Friday Facts #128 - Back down to earth

[Koub]

Well I think vanilla should be meant to fit most, and people with extreme needs should be able to get them through modding.
If you want a sh*tload of spaghetty pipes for lubricant, for every inserter, loader, unloader, mk3 assembling machine, belt, ... Fine, but wanting the game nearly implayable for the sake of challenge should not discourage 99% from the players to actually play.

Also @Carl, please chill, let's keep it courteous .

[Ojelle]

Well I think vanilla should be meant to fit most, and people with extreme needs should be able to get them through modding.
If you want a sh*tload of spaghetty pipes for lubricant, for every inserter, loader, unloader, mk3 assembling machine, belt, ... Fine, but wanting the game nearly implayable for the sake of challenge should not discourage 99% from the players to actually play.

Also @Carl, please chill, let's keep it courteous .

And for those with extreme needs: Just install all bob his mods! (bob's mods)
I did/am doing a run with a friend on multiplayer with those and its kinda different. Lot more different things (for me a bit to much, my friend loves it dough)

[Linux_user404]

I do not see the usefulness of this loader in my factory... Some say it might be overpowered, but for me it looks just redundant. The only use I can think of - if I place a chest in a wrong place and it is full of staff, but I would just use a car for that. It would be much more useful to have some chests that would be able to chain-load, or have some specialized high capacity storage(for ores a coal for example).

[OBAMA MCLAMA]

but wanting the game nearly implayable for the sake of challenge should not discourage 99% from the players to actually play.

If that was a response from me, people that go logistic bot bases, know and will come across problems with bots. As people with belt bases, will know of the limitations like bot players do. And so forth.
I would assume the average player uses a wide mix of belt + bots and everything in between, learning towards bots when the player realizes how useful they are. A loader in this situation as its currently being advertised, would really only benefit people that go heavy on belts. (mostly)

[brunzenstein]

Well I think vanilla should be meant to fit most, and people with extreme needs should be able to get them through modding.

I sign that statement

[Yaua]

I'm a bit late, but for me, the loader can be implemented and interesting. But I would see mainly it as a train unloader : this way, it would ease a little unloading (loading ? Not sure this possibility is a good idea) of trains and this way it would be in a restricted "domain of applications" (so not messing with the global logic of inserters etc). Of course, it should be very costly, in the end of researchs etc... But it can have it's place (so for me it's a yes, if it's done correctly, as usual)

Concerning the tech tree it will be a very good thing. I would have only one suggestion : could it be possible to have an option to directly see over/alongside the image of the technology (like in the images here : https://eu3.factorio.com/assets//img/bl ... tics-2.png ) how much sciences packs of each colors will be needed ? (without needing to click on it I mean) It would help a lot when you're lacking a specific color of science pack, to directly see what is needed in the shortcuts of researchable technologies (because actually you need to click on each icon to see what it "costs" exactly).
If I'm not clear, tell me, I'll try to explain better !

[RobertTerwilliger]

the loader can be implemented and interesting. But I would see mainly it as a train unloader

In fact, this is the main "against". Because it breaks devs' creed "general > specialized", i.e. the whole spirit of Factorio is making complex system from simple elements, and highly specialized element won't fit the concept.

IMO.

[Yaua]

the loader can be implemented and interesting. But I would see mainly it as a train unloader

In fact, this is the main "against". Because it breaks devs' creed "general > specialized", i.e. the whole spirit of Factorio is making complex system from simple elements, and highly specialized element won't fit the concept.

IMO.

I thought about this point this morning too and must admit I didn't thought totally about this ("making complex system from simple elements") when I wrote my suggestion (except that it would make another object to build, and there are already a lot of objects, and so creating a highly-specialized object isn't the ideal). My idea (hope) was that it could fit in the game without breaking the concept of the game. Even if I think that you're certainly totally right I can't prevent myself to be attracted with the idea of a train unloader or something like this But I'm not saying I'm fully for : it's just if the idea can fit in Factorio's spirit. If not, no problem ! And I agree with you, my idea doesn't seems to fit in.

Honestly, I am curious to see how the unloader could be implemented while keeping Factorio's spirit intact, because the idea is interesting. But, of course, Factorio's spirit is the priority ! (because this spirit IS the game and is why the game is awesome)
But this is a complicated decision, I understand why our opinion was asked XD

[golfmiketango]

I'm sure much of this has been said above (haven't read this whole thread and don't plan to) but if the last 30 pages were like the first two, then I'm guessing a majority of those expressing opinions have been tentative or against the loader idea.

I have moderately strong feelings about it so I figured I'd contribute my perspective on this loader thing and then immediately hop the train back to my flame-retardant virtual factory. Executive summary of the following: "yes, please!"

First, why I like the idea: Compressing a belt with product is an interesting puzzle until you figure out how to do it. Then it becomes a repetitive and uninteresting engineering chore. In a game about automation, I would like to automate this away (preferably, not by cut-pasting my repetitive and uninteresting belt-compression designs ad infinitum). Also, it's completely realistic as to what might happen in a real factory (actually in a real factory that loader gizmo would ideally have a "compression" frob the technicians could set to make the loader consistently achieve partial compression ratios -- maybe even two frobs, one for each side of the belt... but I digress).

Finally, and this is the part that is gonna get me flamed, but I can't resist saying it: some people might not like this thing being in their factory, in which case, they presumably would just avail themselves of the opportunity not to craft any, just like they probably decline to craft any number of items already in the game that they don't want or aren't interested in.

Sorry if the above points are hopelessly redundant and every other poster also prefaced by saying they hadn't read the thread

[RobertTerwilliger]

... an interesting puzzle until you figure out how to do it. Then it becomes a repetitive and uninteresting engineering chore

As an IRL construction designer, I'm saying 100% positive this IS realistic - most of engineering workis repetative - luckily with computers we can automate most of it by copy-pasting (can't even imagine how previous generation did all that routine manually...)
And we have a powerfull tool in Factorio for copy-pasting - blueprints. Use them and enjoy not having to do all the work manually.

[Tallinu]

TLDR: Yes to loaders. Please. But just for containers, as I believe was the original intent, not for assemblers and such.

Once you have sufficient production to be able to saturate a belt, a balancer is trivial. It has always been trivial. Just look it up online and copy the design – no challenge at all. It's not difficult to come up with balancer designs, it's not challenging to build them, it's just mildly annoying and fiddly to build. Actually, I guess the biggest challenge would be to find a place where they fit, but you can anticipate that and plan ahead. So, as I understand it, none of that changes. If at some late stage you are suddenly able to replace your old style balancers with a design based loaders, after hours of research ... yea, I don't see the negative side effects. I see potential for new designs and interesting applications.

This right here.

As described in the FF post, the Loader would be for containers. No mention was made of assembling machines -- or furnaces, or chemical plants, or research labs, or anything else. Inserters will always be the king where flexibility is concerned. And if you only want items on one side of a belt -- particularly the opposite side of the belt -- guess what? Loaders aren't going to do the job, you need inserters. But if you need to pack a belt with items, or transfer items rapidly from a belt into a single container? You really can't do those things with inserters.

To me, containers means chests and cargo wagons, both of which require tricky and incredibly annoying builds to load and unload using current mechanics. Inserters are not and can never be an optimal solution for either of these situations, simply because there is not enough space.

With just a chest involved, you have at maximum three sides available for loading or unloading if you want to still be able to do the opposite, and employing three of them to maximize throughput of one of those operations uses up a lot of space surrounding that chest. It's awkward and slow even in the best case. People work around this problem by running multiple small chests in parallel, using up even more space but potentially resulting in a situation where some of the chests are full and some are empty so you aren't getting full throughput anyway.

For cargo wagons, it takes stacks of inserters and those rows of chests on both sides of a wagon, and a bunch of complicated belt layout, to unload them at the current theoretical maximum rate (which is only the theoretical maximum because there's just no space to add any more inserters), to the point where setting up a train loading / unloading station is a big hassle that is only "solved" once you can use construction robots to build them from blueprints... and not everyone is fond of robots.

Having loaders require power would be fine, since they're serving a role similar to inserters. But keep in mind that even belts should require power but don't, and neither do splitters even though they do something that's arguably much more complex than a loader. Therefore the power requirements should not be punishing. Just enough that if your power fails it stops working would be fine (and if power is limited, your output belt won't be fully packed, for instance).

Not to mention that an express splitter can process two full express belts worth of materials (if it has two express output belts). An express loader accepting or outputting a full express belt worth of items does not seem at all unreasonable.

Slower (and relatively cheaper) versions of the loaders should exist for basic and fast belts as well, which would only operate (obviously enough) at the appropriate speeds. Along these lines, I see no reason for loaders to be a particularly high tech item -- as someone mentioned, we've been using systems like these for over a century, long before we had robot arms that could pick up items from a conveyor belt and do anything with them! (The idea of picking up crumbly raw resources like ore and coal using a tool as inappropriately shaped as a grabby claw rather than a shovel is a bit ridiculous to begin with, but we can pretend they have attachments or the resources are in little boxes.) The various speeds of loaders should be unlocked at the same levels where the various belt speed options are unlocked -- for the slowest, probably in the same tech that you get your first splitters and underground belts.

[Venrob]

One of the major problems I have had in factorio is not being able to load/unload fast enough, even with the fastest possible inserter loading/unloading on all sides of a container. It should require you to have researched every inserter stack size bonus and at least red belts, perhaps make it require a small amount of purple research in addition to a medium amount of blue (50 purple, 200 blue). And make that research be the Basic Loader, only fast enough to flood a basic transport belt. Then require you research the fast and express after that.

The basic loader should be relatively cheap (Not require materials such as plastic/advanced circuits). Maybe make the fast loader require some plastic, and make the express require lubricant, plastic, and advanced circuits.

but wanting the game nearly implayable for the sake of challenge should not discourage 99% from the players to actually play.

If that was a response from me, people that go logistic bot bases, know and will come across problems with bots. As people with belt bases, will know of the limitations like bot players do. And so forth.
I would assume the average player uses a wide mix of belt + bots and everything in between, learning towards bots when the player realizes how useful they are. A loader in this situation as its currently being advertised, would really only benefit people that go heavy on belts. (mostly)

I tend to play with a mixture of bots and belts. I use belts for things when I realize that bots cannot handle the throughput I need (a common problem), and bots when I don't need much, but I need from a distance. I almost never use trains, because it simply isn't practical to load/unload a large amount. I can put the maximum number of inserters on it and have it still not work. It simply isn't fast enough! Something like this is pretty much required for trains to be properly useful in many cases.

HOWEVER, this is NOT the only use for the loader. There are so many times where my furnace output is jammed up, not because the inserters cannot pull out of the furnaces fast enough but because they cannot place onto the belt fast enough. I would be able then to pull from the furnaces into chests and then use the loader from there, limiting the throughput at how fast the inserters can pull from a container to another container, rather than from a container to a belt. So it is useful in many different circumstances!

This would also be useful in bots setups! I have had many times where my bots are limited by how fast I can load into a provider chest! This is again, because inserters + belts is NOT A GOOD MECHANIC RIGHT NOW. There simply isnt a way of making this work. However, by adding a loader, you can make the inserter pull from your furnace/assembler/etc into a steel chest and then loader onto the belt from that chest.

Because it breaks devs' creed "general > specialized", i.e. the whole spirit of Factorio is making complex system from simple elements, and highly specialized element won't fit the concept.

IMO.

So I can see this being useful for bots, trains, and belts. That's pretty GENERAL if I do say so myself.

[RobertTerwilliger]

One of the major problems I have had in factorio is not being able to load/unload fast enough, even with the fastest possible inserter loading/unloading on all sides of a container.

If you really want THAT fast unload, you should use multiple containers, but, well, if you like to store only at single (or maybe you have tough L.bot delivery there, which I think is more likely), then unloading setup should look like this:

storage chest -> fast inserter -> intermediate chest -> 3 fast inserters -> belts
--- copy to all sides you have access to ---

this will provide you 145*3 = 435 items/min per side, 1740 items/min if all sides are avilable.
consider Yellow belt's throughput is about 820 items/min, Red one's 1340, Blue one's 2200
So this setup allows you to almost fill (80%, actually) blue belt from a single storage chest.

Yeah, a bit big, but for such enormous flow it's probably okay. Well, it's just a rhomb with 9-length diagonals))

[AndaleTheGreat]

I just want the end of the line to be able to dump into a storage. My interest in something like what is shown would not be for pulling things from storage but for dumping at the end. If I could I would love to be able to run all my lines into a chest then use an insterter to place items from chest into machine. That would be my preference.

[Deity Link]

As I was bored yesterday, I spent some time creating a new Technology Tree for the wiki until we get the one in 0.13

big image

And I noticed a few oddities. For instance, the Flying tech requires both Advanced Electronics and Electrical Engine. However, Electrical Engine itself requires Advanced Electronics. Couldn't Flying be simplified to only require Electrical Engine? Likewise, Rocket Silo requires both Rockets and Rocket Shooting Speed 5 (which you cannot research before having Rockets in the first place).
Are you guys going to simplify all that?

[vanatteveldt]

In other words...please change the tech tree into a proper (poly)tree, i.e. a fully connected graph that has no (undirected) cycles.

They could also consider dropping other redundant prerequisites, e.g. power armor mk2 doesn't need to depend on alien tech explicitly (as it requires purple tech to research anyway) and doesn't need to depend explicitly on the modules (as you need those to construct the armor itself anway). Similarly, the rocket silo tech doesn't need to depend on rocketry, as it requires purple tech and rocket damage 5, which both depend (indirectly) on rocketry as well. There are many examples like this.

Do you (or anyone) have the tech tree in any "computer readable" format? (ie json, csv, ...) It would be nice to have a go at simplifying and visualizing it automatically...

[Deity Link]

I believe that's what the devs are currently trying to do.

Sadly I made that picture with Gimp and don't have the motivation to transcribe all the data in another format.

[bobingabout]

They could also consider dropping other redundant prerequisites, e.g. power armor mk2 doesn't ... need to depend explicitly on the modules (as you need those to construct the armor itself anway).

That would be undoing a recent change that added in those prerequisites.
People would research the power armor 2 and... oh wait, can't build it because they haven't researched modules high enough yet.

In the opinion of many people, you should have researched everything you need to build a thing before you can research the thing itself.

power armor mk2 doesn't need to depend on alien tech explicitly (as it requires purple tech to research anyway)

in fact in another topic, they are suggesting the exact opposite. Battery tech unlocks science pack 3, they are suggesting that everything that needs science pack 3 as an entry level to research should have batteries technology as a prerequisite.


A difference of opinion in these areas makes it hard to go a route that pleases everyone... I'd rather be forced to research modules before power armor, rather than find I can't build my newly researched armor because I can't make the components. Having to research alien technology... I don't mind either way, it's fairly obvious you can't research it if you can't make the science pack yet, but at the same time, having the prerequisite points you to what you need to research to unlock the science pack.

[Ghoulish]

In the opinion of many people, you should have researched everything you need to build a thing before you can research the thing itself.

And it shouldn't show as available in the research screen if prerequisite research hasn't been done.

[voidei]

I think that the loader could be good if coupled with different tiers, that matches the colour of the belt tiers...but I think it could be a very valuable design!