforms, or limited space like Fulgora or Aquilo.

Big mining drill are A tier for me.

I think there are places that incentivize or items that does too toward a C option for production, only roll the final dice like accumulators, use the tier 1 for science, and the tier 2 for building, but then you also recycle the tier 2 and you use the tier 3 for building, and later you can do with 4 or 5, without ever setting a dedicated production for it, you sill end up with lots of quality that you need to stop producing when you have too many. This in Fulgora.

You can also mine quality scrap or iron or coal or copper , to get some raw ressources of the choosen level of quality ,and only build whatever you need from them. ( for a few belts that carry biterss eggs and risk damage in case of hatching, or that one turret you put at the back of your ship that need higher quality than you other turret to properly cover things , or that one reactor for the platform or the one you bring in Aquilo). Sort of mutualization of the raw ressources of "quality" not in relation with the production chain of a particular item but rather for them to be available for "any" item that you may need, (not on a sustainable basis).

[Khagan]

I think some items for energy production may be of higher tier for platforms, or limited space like Fulgora or Aquilo.

Instead of making the classification variable, I'd rather say that limited space is what makes class C items useful. In contrast, class A and B are valuable even when space isn't limited.

[Makka77]

I mentioned my thinking in another thread regarding order of production importance of quality items. It would seem sensible that the first quality items you attempt to produce are quality modules themselves (A tier) followed by labs to make the most of science production. But from there it would seem to be variable depending on playstyle.

One FFF mentioned the devs LAN game had a facility dedicated to producing legendary bulk inserters, which sounds like endgame minmaxing strategy to get everything as streamlined as possible with all the best gear.

I agree with your classifications on the whole, and also the variable value per application. Good discussion!

Edit: I misread the lab quality stats. There is only a speed increase, no productivity bonus. So next candidate for quality production would be Modules and beacons for small local boosts in key areas.

[MeduSalem]

Add rocket thrusters to A-"tier".

Reason being... Because without quality thrusters you can not increase travel speed beyond a certain cap. I found out yesterday that having more than ~5 thrusters have no benefit to increase top speed because the increasing width of the platform will counter-act it. That is why you cannot really increase top speed beyond a certain point no matter how many thrusters you place because for every thruster you place, the platform also become 4 tiles wider. For that reason the highest values for top speed you can reach cap out at around ~300km/s for standard quality thrusters. Mass barely plays a role if it is not really huge (like several thousand tons),

To overcome that speed limit requires to use quality thrusters; those allow for even higher top speeds because they offer more thrust without increasing the platform width that would counter-act it. That is why they are totally A-tier for use with spoiling stuff since you can't achieve it with "throwing more of the same" at it.^^

[mmmPI]

I found out yesterday that having more than ~5 thrusters have no benefit to increase top speed because the increasing width of the platform will counter-act it. That is why you cannot really increase top speed beyond a certain point no matter how many thrusters you place because for every thruster you place, the platform also become 4 tiles wider. For that reason the highest values for top speed you can reach cap out at around ~300km/s for standard quality thrusters.

Although i share the opinion that thrusters are somewhat A tier, for the same reasons as beacons, in a way. I would like to point out that you CAN have more than 5 thrusters, and reach a top speed superior to 300 km/s with normal quality only. It is incorrect to say that a platform becomes 4 tile wider when you add a thruster. This is because you can have several decks of thrusters, albeit the rules were made harsher, it is still possible, i made a design to illustrate what i mean, reaching 377 km/s : viewtopic.php?p=632307#p632307

[shopt]

This is a very interesting discussion. Categorizing this way is interesting, but to me it's a bit academic. I don't think you can then go straight to a gameplay application based purely on these categorizations.

First thing I would say is that large power poles are technically in the A category IMO. There are islands on Fulgora that can only be connected with higher quality poles. Regular poles will be able to eventually with lots more tech, but for a large part of the game the higher quality poles let you do things that are otherwise impossible.

While vehicle fuel technically lets you do things that are otherwise impossible, I would say that's of limited practical benefit. Definitely not enough for me to set up recycling loops or high quality intermediates for guaranteed quality. Similarly not all space saving is equal. Inserters aren't on the list but a case could be made for them to be in A, B or C. Sometimes the space between a belt and a machine is otherwise dead, and just putting more inserters is no big deal (ignoring UPS concerns). Sometimes there isn't enough room to put enough regular quality inserters to properly feed the machine, which effectively puts inserters in A tier (no other way to get that machine running full speed) or B tier (well you could just let a machine run below capacity and build more machines).

In another example roboports can often take up high value space where it's not possible to "just build more". Or when they are used just to extend range they usually take up low value space.

I'm also not sure I'm sold on the conclusion that putting quality in class C is more or less a free bonus. There's the slight speed penalty, there's the opportunity cost of using speed/efficiency modules instead, and there's the blueprinting and logistics complications of dealing with multiple quality levels. For those who pulls walls off the end of their grey science line, making quality walls would be downright frustrating. Reiterating the "not all space is equal" point, the space walls take up is usually very low value.

[BraveCaperCat]

Maybe productivity modules should go down to at least B tier, because of the 600% cap on them. Until you use mods to add additional qualities however, this won't matter too much.

[MeduSalem]

Although i share the opinion that thrusters are somewhat A tier, for the same reasons as beacons, in a way. I would like to point out that you CAN have more than 5 thrusters, and reach a top speed superior to 300 km/s with normal quality only. It is incorrect to say that a platform becomes 4 tile wider when you add a thruster. This is because you can have several decks of thrusters, albeit the rules were made harsher, it is still possible, i made a design to illustrate what i mean, reaching 377 km/s : viewtopic.php?p=632307#p632307

I know that it is possible; however multiple rows of thrusters are a special situation that I think most people will likely never do.

If you would just have put 5 Rare thrusters at the back you could spare yourself all the other ones. You would also have gotten around 380km/s with a muuuch easier design and not having to sustain 13 thrusters and build the piping to get it up all the way up there just to overcome the thruster build-restriction.

The 5 rare thrusters would also be way more fuel efficient (there it is again) to achieve that speed than the total whopping of 13 normal thrusters are. Because 5 rare thrusters consume roughly 900 units/s; 13 normal ones consume a whopping 1450 unit/s if you run them at 75% fill-rate to get 100% thrust out of them. You are just doing it by sheer brute force at that point. And yea, I did that comparison for different qualities in an excel table the other day too when I wrote I did many different combinations. Just didn't write it in the other thread because I thought it is way out of scope. ^^

As said, the additional thrusters you put there because you made the platform even wider are partly just battling its own added width; you add tons of overhead consumption just to be able to put 4 more thrusters in a 2nd row which then allow you to surpass the top-speed.

I mean sure, if you would replace all those 13 thrusters with rare ones then you can go even way beyond the 380km/s probably into the 450-480km/s range or something. But then you would have to make a couple more chemplants to deal with the higher consumption of that many higher quality thrusters as well. And then one could argue that you can achieve the same speed yet again with just 5 legendary ones that are also more fuel efficient for the thrust they provide.

I want to see you put 13 legendary ones on the ship. Must be ridiculous at that point. Sure do-able and even sustainable too with all the chunks & infinite asteroid processing productivity. xD


Another thing would be, that if you would do a 3rd row of thrusters and per row just 1 "hang-over" engine per side, it would actually be even faster & more fuel efficient than 2 rows with 2 "hangover" engines per side because... as said... platform width would be narrower by 8 tiles and that would cause less overhead from width. That I would actually find even an acceptable approach. It is the most efficient to actually stack more rows than it ever is to build more wide.


Anyway, this is getting off-topic.

[mmmPI]

I know that it is possible; however multiple rows of thrusters are a special situation that I think most people will likely never do.

Showing it is possible was kind of the goal ,i'm glad it worked , because people restricting themselves to 5 thrusters would be a bit boring,i think people will post such designs.

[Khagan]

I don't think you can then go straight to a gameplay application based purely on these categorizations.

Absolutely not. The classification is just a starting point.

First thing I would say is that large power poles are technically in the A category IMO. There are islands on Fulgora that can only be connected with higher quality poles.

An interesting example. I would still prefer to classify them as C class, since this is a rare special case. As I said further up, A or B class are always valuable, while C class are occasionally valuable. The most obvious circumstances when they are are those of very restricted space (e.g. on a space platform), but you have pointed out another type of case. It still has the occasionalness that allows you to obtain quality items by 'buying lottery tickets', using the winners for the special case, and using the losers for everything else without regret. For class A and B you want a more systematic approach, since quality items are always what you want.

While vehicle fuel technically lets you do things that are otherwise impossible, I would say that's of limited practical benefit.

I think I was inconsistent in having two different classification for fuel. Fuel should probably be classed as C grade, since quality fuel is useful for vehicles but not for buildings. So again, you can make rocket fuel with a small chance of quality, use the good stuff for trains or tanks, and use the bulk of normal quality for rockets.

Inserters aren't on the list but a case could be made for them to be in A, B or C. Sometimes the space between a belt and a machine is otherwise dead, and just putting more inserters is no big deal (ignoring UPS concerns). Sometimes there isn't enough room to put enough regular quality inserters to properly feed the machine, which effectively puts inserters in A tier

I think the latter case is much rarer, making C class again the right assignment.

I'm also not sure I'm sold on the conclusion that putting quality in class C is more or less a free bonus. There's the slight speed penalty, there's the opportunity cost of using speed/efficiency modules instead, and there's the blueprinting and logistics complications of dealing with multiple quality levels.

Yes, 'free' was hyperbolic. But the costs are generally small compared to the benefit.

For those who pulls walls off the end of their grey science line, making quality walls would be downright frustrating. Reiterating the "not all space is equal" point, the space walls take up is usually very low value.

I'm not sure where you think the frustration would come from, but I agree about walls generally being in low value space. It has occurred to me that one might supply higher quality walls as replacement items for use by repair bots. That way, precisely those spots on the wall that have been shown to be at risk of destruction would get a quality upgrade.

Anyway, thanks for some thought-provoking comments!

[BraveCaperCat]

I don't think you can then go straight to a gameplay application based purely on these categorizations.

Absolutely not. The classification is just a starting point.

First thing I would say is that large power poles are technically in the A category IMO. There are islands on Fulgora that can only be connected with higher quality poles.

An interesting example. I would still prefer to classify them as C class, since this is a rare special case. As I said further up, A or B class are always valuable, while C class are occasionally valuable. The most obvious circumstances when they are are those of very restricted space (e.g. on a space platform), but you have pointed out another type of case. It still has the occasionalness that allows you to obtain quality items by 'buying lottery tickets', using the winners for the special case, and using the losers for everything else without regret. For class A and B you want a more systematic approach, since quality items are always what you want.

While vehicle fuel technically lets you do things that are otherwise impossible, I would say that's of limited practical benefit.

I think I was inconsistent in having two different classification for fuel. Fuel should probably be classed as C grade, since quality fuel is useful for vehicles but not for buildings. So again, you can make rocket fuel with a small chance of quality, use the good stuff for trains or tanks, and use the bulk of normal quality for rockets.

Inserters aren't on the list but a case could be made for them to be in A, B or C. Sometimes the space between a belt and a machine is otherwise dead, and just putting more inserters is no big deal (ignoring UPS concerns). Sometimes there isn't enough room to put enough regular quality inserters to properly feed the machine, which effectively puts inserters in A tier

I think the latter case is much rarer, making C class again the right assignment.

I'm also not sure I'm sold on the conclusion that putting quality in class C is more or less a free bonus. There's the slight speed penalty, there's the opportunity cost of using speed/efficiency modules instead, and there's the blueprinting and logistics complications of dealing with multiple quality levels.

Yes, 'free' was hyperbolic. But the costs are generally small compared to the benefit.

For those who pulls walls off the end of their grey science line, making quality walls would be downright frustrating. Reiterating the "not all space is equal" point, the space walls take up is usually very low value.

I'm not sure where you think the frustration would come from, but I agree about walls generally being in low value space. It has occurred to me that one might supply higher quality walls as replacement items for use by repair bots. That way, precisely those spots on the wall that have been shown to be at risk of destruction would get a quality upgrade.

Anyway, thanks for some thought-provoking comments!

Posting long replies to long replies are we? I guess I need to make this post extremely long, so that you will read it then. Oh wait, it's off topic now. I guess I have to mention that all modules should be moved to A tier - as, even with beacons, there is a limit to how many modules there can be affecting a machine. The quality of the modules helps with this issue. Now it's on topic!

[Frogtorio]

Does anyone know how the miners are affected by quality? The wiki about quality mentions this possible bonus: "Reduced resource depletion on miners (likely multiplicative in effect with productivity)". But the pages about the miners only mention a health bonus. Which is right ?

[BraveCaperCat]

Does anyone know how the miners are affected by quality? The wiki about quality mentions this possible bonus: "Reduced resource depletion on miners (likely multiplicative in effect with productivity)". But the pages about the miners only mention a health bonus. Which is right ?

Factoriopedia and the page about quality. Quality miners will deplete resource deposits slower, similar to productivity. The difference between productivity and the quality bonus is that the quality bonus does not improve the rate of resource output.

[armagin]

Minor side note, power poles give increased coverage with quality, making some funny builds possible where you need the extra range. I imagine there are other quality tiers on other items that make quirky and specific builds possible.

[shopt]

For those who pulls walls off the end of their grey science line, making quality walls would be downright frustrating. Reiterating the "not all space is equal" point, the space walls take up is usually very low value.

I'm not sure where you think the frustration would come from, but I agree about walls generally being in low value space. It has occurred to me that one might supply higher quality walls as replacement items for use by repair bots. That way, precisely those spots on the wall that have been shown to be at risk of destruction would get a quality upgrade.

The frustration would come from the fact that grey science will only use normal quality walls. If you put quality walls on the same belt, it will all work until the buffer at the end backs up. Then your belt would become full of quality walls and grey science stops working. There are lots of ways around this, but it doesn't change the fact that something that was viable becomes unviable once you take advantage of "free" quality.

Regarding bots replacing destroyed walls with higher quality, I don't think it works like that. I'm fairly sure a destroyed normal quality wall will have a normal quality wall ghost, and bots will never use a high quality wall to replace it without player intervention. You will need to be paying attention to the destroyed entity alerts and running an upgrade planner over the ghost.

[BraveCaperCat]

For those who pulls walls off the end of their grey science line, making quality walls would be downright frustrating. Reiterating the "not all space is equal" point, the space walls take up is usually very low value.

I'm not sure where you think the frustration would come from, but I agree about walls generally being in low value space. It has occurred to me that one might supply higher quality walls as replacement items for use by repair bots. That way, precisely those spots on the wall that have been shown to be at risk of destruction would get a quality upgrade.

The frustration would come from the fact that grey science will only use normal quality walls. If you put quality walls on the same belt, it will all work until the buffer at the end backs up. Then your belt would become full of quality walls and grey science stops working. There are lots of ways around this, but it doesn't change the fact that something that was viable becomes unviable once you take advantage of "free" quality.

Regarding bots replacing destroyed walls with higher quality, I don't think it works like that. I'm fairly sure a destroyed normal quality wall will have a normal quality wall ghost, and bots will never use a high quality wall to replace it without player intervention. You will need to be paying attention to the destroyed entity alerts and running an upgrade planner over the ghost.

Because of this, I request that Class E be added - items which suffer because of quality. Walls might be put here, though I personally think they should go in Class B or C.
Class E would mostly be items which aren't used much, or intermediaries like Iron Gearwheels. (items which place) Tiles might also be put here.

[Khagan]

The frustration would come from the fact that grey science will only use normal quality walls. If you put quality walls on the same belt, it will all work until the buffer at the end backs up. Then your belt would become full of quality walls and grey science stops working.

This is solved by one (1) splitter. Exactly as I am already doing on my purple science production line to pull out the better quality electric furnaces and productivity modules. I would call it rewarding rather than frustrating (though high quality walls would be less interesting than high quality modules and furnaces).

Regarding bots replacing destroyed walls with higher quality, I don't think it works like that. I'm fairly sure a destroyed normal quality wall will have a normal quality wall ghost, and bots will never use a high quality wall to replace it without player intervention.

Yes, you are right. Never mind, it was just a bit of brainstorming.