Conveyor belt speed (blue)

[dood]

40. You get 40.

No you do not get 40.
1) If you have a red belt main bus and siphon off a full red belt
2) You upgrade only the main bus but not the side lane, with the old splitters, you would get a maximum of half the blue belt, which is 40/2 = 20 into the side red belt. Which isn't full capacity.

If you got a full red belt before, you would get a full red belt after.
Where's the part that supposedly breaks?

[Engimage]

Belt speed does not effect game performance much. Base belt speed is 1 pixel/tick. Tile sprite is 32 pixels long and this resulted in belts having 32 "slots" for items on each lane. Item occupies 9 slots (one for itself and 4 reserved free spaces on both sides). So basicly belt speed is a multiplier of base speed causing just more shift iterations per tick than base.
However with belts having increased speeds compared to current blue belt inserters start to fail as they have limited rotation speeds. So increasing maximum belt speed requires increasing inserter speeds slightly.

Other from that there is no other problems with belt speeds. It is a purely design choice currently.

[Sir3n]

Belt speed does not effect game performance much. Base belt speed is 1 pixel/tick. Tile sprite is 32 pixels long and this resulted in belts having 32 "slots" for items on each lane. Item occupies 9 slots (one for itself and 4 reserved free spaces on both sides). So basicly belt speed is a multiplier of base speed causing just more shift iterations per tick than base.
However with belts having increased speeds compared to current blue belt inserters start to fail as they have limited rotation speeds. So increasing maximum belt speed requires increasing inserter speeds slightly.

Other from that there is no other problems with belt speeds. It is a purely design choice currently.

Thanks so much, this was the answer i was looking for.

[JimBarracus]

If you got a full red belt before, you would get a full red belt after.
Where's the part that supposedly breaks?

The point is that if you split one compressed blue belt you won´t receive two compressed red belts.
Which would be possible, if blue belt speed was x 4 and not x 3.

[fiery_salmon]

If you got a full red belt before, you would get a full red belt after.
Where's the part that supposedly breaks?

The point is that if you split one compressed blue belt you won´t receive two compressed red belts.
Which would be possible, if blue belt speed was x 4 and not x 3.

You can easily split two blue to get three red (with priority splitters).

[dood]

If you got a full red belt before, you would get a full red belt after.
Where's the part that supposedly breaks?

The point is that if you split one compressed blue belt you won´t receive two compressed red belts.
Which would be possible, if blue belt speed was x 4 and not x 3.

He said he pulls a full red belt off a red bus and that changing the bus to blue would somehow no longer get him a full red belt.



How?

[Sir3n]

LIke i said, with the old splitters, not the newer ones with inbuilt priority selection. The number 40/s was before the new splitters so i wasn't taking those into account

EDIT
An example is, you have your main bus set with like lanes on all sides built with red belts. Maybe you take a full red lane off without splitting or anything, now you realise, later down the bus, you need more iron / copper than you planned for, and you don't want to tear the whole bus down and rebuild it just yet or something
At this point, you'd replace the main bus red belt with blue belt, if you used the old splitters to split it, you wouldn't guarantee the side belt gets full compression red belt it used to have unless you used circuitry / funny contraptions.

If blue belt was 2x red belt speed, you can just fully replace the main bus and wherever you used to take a red belt off, you just get a blue splitter and you keep the side belt unchanged.

[darkfrei]

However with belts having increased speeds compared to current blue belt inserters start to fail as they have limited rotation speeds. So increasing maximum belt speed requires increasing inserter speeds slightly.

High speed belts are not for inserters, but for main bus. If you need some resources, just take the part from main bus and connect with your production with low speed belt.

[dood]

LIke i said, with the old splitters, not the newer ones with inbuilt priority selection. The number 40/s was before the new splitters so i wasn't taking those into account

EDIT
An example is, you have your main bus set with like lanes on all sides built with red belts. Maybe you take a full red lane off without splitting or anything, now you realise, later down the bus, you need more iron / copper than you planned for, and you don't want to tear the whole bus down and rebuild it just yet or something
At this point, you'd replace the main bus red belt with blue belt, if you used the old splitters to split it, you wouldn't guarantee the side belt gets full compression red belt it used to have unless you used circuitry / funny contraptions.

If blue belt was 2x red belt speed, you can just fully replace the main bus and wherever you used to take a red belt off, you just get a blue splitter and you keep the side belt unchanged.

You run bus lanes directly into things without splitting?
What are you doing with your life? You always split.

And your issue is only an issue if the entire bus gets constantly consumed in its entirety nonstop so no backup happens in which case it's time for more bandwidth so the bus backs up and you get fully compressed belts everywhere.
It has only do with uneven belt speeds on a superficial level, even if blue belts were 2x red speed and a splitter off of a compressed, moving blue belt would produce a red belt, sure, that would look pretty but it wouldn't change anything about the fact that this bus is starved.

The new priority splitters don't change much about this concept.
All that changed is that they completely starve factories down the line rather than slowly dissolving the belts into nothingness if you consume more than you push through in which case the solution is the same, bigger bus, more throughput. Chop chop.

[McDuff]

I use the mod that does this and it makes a lot more sense for me - it's actually how I thought it worked before and was confused when it wasn't. 4 yellows compress to 2 reds compress to 1 blue. I also use the "vaccuum belts" mod because I like the idea of a higher tier of belt that works slightly differently. I'm also trying to work out how to properly use deadlock's stacking beltboxes. For me the idea of branches and fractal designs is a great part of Factorio that produces interesting and readable designs, so I would be perfectly happy with this logical change.

[Sir3n]

It has only do with uneven belt speeds on a superficial level, even if blue belts were 2x red speed and a splitter off of a compressed, moving blue belt would produce a red belt, sure, that would look pretty but it wouldn't change anything about the fact that this bus is starved.

While that is true, you are assuming that every machine has equal value at any given moment. Sometimes you know you don't have enough production just yet but just want to build it already so you don't have to think about it again later on or maybe you just don't care very much about it and just want leftover stuff to filter down there once you satisfy all the important stuff.

There was no elegant way of doing that before priority splitter.

It is also a lot harder to immediately visually differentiate between not having enough production and insufficient belt speed when you can't guarantee a fully compressed belt. You have to let things run for a while until everything backs up like you said, which can take a very long time depending on how your build is set up.

I guess i worded what i meant in a somewhat unclear manner.

[dood]

It has only do with uneven belt speeds on a superficial level, even if blue belts were 2x red speed and a splitter off of a compressed, moving blue belt would produce a red belt, sure, that would look pretty but it wouldn't change anything about the fact that this bus is starved.

While that is true, you are assuming that every machine has equal value at any given moment. Sometimes you know you don't have enough production just yet but just want to build it already so you don't have to think about it again later on or maybe you just don't care very much about it and just want leftover stuff to filter down there once you satisfy all the important stuff.

There was no elegant way of doing that before priority splitter.

It is also a lot harder to immediately visually differentiate between not having enough production and insufficient belt speed when you can't guarantee a fully compressed belt. You have to let things run for a while until everything backs up like you said, which can take a very long time depending on how your build is set up.

I guess i worded what i meant in a somewhat unclear manner.

I don't really understand what you're building.
The only constant resource drain you really have to worry about is science.
Everything else comes and goes in bursts unless you forgot to limit chests or inserters.
Are you trying to build an early 500 spm facility off of 4 iron, 4 copper belts or something?

[Jap2.0]

LIke i said, with the old splitters, not the newer ones with inbuilt priority selection. The number 40/s was before the new splitters so i wasn't taking those into account

EDIT
An example is, you have your main bus set with like lanes on all sides built with red belts. Maybe you take a full red lane off without splitting or anything, now you realise, later down the bus, you need more iron / copper than you planned for, and you don't want to tear the whole bus down and rebuild it just yet or something
At this point, you'd replace the main bus red belt with blue belt, if you used the old splitters to split it, you wouldn't guarantee the side belt gets full compression red belt it used to have unless you used circuitry / funny contraptions.

If blue belt was 2x red belt speed, you can just fully replace the main bus and wherever you used to take a red belt off, you just get a blue splitter and you keep the side belt unchanged.

That's why you had things like these.

[quyxkh]

You can easily split two blue to get three red (with priority splitters).

That's always been easy. A 2:3 splitter is

[dragontamer5788]

In my experience, blue-belts aren't worth it until well into the post-game. Red is already expensive at 11.5 iron per tile for only 26.6 items/sec. This is 7.66x more expensive than yellow, for only 2x the performance. Blue is 31.5 iron per tile + lubricant for 40 items/sec. This is 2.7x the iron cost (ignoring the costly crude oil costs!) for only 1.5x the performance.

Blue improves the speed of train station unloading, it also allows mines with high-productivity to output more quickly. But these are endgame issues. Mining patches without modules at productivity +20% (Level 10 Mining Prod. research) are handled by red belts rather easily, and Train stations are still being fed by blue inserters. (Green inserters are likely too costly, and Inserter Bonus 7 is also endgame tech). As such, red belts are more than sufficient for midgame, and even early post-game levels of base building.

Blue belts should only be used in poorly-planned mainbus which cannot have parallel lanes added to them. Beyond that, its far more cost efficient to build your base bigger (ex: add another lane of belts)

[dood]

In my experience, blue-belts aren't worth it until well into the post-game. Red is already expensive at 11.5 iron per tile for only 26.6 items/sec. This is 7.66x more expensive than yellow, for only 2x the performance. Blue is 31.5 iron per tile + lubricant for 40 items/sec. This is 2.7x the cost for 1.5x the performance.

Resource cost is immaterial for 1 time investments.

The longer you cling to red belts, the higher the opportunity cost of having to replace thousands of less than max tier belts gets and if you juggle multiple tiers of belts in your inventory, it costs you precious inventory space and logistics slots.

The comfort of only having to worry about 1 tier of item type far outweighs that drop in the ocean lubricant and iron cost.

[dragontamer5788]

In my experience, blue-belts aren't worth it until well into the post-game. Red is already expensive at 11.5 iron per tile for only 26.6 items/sec. This is 7.66x more expensive than yellow, for only 2x the performance. Blue is 31.5 iron per tile + lubricant for 40 items/sec. This is 2.7x the cost for 1.5x the performance.

Resource cost is immaterial for 1 time investments.

Only true in the endgame.

The longer you cling to red belts, the higher the opportunity cost of having to replace thousands of less than max tier belts gets and if you juggle multiple tiers of belts in your inventory, it costs you precious inventory space and logistics slots.

Why would I "upgrade" when I simply build a better sub-base?

"Upgrade" cycles of blowing away old factories is a solved problem. Deconstruction planner everything to storage chests, then nuke the storage chests. Hell, its probably faster to just nuke the sub-bases (although you still need construction bots to move the dropped items somewhere). As you've said earlier: in the post-game, resource costs are completely immaterial for 1-time investments (but only in the endgame).

EDIT: If you really care about saving costs in the endgame (probably for OCD reasons, but aiming for "saving resources" when you're launching ~300,000 iron ore worth of rockets every few minutes is... suspect), then use logistic bots to move the old belts to your shopping mall district for upgrading.

---------------------

EDIT2: The goal of the midgame is to grow to endgame-size as soon as possible. That's why costs are important. Yellow and Red belts (with proper planning) allow you to grow faster. Red Belts save on inventory space by 2x, although if you've got a well sized logistic roboport area, maybe yellow belts will remain fine (since construction bots can pickup yellow belts from the mall automatically). So its not like inventory space is a problem with proper gameplans past ~purple science. Construction bots are more than usable in blue-science tier, very early midgame, and that's roughly the time a good player should be starting to spam blueprints.

[dood]

Only true in the endgame.

The "endgame" has to start at some point.
You can stay there, fretting over every broken egg or get over here and make the mother of al omelettes.

Also it's not only the endgame.
People have been all like "but that's too expensive" when I tell them to box everything they need in with turrets right from the beginning and would rather make small "cheap" boxes to put outposts in, let their trains run through biter territory and then get on here asking for silly things like turrets on trains when that causes all sorts of problems that would have been solved already had they aimed higher from the very beginning.

"Upgrade" cycles of blowing away old factories is a solved problem. Deconstruction planner everything to storage chests, then nuke the storage chests. Hell, its probably faster to just nuke the sub-bases (although you still need construction bots to move the dropped items somewhere).

Wow. Now that is an actual waste.
People do that?

[dragontamer5788]

Only true in the endgame.

The "endgame" has to start at some point.

The endgame is roughly the time when you can afford Productivity3 modules in every machine, and every smelter. If you aren't at that point, you need to accelerate and grow your base faster. Using red-belts and yellow-belts temporarily, Steel Furnaces, AM2 machines, blue-inserters instead of stack inserters, etc. etc. gets you there faster.

Each PM3 module requires 2000 copper, 1080 iron, and 120 plastic. Two go into each smelter, four goes into each AM3. You'll also need to mass-produce speed modules and beacons, which are similar in costs. If you're aiming for 1kspm, you'll also need to mass produce 6GW of solar panels (That's 142,857 solar panels + a large number of accumulators).

"Upgrade" cycles of blowing away old factories is a solved problem. Deconstruction planner everything to storage chests, then nuke the storage chests. Hell, its probably faster to just nuke the sub-bases (although you still need construction bots to move the dropped items somewhere).

Wow. Now that is an actual waste.
People do that?

That's my plan personally. Why should I care when resources are basically limitless in the endgame? The issue is getting to the endgame faster. Midgame is an incredibly difficult time and resources are quite constrained. Its difficult to bootstrap yourself to the point where you can afford the modules needed for endgame builds.

But seriously, add it up. 1000 red belts is what? 11,500 iron.

In contrast, the 26 PM3 modules needed for a blue-belt of iron (~13 Elec. Furnaces x 8x beacons per furnace with 2x Speed3 modules per beacon) is 28080 iron + 52000 copper. Oh, for PM3 alone. The speed modules double that cost to 56160 iron + 104000 copper. That's for a singular blue-belt of iron, and a 1kspm base needs like 40x that design. An order of magnitude more expensive. Endgame is incredibly expensive, and the thousands of red belts you used during the midgame is basically an ignorable cost.

Endgame begins when you stop caring about resources, because you have enough PM3 / Speed3 modules to accomplish your goals.

[dood]

The endgame is roughly the time when you can afford Productivity3 modules in every machine, and every smelter.

Well I've never reached "endgame" then. I get bored when starting to prod3 the rocket silo and climbing down the chain of intermediates.
Prod mods in smelters is some new level of desperate.

Blue belts are nowhere near that level.
You secured a big chunk of iron ore and some oil, you're good to go and you should take the plunge earlier rather than later.
When you start getting t3 assemblers sounds about right. Nowhere near endgame.
Fun fact: changing the t2 assemblers and belts of a build to t3 makes the entire thing 50% faster without starving the belts.