Friday Facts #393 - Putting things on top of other things

[Shadow_Man]

Exactly like they interact with belts now, why not? The hands phase right through the walls of assemblers but you aren't worrying about that. It's a non-issue.

Ok, let's say. =)
But then how to evaluate the compression, content and proper sides of belt? Do I have copper cables, are these copper cables (or iron plates), do I have enough of them, and on which side of belt they are?

2 words: Transparent cover

[Losash]

[Moderated by Koub : Off topic]

[POPISowyNumer]

Not gonna lie, I do not vibe with the graphic aspect of stacking at all.
Something about building tall towers of loose and/or unwieldy times makes me uneasy.

This will be one of parts of this update that I hope will be modded soon

[crast_b]

Watching the interaction between belt and inserter is one of the fun parts of playing FACTORIO. I think the bulk inserter will increase that fun.

I'm confused by how it's written as if this update increases belt transport.
The maximum belt transport in vanilla factorio should be vehicles on a belt.
Of course, this update will scale up space factory. I'm excited now because it looks like I will be able to build bigger space factories than what I had imagined at FFF385. In my factory system, perhaps, space factories will replace some blocks of the rail grid-based factories.
And, the speed difference between a blue belt and vehicles on a belt is so big that it makes sense to add middle speed belt.
However, when it comes to production line design, a comparison to a production line with a bot group would be a production line with vehicles on a belt.

[Zavian]

Not gonna lie, I do not vibe with the graphic aspect of stacking at all.
Something about building tall towers of loose and/or unwieldy times makes me uneasy.

This will be one of parts of this update that I hope will be modded soon

I agree that stacking probably only makes sense for some items. eg plates and circuits are relatively flat, and stacking multiple items on top of each other is not unreasonable. Ore isn't flat, but to a certain extent you can probably just pile more on top.

But for many taller items like barrels, stacking makes less sense, and for things like nuclear reactors it makes no sense. (It make no sense to even be able to put those on belts in the first place, but whatever. Also they probably should have a stack size of 1).

I'm hoping that with the release of the DLC, not all items will be stackable on belts, and that at least for modding purposes, there will be an item property that specifies the how high individual items can be stacked on belts, so that mods can change that for individual items, without changing stack size for chests/trains.

[bobucles]

Stack inserters move lots (a bulk load) of items, but do not stack items

Bulk inserters do stack items when moving them in bulk.

Underrated post. Stack inserters move bulk, and bulk inserters move stacks. Why do game devs do this?

[mrudat]

I wonder how it would look if items on belts were offset randomly from the center of the lane instead of perfectly centered; that would add some easily visible low-frequency noise to items on belts.

Another thought could be to add random noise (something like a 'film grain' shader, but tied to the item) to any icons with only a single variant.

The ideal would be to add variants to more icons.

I have a vague thought to render a collection of various gears; worm, bevel, internal gear, a segment of a big gear, etc. For circuits, add 1 or 2 letter codes, and short words on processing units. I figure 'W', 'U', 'B', 'E', 'WU', 'BE' and 'WUBE' would be a good start, along with '555', '4004', '741', '8088', '7805', etc.

[Amarula]

I like the idea of adding some 'visual noise' to items. Higher quality items are checked by QA (quality assurance) teams. They could have a 'verified by XX' mark, where the initials XX are drawn from the backer list (the one that is used for radar, lab, and train names).

[Tertius]

I like the idea of adding some 'visual noise' to items.

I'd like to extend that idea of adding a small amount of visual noise to everything shown on the map, at least if zoomed out, to avoid flickering and moire if large similar looking structures are shown.

[Darkona]

I just remembered a previous FFF about Underground Belts not being distinguishable in blueprints, and this new olive color of the highest Tier Belts is insanely close to the yellow of first Tier Belts to the point that even with normal 100% color vision it can be difficult to distinguish those two colors at times. I highly suggest choosing a color that is sufficiently different from the Yellow, Red and Blue that already exist. Sadly the little bit ugly "Uranium Green" and "Cheaty Purple" that most Mods use for highest Tier Belts already, are some of the very few colors that are different enough to be distinguishable in this context.

On the subject of color randomization on displays, I personally have my display as dark as possible to not strain my eyes, so certain things like weird movement are showing very bad ripple effects, worse than lack of vsync, like that last video did somehow.

Hi, new forum account, very old lurker/player.

I'm totally fine with green, but please make it a pretty green. I always like the Tetrad of Yellow/Red/Blue/Green, as I feel like it is "complete" in a sense. But Uranium annoying green is too loud. Maybe some Apple Green with a pastel-like tone would make it distinguishable enough without being annoying. You can try many colors with the color picker used for the player on a mod until it looks ok I guess.

[Henry Loenwind]

One solution for faster belts that don't move faster would be to give us multi-layer belts. Simple rule:

• Inserters can only access the top layer of a multi-layer belt. (Add a flag on the prototype to allow accessing all layers for modded loaders.)
• When a multi-layer belt ends and feeds into a single-layer belt, it behaves like a merging splitter that has the lower level set to priority. (i.e. top-level items drop down if and only if there's space to drop to)
• When a single-layer belt feeds into a multi-layer belt, it feeds onto the same (i.e. the lower) level normally.
• Splitters. Either all splitters are single-level (easy way out), or multi-layer splitters have their UI doubled, and the top and bottom splitters work independently.
• Top and bottom layers are one belt, so they always go in the same direction. Otherwise, this would become insane.

Done. Now, a two-layer belt can move twice as many items as a normal one. And it even fits into the 2.0 theme---there are already multi-level train tracks (just not stacked directly above each other).

Graphics would also be no nightmare. A copy of the belt graphics shifted up, and 2 connecting pillars to hold up the upper layer. Undergrounds need completely redrawn entries and exits. Yes, items in the back row of the lower level would be almost completely hidden, but that's still better than too fast or enclosed belts. And the sidebar preview could show both levels side by side.

(Side note: I wouldn't implement multi-layer belts with more than 2 layers in vanilla, but put the logic for them in for mods. The logic for connecting layers that are beneath other layers 1:1, and merging all layers that feed into the top layer of the target is trivial. The only thing needed is chaining implicit splitters in place.)

[phooenix]

Is it just me or should the names for the "stack inserter" and the "bulk inserter" be swapped now? Seeing as how one of them puts stacks of items onto belts, and it's not the stack inserter

You have a point good sir.

A comment I saw several people mention (though am not worried about it because I love trains) is that these new belts can possibly outperform a train. I'd love to see the math on that.

Another point I'd like to make is that whilst sure, we can restack items via the bulk inserter, I feel that's a very awkward solution. A stacker would be a very welcome addition. Simply place it on a belt and it automatically stacks any items it gets inserted, outputting the resulting item on the side of the belt where the first item of that type got in.

Yeah it got me thinking, new belt 60/sec with 4 items tall would be 240/sec, 4.8 stacks of ore (storage stacks) per second. However that's the easy math, for calculating vs a train would obviously be a 3'rd degree equation where avarage speed is one variable, number of wagons the second variable and distance the third. Assuming loading and unloading speed are the same (absolute max). Just intuitively on short distances from base it could make more sense to have a belt instead of trains but considering that different trains can use the same network obviously makes it superior in the long run.

However i think it feels wierd to "stack" loose pieces of ore on a belt, it would just fall down, it looks wierd aswell (and it clipped through the underground belt in the video even if this is fixable ofcourse). Intermediate produce and up like plates, gears, wires and circuits makes more sense to be stackable.

[FuryoftheStars]

However i think it feels wierd to "stack" loose pieces of ore on a belt, it would just fall down

I kind of agree with you here, at least in terms of the perfect stacks, but it does make sense to make them stackable anyway. It might be possible for them to tweak the graphics of it so that some items, such as ore, look more "piled" on vs stacked on.

[phooenix]

One solution for faster belts that don't move faster would be to give us multi-layer belts. Simple rule:

• Inserters can only access the top layer of a multi-layer belt. (Add a flag on the prototype to allow accessing all layers for modded loaders.)
• When a multi-layer belt ends and feeds into a single-layer belt, it behaves like a merging splitter that has the lower level set to priority. (i.e. top-level items drop down if and only if there's space to drop to)
• When a single-layer belt feeds into a multi-layer belt, it feeds onto the same (i.e. the lower) level normally.
• Splitters. Either all splitters are single-level (easy way out), or multi-layer splitters have their UI doubled, and the top and bottom splitters work independently.
• Top and bottom layers are one belt, so they always go in the same direction. Otherwise, this would become insane.

Done. Now, a two-layer belt can move twice as many items as a normal one. And it even fits into the 2.0 theme---there are already multi-level train tracks (just not stacked directly above each other).

Graphics would also be no nightmare. A copy of the belt graphics shifted up, and 2 connecting pillars to hold up the upper layer. Undergrounds need completely redrawn entries and exits. Yes, items in the back row of the lower level would be almost completely hidden, but that's still better than too fast or enclosed belts. And the sidebar preview could show both levels side by side.

(Side note: I wouldn't implement multi-layer belts with more than 2 layers in vanilla, but put the logic for them in for mods. The logic for connecting layers that are beneath other layers 1:1, and merging all layers that feed into the top layer of the target is trivial. The only thing needed is chaining implicit splitters in place.)

If you are looking at the belt going north-south you wouldn't se anything of the belt under it.

[phooenix]

But for many taller items like barrels, stacking makes less sense, and for things like nuclear reactors it makes no sense. (It make no sense to even be able to put those on belts in the first place, but whatever. Also they probably should have a stack size of 1).

Damn this post made my tummy hurt, now i wanna see someone build a megafactory that mass produces nuclear reactors and fills the new transport belt with 240 reactors per second just for shits and giggles xD

[guy-732]

But for many taller items like barrels, stacking makes less sense, and for things like nuclear reactors it makes no sense. (It make no sense to even be able to put those on belts in the first place, but whatever. Also they probably should have a stack size of 1).

Damn this post made my tummy hurt, now i wanna see someone build a megafactory that mass produces nuclear reactors and fills the new transport belt with 240 reactors per second just for shits and giggles xD

That... would be huge
https://kirkmcdonald.github.io/calc.htm ... 6VXUN/4wM=
16k concrete assemblers... with 4 speed module 3 and 12 speed beacons, I know quality would bring those numbers down but still, never thought we'd need 2k offshore pumps though.

[phooenix]

But for many taller items like barrels, stacking makes less sense, and for things like nuclear reactors it makes no sense. (It make no sense to even be able to put those on belts in the first place, but whatever. Also they probably should have a stack size of 1).

Damn this post made my tummy hurt, now i wanna see someone build a megafactory that mass produces nuclear reactors and fills the new transport belt with 240 reactors per second just for shits and giggles xD

That... would be huge
https://kirkmcdonald.github.io/calc.htm ... 6VXUN/4wM=
16k concrete assemblers... with 4 speed module 3 and 12 speed beacons, I know quality would bring those numbers down but still, never thought we'd need 2k offshore pumps though.

Yeah thats what i figured, i mean just look at those iron ore miners, 130k of them, just for that you'll need to fill up several thousands of ore deposits. It would definitely save you time to first reach max quality of all items and then proceed with building the actual thing. Or you could save up a couple million of ore and just have burst of 240 reactors/sec but that's cheating

[GrimAarkan]

Now we just need long handed bulk insterters.

[varundevan]

since the overlay of parts is depending on the inserters ,
will there is be any new item that compacts the belts to its max limit , A BELT COMPACTOR ?
or will there be an option for the splitter to do it ?/

[YanMaojie]

I hope you make the long arm inserters colored white, and other ones range from yellow for normal, red for fast, blue for stack and green for bulk. This way one inserter color will correspond to one tier of belt.