Loader Poll - Trying to get an idea of what the opinions are

[Koder]

How is it more elegant to have them limited by the devs? If they end up being chest only, then anybody who wants to use them for rail cars is now forced to play the game the way the chest people want to play it. Whereas if up to the player, both parties would be happy instead of only one party.

One of the reasons this game is so engaging because of the existing balances/challenges put in place by the developers that we players need to engineer ways to work through. I guess I just hope it stays that way. As I mentioned in a recent post, I will wait until I play the first iteration before I post further opinions about it

[Zeblote]

Please not... More Splitters due to a Loader not being able to run at Fast/Express belt speed would render the Loaders plain and simple horrible.

Don't know if you have seen the screenshots of my experimental setups a few pages earlier (viewtopic.php?p=133747#p133747), but the setups will already be "challenging enough" without imposing a speed limit on the loaders. Also there is simply no room in layouts for the additional loaders/splitters as you suggest. If you would force an artificial speed limit at loaders then nearly nobody will use them because of how it is so much easier and more space efficient to just use regular inserters grabbing directly from the belt when comparing the necessary splitter contraptions. Due to how a Splitter is required with each Loader it is already like as if the Loader is 2x2 in size due to the Splitter overhead.

And the Loaders also try to avoid the splitter-balancing mess in the first place... So having to deal with 1/3s would be counterproductive.

Loaders are supposed to feed into containers like chests, assemblers and trains, so you wouldn't have inserters grabbing stuff from them.

Like this:



Or this:





Now you could use them to make this



But is that really better than this?





In my opinion they are already balanced fine, as it makes no sense to use them in many places people are worried about.

[zytukin]

How is it more elegant to have them limited by the devs? If they end up being chest only, then anybody who wants to use them for rail cars is now forced to play the game the way the chest people want to play it. Whereas if up to the player, both parties would be happy instead of only one party.

One of the reasons this game is so engaging because of the existing balances/challenges put in place by the developers that we players need to engineer ways to work through. I guess I just hope it stays that way. As I mentioned in a recent post, I will wait until I play the first iteration before I post further opinions about it

Very good points. Always good to test and experiment before making a final decision.
And it is fun to work around limitations.

Putting some thought into it in regards of realistic and role playing limitations:
Rail cars are big, so something needs to be able to reach or extend into them to stack and grab things at the ends of the car away from the doors. So I could see not being able to load or unload a rail car plausible. Unless it *only* worked on them and (if you used your imagination) extended *into* the rail car to stack or un-stack items.

If no to a rail car, a belt feeding into it and it feeding into an assembler or chest is plausible. It would just push everything into it. Pulling stuff out of an assembler or chest is also plausible for the same reason. But few assemblers would be able to make use of it efficiently. Assemblers with speed boosts making Gears, Wires, or Stone Walls are the only things I can think of off hand. An Assembler making stone walls with 2 level 2 speed boosts needs 4 fast inserters putting stone bricks into it non-stop from a fast belt to keep it running full speed.


So, taking the above thoughts into consideration, I am now actually leaning more to rail cars only, simply due to the higher capacity and being more useful. A device that extends into them to stack and un-stack large qtys of items.


Still against the "shouldn't be implemented" option as anybody who doesn't like them could just not use them at all.

[MeduSalem]

Loaders are supposed to feed into containers like chests, assemblers and trains, so you wouldn't have inserters grabbing stuff from them.

Like this:

[...]

In my opinion they are already balanced fine, as it makes no sense to use them in many places people are worried about.

If they are going to implement them like that I will never use them because I can't stand the layouts resulting from this horrible mess of direct insertion for the sake of my life. I would rather continue using 14 inserters or whatever to feed from belts into chests, trains etc than using direct-inserting Loaders. And I would condemn it as a cheap cheat for the rest of Factorio's development, like I already do with certain other things like for example the notorious Underground Belt Cheat.

I like Klonan's idea of using Inserters as intermediate step to move Items from the Loader into the target machine/item much more. MUCH more because that would at least give some interesting layout options instead.

[sillyfly]

...And I would condemn it as a cheap cheat for the rest of Factorio's development...

Feel free to condemn whatever aspect of the game (whether implemented or theorized) you want

[Zeblote]

If they are going to implement them like that I will never use them because I can't stand the layouts resulting from this horrible mess of direct insertion for the sake of my life. I would rather continue using 14 inserters or whatever to feed from belts into chests, trains etc than using direct-inserting Loaders. And I would condemn it as a cheap cheat for the rest of Factorio's development, like I already do with certain other things like for example the notorious Underground Belt Cheat.

I like Klonan's idea of using Inserters as intermediate step to move Items from the Loader into the target machine/item much more. MUCH more because that would at least give some interesting layout options instead.

Where did you plan on using the other ones?

I mainly want these to unload trains without inconvenient inserter rows, make small buffers without having to build rediculous contraptions to not lose compression, and get those damn copper cables around fast enough

[The Phoenixian]

So I have another idea to proffer:

If we have just one level of loader and unloader, and it feeds onto a yellow belt at full speed, a red belt at half speed and a blue belt at one third speed, then as your factory advances, the complexity of using loaders increases if you wish to maintain full throughput. At the beginning of the game you have something you can just slap down. In the mid game, you are asked to use a splitter. And at the end of the game, the game now asks you to not just split, but split into thirds for best material efficiency.

Certainly any experienced player can do so easily, assuming they choose to bother, but it's a nice little niggling problem for newbies and for the rest, there's a lack of space efficiency that now comes with loaders on high level belts is a consistent challenge.

Please not... More Splitters due to a Loader not being able to run at Fast/Express belt speed would render the Loaders plain and simple horrible.

Don't know if you have seen the screenshots of my experimental setups a few pages earlier (viewtopic.php?p=133747#p133747), but the setups will already be "challenging enough" without imposing a speed limit on the loaders. Also there is simply no room in layouts for the additional loaders/splitters as you suggest. If you would force an artificial speed limit at loaders then nearly nobody will use them because of how it is so much easier and more space efficient to just use regular inserters grabbing directly from the belt when comparing the necessary splitter contraptions. Due to how a Splitter is required with each Loader it is already like as if the Loader is 2x2 in size due to the Splitter overhead.

And the Loaders also try to avoid the splitter-balancing mess in the first place... So having to deal with 1/3s would be counterproductive.

From a quick look, I don't think that any of those theoretical setups are affected by a speed limit on loaders because they already use two or three splitters per belt line. Likewise the train pictures posted earlier. The very fact that you used three or more splitters already solves the problem.

But you're right that it's worth looking into what it takes to break it.

If the loader is working at a speed of a yellow belt, then it has a throughput of a little more than 10 items a second, 5 items a second from half a belt. So, what you're feeding needs to produce or consume 10 items a second or 5 items a second from half a belt for loaders to be a problem. That in mind, I it's plausible for electronic circuits with wires on the belt and more than one smart inserter assembly machine with electronic circuits on half the belt, but not for anything else without speed modules. Even iron gear wheels in yellow assemblers are just about equal to the speed of the loader with a 2 assembly machine to 1 loader ratio.

For recipes that produce at 2/sec in a yellow assembler or chemical plant, you will need 4 machines to equal the rate of a loader. So you should be able to happily feed a belt with sulpher, plastics, electronics, and gears with no problem. Only copper wires, which are produced at 4/sec, would make it lock up.

So for normal use it's borderline at worst but if what you're after however is a way to use loaders to speed boost a small number of assembly machines far beyond the normal production rate then yes, it breaks.

So, with all that laid out, do you still say that breaks too much unique utility to be worth experimenting with?

(For reference, the main intent of the idea of loader limits is that they'd apply to things like buffers, places where they are always going to be the limiting factor.)

EDIT: Ending question seemed too condescending, nonsensical, hopefully it reads more like an honest question now.

[Mad Inventor]

I like the idea of a loader with small internal buffer. It should only interact with belts, and you will still need an inserter to move items. Basically, it should be a part of the belt system that applies inserter stack bonus to belts.

I think this is the most reasonable suggestion so far. It's been elaborated on by others in this thread but the early quote seems to sum up the idea.

The main objection to implementing loaders seems to be that they will trivialize Inserters, and more importantly, trivialize the challenge/fun of making layouts for resource buffers, sorters, routers, junctions etc.
I think there's an important difference between making these layouts simpler, though, and just making them more compact for a given throughput.
If loaders/unloaders just act as intermediaries between belts and inserters, I think it might make designing these layouts even more interesting, because there will be another kind of basic element to integrate into them. It won't replace the need for inserters, chests etc, and the gameplay that's already built around these will remain.

The benefit that the loader DOES bring - when you choose to make it part of a layout - is saving space and repetition. Instead of replacing the need for an inserter entirely, it will reduce the need for 6/12/18 inserters to fill or empty a belt, dividing the number of inserters you need by the inserter stack size.
The designs I see now all seem to have long rows of identical inserter/belt/inserter elements, and I don't think it will detract from the gameplay to make these rows shorter. But it will add to the gameplay in several ways -

First, depending on the dimensions of the loader and the space/energy it needs, it won't always be worthwhile to use it versus taking things on/off the belt directly. This adds another tradeoff/calculation to make, finding the point at which you need to add loaders to get more throughput. And it adds more design complexity in fitting the loader in with the other things in the efficient ways.
Second, by making similar designs more compact, there will be new possibilities for fitting these into the larger layout of the factory. If you need less space for rows of inserters you can use that space for things that weren't practical before, like more elaborate control systems with circuits and combinators, more intricate production layouts with beacons and 4-6 ingredients, etc. Less repetition in one part of the factory could allow more complexity in other parts.
A related benefit is that you can also be more flexible fitting in these setups where the terrain, not the factory, is the limitation. The screenshots of the huge belt layouts I see never seem to have any bodies of water, forests, or large resource patches (that you could build over but would rather mine). This will allow you to build your layouts in places where you couldn't or wouldn't put them before.

Quite apart from the question of whether the loader adds interesting gameplay itself is also whether it reduces tedious, repetitive play that is not interesting or challenging. To me, it IS interesting, and challenging, to tinker with combining belts, chests, inserters etc. But building 9 inserters all doing the same thing is not.
Building loaders to "boost" the inserters is more interesting than either "only loaders" (if they make inserters redundant) OR "only inserters" (if loaders aren't implemented at all). I don't want the loader as it was shown in the FF post, because it wouldn't do anything that inserters don't already do. But I really like the idea of the loader as an accessory to inserter-belt systems.

Correct me if I've misunderstood any of the mechanics or gameplay involved. My perspective on this is that of a very new player.
But then I don't like the assumption that newer players want less complexity/challenge and the more experienced crowd are always in favor of more of it. I've only been playing Factorio since the Steam release. I haven't yet managed to make any of those complex designs I see on the forums work for me. But I'm enjoying the planning, trying, failing and return to planning immensely. I wouldn't still be playing, and reading the guides, forums, wiki etc if I didn't.

[Khaylain]

Quotes

[quote="Zeblote"][quote="MeduSalem"]Please not... More Splitters due to a Loader not being able to run at Fast/Express belt speed would render the Loaders plain and simple horrible.

Don't know if you have seen the screenshots of my experimental setups a few pages earlier (viewtopic.php?p=133747#p133747), but the setups will already be "challenging enough" without imposing a speed limit on the loaders. Also there is simply no room in layouts for the additional loaders/splitters as you suggest. If you would force an artificial speed limit at loaders then nearly nobody will use them because of how it is so much easier and more space efficient to just use regular inserters grabbing directly from the belt when comparing the necessary splitter contraptions. Due to how a Splitter is required with each Loader it is already like as if the Loader is 2x2 in size due to the Splitter overhead.

And the Loaders also try to avoid the splitter-balancing mess in the first place... So having to deal with 1/3s would be counterproductive.[/quote]

Loaders are supposed to feed into containers like chests, assemblers and trains, so you wouldn't have inserters grabbing stuff from them.

Like this:



Or this:





Now you could use them to make this



But is that really better than this?





In my opinion they are already balanced fine, as it makes no sense to use them in many places people are worried about.[/quote]

I'll add my voice to the camp of saying it's probably pretty good as is, but it should probably consume power, seeing as they seem to me like a more powerful inserter (useful in the late game, where even fast inserters seem lacking at times). The idea of it also autobalancing the lanes of a belt is a very good feature, I don't know if this is intended to work with belt-->feeder-->belt, but if it is I would praise the possibility. I know I could balance the lanes with splitter to double splitter to sideload each half into single splitter to recombine, but I feel like adding the option in the late game to simplify it would make a lot of sense.

I'm not certain about use-restrictions, like only feeding chests or only feeding belt from chest. There might be some "gamebreaking" uses for it, but that's just human nature. We've always tried to find a better way to do something, and we haven't stopped yet. I think at least feeding to and from belts and chests is the minimum for it to be usable, and train wagons might also be a very good application. It would allow you to stock up from miners into chests at a train station, then when the train comes the feeders will "feed" the train as much as it can in a short amount of time, and the time a train needs to be stopped goes down.

I'm not sure about feeding into assemblers, I can see uses for it in making wire and other quick recipies as others has stated, but it will of course use more space than just inserters. You also won't be able to use the trick of having two belts be accessible for quick feeding as easily, as you'd need a splitter per belt per feeder. Now you can just snake one belt in and out of range of the fast inserters and have the other belt go under that portion.

TL;DR: Like the idea of it being more powerful inserter, like the idea of autobalancing lanes on belts, not sure about restrictions in input and output locations.

[taiiat]

they're pretty insane, and the sorts of costs and upkeep to make balanced... would also probably make them not worth bothering for the first 99% of the game.

resource costs and crafting time alone doesn't do it, since that's just a time and logistics gate... if to exist at all, has to have some sort of upkeep. probably... something that's annoying so that you use them sparingly and only in places/situations where you have the space and sanity to spend on the massive throughput.

- - - - -

Nope, NO barreling for every fluid...! The amount of additional items would be ridiculous (one for each barreled fluid type) and even more so for mods adding additional fluids. I'd rather get rid of the damn barrels in the first place.

that's a solvable problem. the crafting menu and the like is fine for the number of things that exist but that's guaranteed to go up - so inevitably things are going to probably start becoming drop down / breakout menus in order to avoid having the entire menu scroll.

- - - - -

hm, making Loaders a buffer that works with Inserters, rather than being a direct upgrade type thing, could be interesting.
i'm always for those sorts of things in games, direct upgrades are... not as interesting as changing the way you interact with things and think about using them.

[byronczimmer]

I keep seeing the desire to make these "Loaders" cost a lot in terms of science, materials, upkeep or some esoteric combination of all three.

Let's dial it down a notch and ask ourselves what this technology is compared to, say, the following:
1/ Conveyor belts
2/ Robotic Arms
3/ Robotic Arms smart enough to only pick up certain objects
4/ Trains
5/ Flying Robots that carry stuff

Just to be sure, we're talking about a 'hopper' or 'chute' system which temporarily stores stuff in a bin and then makes it available to the next step of the process, which essentially boils down to 'move it someplace else" or "use it".
These things have existed in one form or another since ancient times. Hoppers/Chutes certainly existed before Trains. By my reckoning, they existed before Robotic Arms, which are one of the first available technologies in the game.

A conveyor belt that 'ends' either gets stopped (things stop moving) (current default behavior), or it dumps into a hopper until it's full.

A hopper feeds another belt or can have items taken from it by another means (inserters) if it contains items.

It really doesn't need to be much more difficult yet there this strange effort to make it overly complex and prohibitively expensive to utilize.

If anything, look at a system that allows the player to put these 'hoppers' on conveyor belts (or some other kind of movement system) and then be manipulated as units. Similar to oil barrel filling/emptying, have a way to package one stack of a 'thing' into a 'bundle of that thing'. Then allow a bundle to be picked up and inserted into the machine intended to process it when it needs material to keep working.

That is the goal, right? To be able to handle larger volumes of material in a logical way.

[againey]

GoliathMrk1 suggested a "belt chest" idea in the FF133 thread which sounds quite a bit like the hopper idea mentioned in this thread, and it inspired me to make a mockup.

Expanding on my description in that other thread, the essence is that it is a one square chest-like object which has a single input which can be supplied by a belt, a single output on the opposite side which can supply a belt, and which has a relatively small storage capacity. Beyond that, it would behave just like a chest, and therefore the stack size upgrades will apply to any inserters which exchange items between it and another container. So unlike the loaders as presented previously, inserters are still used in conjunction with belt chests. And unlike the recently proposed heavy inserter, the stack size bonus would behave exactly as it currently does, rather than in the additional way of dropping multiple items onto the belt in quick succession.

Since this object replaces a chest rather than an inserter, you're free to use whatever inserter type you wish, but now you're stuck with a single chest type. To overcome that, you can just use an inserter (with stack size upgrades) to move items between the belt chest and the desired chest type, as shown in the following mockup (the yellow highlighted smart chests with arrows are to be interpreted as oriented belt chests):

Factorio - 2016-04-11-15-42 - Belt Chest.jpg (124.68 KiB) Viewed 8833 times

• (1) Pulling stuff from a belt into a belt chest, and then into a larger steel chest.
• (2) Pulling stuff out of a large steel chest into a belt chest to feed the belt.
• (3) Pulling stuff from a belt into a belt chest, and then using three smart inserters to separate iron, copper, and steel into three steel chests.
• (4) Pulling stuff from a belt into a belt chest, and then using two smart inserters to separate green and red circuits into two other belt chests which each feed an output belt.
• (5) Passing stuff through a belt chest, but pulling some of the items out into two steel chests, one on each side.
• (6) Passing stuff through a belt chest, and also inserting stuff into the belt chest from a steel chest, which will get merged with the pass-through items.
• (7) A train station using (2) in order to provide a larger buffer due to steel chests than is provided by the belt chests.
• (8) A simple train station using just belt chests. If the belt chests fill up from the train faster than the belt empties them, then the station won't be as efficient as it could be.

As a small detail, there perhaps could be a small delay (0.1s? 1s? 5s?) when items are fed into and out of the belt chest via belts, though items exchanged via inserters would behave as they always have with chests. So if it's pulling items from a belt, each item would be unavailable for that delay interval between when the item entered the belt chest's square and when it is registered as an item in the chest. Imagine that it's occupied in some internal belt/organizing system that makes it arranged conveniently for usage of the stack size upgrades. Likewise, when an item is pulled from the belt chest in order to be dumped onto the output belt, it is queued for that interval before it finally appears on the output belt. If it's a pass-through belt chest (that is, it has both an input and output belt), then perhaps an additional delay is added where the item sits in the chests momentarily for any inserters that might grab the item before it gets dumped onto the output belt.

[Omarflyjoemacky]

Ok - I don't know if this discussion is dead. But simple point...

Loaders are for us folks who don't want the complexity of circuits. I've played over a thousand hours now, and I use the loader mods only because they aren't "part" of the game. Loaders and inserters have their own uses in any factory - the coexist quite nicely. I don't think they are overpowered either. Watch any of Arumba's campaigns - he agrees.

However - to my point. I stumbled upon this thread looking for any kind of "authentic" graphics. Fell in love instantly with the images posted. Does anyone have a link to those images for loaders? Or are they a developer secret that I should make a kickstarter for? I'd pay $20 as a DLC for the loader graphics on their own...

Are those loader graphics available at all? As a download? Purchase? There are other graphic sets out there, but these are beautiful.

lLzl78h.png (341.77 KiB) Viewed 8446 times

Please forgive me if this request has already been addressed!

[Zeblote]

I made that in gimp by patching a bunch of other sprites together. (specifically, 2 underground fast belts and some combinator)

Not really sure what graphics loaders needs, but probably at least 8 of those... 4 directions, load and unload.

[steinio]

What should the yellow > sign mean? Is this a combinator function?

Greetings steinio

[Zeblote]

What should the yellow > sign mean? Is this a combinator function?

Greetings steinio

It's from the compare combinator, just thought it looked fancy with the arrow showing the direction the loader is loading.

[Omarflyjoemacky]

I actually didn't realize that was the end of a combinator ... again I don't do circuit stuff. That in itself is hours and hours, especially with .15 coming.

It sure looks great. You could definitely post them in the mods section - I think they would be very popular.

[Zeblote]

It'd be easy to make the full set if we had the original 3d models available, but not so much by mashing sprites together... if someone else wants to do the same tedious thing 8 times, go ahead

Oh, and it was made using the older, big combinator sprites that don't seem to exist anymore.

[Omarflyjoemacky]

I'll definitely take a crack at it. I did something like yourself, mashing sprites together (see below). It's simple, for sure. But man - when I saw yours I almost choked.

I'll take a shot at it when the high res graphics come out for combinator... perhaps it'll be a little easier than coloring every pixel.

loader-structure.png (22.5 KiB) Viewed 8269 times

What I find amazing is that most of the community is happy with that grey placeholder arrow!

[xng]

I really like the idea of loaders, no matter what graphics it uses, but I would like to see the limitation like how you connect pipes to the assemblers. What I mean is that you have 2 spots, one input and one output, if you choose to connect loaders/offloaders there you get away free of electricity (like with pipes), but you are limited to how you can use them.