Your primary issue with the first refinery is that the pipes are not through-put optimise it 1. Use underground pipes for any distance farther than a length of 2, as eatch pipe object adds additional resistance to flow 2. Use pumps as valves and through put boosters, pumps allow incredible amounts of fluid to be forced into pipes upwards of 12k/s 3. Use minimal amount of buffer storage, eatch tank without direct connection to a pump reduces how much peak demand can draw as there will be some fluid that cannot be drawn
Great advice sakari_119. I was thinking he should have been using pumps. Also, if are coming upon on limits of fluid throughput, you don't have to put all your refineries in one long serial chain. You probably have lots of oil fields. Just route them into a more parallel setup. This also creates redundancy should a pipeline succumb to biters, you have an alternate source of petroleum products without a single point of failure. Finally, lots of robots flying around can bring even the strongest CPU to its knees. I try to limit the jobs my robots are doing, like fulfilling my personal logistics requests or moving expensive outputs (e.g. electric engines, blue circuits, etc.) to where they are needed in my factory so I don't have to buffer hundreds or thousands of them on my main bus via belts.
I've been trying to understand robots for what seems like ages. Thanks for the explanation, who knew all I needed was someone to show me simply what each chest is for ( which I saw in your other tutorial ) and with this video my excitement for factorio has returned!
RIP UPS. The piped version seems constrained on the pump jack side and the bottleneck apparently is not the pipes. IOW, add more jacks. Not an apples-to-apples comparison.
I want to be able to recycle my barrels by melting them back into steel. Not for any kind of chenanigans, but because I am going to overproduce and end up with boxes full of them. And it makes sense you could do that, or at least melt them back for a few iron plates.
Super crisp mic you're using mate. Thanks for the wicked video! I never considered barreling before. Now I have another question: using pipes only, how many refineries would it take before non-beaconed production is capped? How many if using in-line pumps? I feel that I'd still want to pipe to the max of a pipe's efficacy before using bots.
I think pipes fall around 1800 liquid/second before they start to have issues. So if oil has a 4 second craft time then take the crude oil cost/4 * (x) =1800 then solve for x to get your number. I just woke up so if my maths wrong there im sorry lol
I love this idea! although I think pipes and this new barrel way of oil prossessing are kinda equal in effecincy, I think some sort of hybrid setup may be even more effecient than either of them alone. Thank you for the ideas to use im my own factory!
How do you know how many bots to add? Per refinery chunk i suppose. Since you'd want to have X number of bots per refinery. I guess it could just be eyeballed to see if the chests are getting too full or too empty
Well robots move items on an as needed basis. If you have "too many" they will sit idle until its their turn to be slaves. Technically speaking you would want just under "enough" so they pick up 4 at a time and you dont have bots getting 1 item, but thatd outside of the scope of what i care to do lol
@@spiritwind02 comlete rework of fluid system will nullify all of the pros of this setup leaving only cons. And when the design only has disadvantages I can confidently call it unusable
Except that by isolating the network like this, you're highlighting the major issue with robots: They're good for a couple of things over short distances (until you start unlocking white science upgrades to them), but you ideally don't want one large robot network. The problem here, is that to really use this effectively, i.e. sending those barrels to somewhere you'd need them, you'd need either the large logistic network, or a combination of train networks and local logistics networks. Also, adding these extra steps means you need to set up an exact production of barrels. Make too few, and this is slower than pipe spaghetti, too many and you've got barrels every where, even when you don't need them to be. Furthermore, now, instead of just having trains (if you go the mixed logistics route) transporting your fluids to where they're needed, you're adding trains to transport barrels to drop offs, as well as trains to transport the barreled fluids AND the machines to both fill and empty the barrels. I could see this being very usable with a good circuit network controlling things like barrel production and fluid requests and such. However, a VERY large number of factorio players know little more about the circuit network aside from turning on lights. You're also trading one UPS problem for another here. Would be interested to see this set up when the DLC finally drops, though, as with all the improvements to robots and such, one massive logistics network might be a feasible thing, and this would really shine there.
I tried factorio, won 3 times. It is enough to build 6 refineries to produce enough materials to launch a rocket. What for do you need so many refineries and even speed them up? Wha is that for?
launching the rocket is far from the actual goal. Sure, you "win" by doing it, but it also unlocks a new science to research. The end goal of factorio is to have enough of a factory that your computer can't handle it any more.
I’m not sure this system truly addresses the issue of one oil type not being consumed as much as the others, resulting in the dreaded ‘Output Full’ scenario. I think this system is interesting and I’d like to see if a little circuit controlled balancing of cracking couldn’t be added to the system too…
I wouldnt call it a stupid comment :o the setup definitely doesnt do that. Its not a mathematically correct version of it. I personally almost never do math with factorio. I also dont do blueprints so its an inexact science on this channel
For me personally, the pipe-way of doing it is still superior. Your version is better than your early example, but that was pretty poorly optimized. Using so many speed buffs on only one pipe input was the problem. Just use some pumps or a maximum refinery per input pipe, and it should solve itself. The barrel stuff does take up more space overall, and you still have the problem with one overfilling. One wire from tank to pump that goes to cracking, is so simple to set up.
Don't agree with this solution. Use prod modules, so it uses less oil. And if you are pipe constrained, you will probably not have enough oil to feed it anyway.
Interesting approach but the video is all over the place: repetition, clicking frantically everywhere while really not doing anything, and for the times you do make something you don't allow us to look at it for more than 0.5 sec -- very frustrating. I'd recommend announcing (even in the big lines) what you're going for and then bring us there slowly. It was hard to follow, not because of the complexity, but because you're scattered around like if you were just put on the spot. I see big potential but you need to script your videos and trim the fat (narrative + video edits). I forced myself to get to the end of the video and see your build, but you totally overlooked the problem you tried to tackle. Input/output flow rates -- what are the numbers? What about pumps? Did your robot stuff improved on the pipes?
I think of this more is a demonstration rather than an explanation. I make tutorials separately that I'll teach people and explain how certain mechanics work. Most of the factorio community just looks up blueprints and how to do stuff when they don't know what to do.. So ultimately making an independent explanation as to why this function's better than using regular pipes is not necessary
Your primary issue with the first refinery is that the pipes are not through-put optimise it
1. Use underground pipes for any distance farther than a length of 2, as eatch pipe object adds additional resistance to flow
2. Use pumps as valves and through put boosters, pumps allow incredible amounts of fluid to be forced into pipes upwards of 12k/s
3. Use minimal amount of buffer storage, eatch tank without direct connection to a pump reduces how much peak demand can draw as there will be some fluid that cannot be drawn
Great advice sakari_119. I was thinking he should have been using pumps.
Also, if are coming upon on limits of fluid throughput, you don't have to put all your refineries in one long serial chain. You probably have lots of oil fields. Just route them into a more parallel setup. This also creates redundancy should a pipeline succumb to biters, you have an alternate source of petroleum products without a single point of failure.
Finally, lots of robots flying around can bring even the strongest CPU to its knees. I try to limit the jobs my robots are doing, like fulfilling my personal logistics requests or moving expensive outputs (e.g. electric engines, blue circuits, etc.) to where they are needed in my factory so I don't have to buffer hundreds or thousands of them on my main bus via belts.
I've been trying to understand robots for what seems like ages. Thanks for the explanation, who knew all I needed was someone to show me simply what each chest is for ( which I saw in your other tutorial ) and with this video my excitement for factorio has returned!
I love seeing comments like these. In so glad you are enjoying the game again :) this game is honestly awesome and im glad i could help
RIP UPS. The piped version seems constrained on the pump jack side and the bottleneck apparently is not the pipes. IOW, add more jacks. Not an apples-to-apples comparison.
Very clever design! Always nice to make use of a typically-ignored gameplay element (barrels).
Glad you like it :D I hope it helps build you a new system that works well
I want to be able to recycle my barrels by melting them back into steel.
Not for any kind of chenanigans, but because I am going to overproduce and end up with boxes full of them.
And it makes sense you could do that, or at least melt them back for a few iron plates.
Use the reverse factory mod
Space age has a recycler :D
Super crisp mic you're using mate. Thanks for the wicked video! I never considered barreling before.
Now I have another question: using pipes only, how many refineries would it take before non-beaconed production is capped? How many if using in-line pumps? I feel that I'd still want to pipe to the max of a pipe's efficacy before using bots.
I think pipes fall around 1800 liquid/second before they start to have issues. So if oil has a 4 second craft time then take the
crude oil cost/4 * (x) =1800 then solve for x to get your number. I just woke up so if my maths wrong there im sorry lol
I love this idea! although I think pipes and this new barrel way of oil prossessing are kinda equal in effecincy, I think some sort of hybrid setup may be even more effecient than either of them alone. Thank you for the ideas to use im my own factory!
absolutely. Im glad this interests you. I hope it serves your factory well :D
How do you know how many bots to add? Per refinery chunk i suppose. Since you'd want to have X number of bots per refinery. I guess it could just be eyeballed to see if the chests are getting too full or too empty
Well robots move items on an as needed basis. If you have "too many" they will sit idle until its their turn to be slaves. Technically speaking you would want just under "enough" so they pick up 4 at a time and you dont have bots getting 1 item, but thatd outside of the scope of what i care to do lol
This setup will be usable only for the next 51 days before 2.0 releases
What changes makes it unusable?
@@spiritwind02 comlete rework of fluid system will nullify all of the pros of this setup leaving only cons. And when the design only has disadvantages I can confidently call it unusable
Do you have blueprint book for all those logistic bots based set ups?
Im a nerd and never use blueprints. Sorry. :(
Just use pumps to provide fluid everywhere
Except that by isolating the network like this, you're highlighting the major issue with robots: They're good for a couple of things over short distances (until you start unlocking white science upgrades to them), but you ideally don't want one large robot network. The problem here, is that to really use this effectively, i.e. sending those barrels to somewhere you'd need them, you'd need either the large logistic network, or a combination of train networks and local logistics networks. Also, adding these extra steps means you need to set up an exact production of barrels. Make too few, and this is slower than pipe spaghetti, too many and you've got barrels every where, even when you don't need them to be. Furthermore, now, instead of just having trains (if you go the mixed logistics route) transporting your fluids to where they're needed, you're adding trains to transport barrels to drop offs, as well as trains to transport the barreled fluids AND the machines to both fill and empty the barrels.
I could see this being very usable with a good circuit network controlling things like barrel production and fluid requests and such. However, a VERY large number of factorio players know little more about the circuit network aside from turning on lights. You're also trading one UPS problem for another here. Would be interested to see this set up when the DLC finally drops, though, as with all the improvements to robots and such, one massive logistics network might be a feasible thing, and this would really shine there.
I tried factorio, won 3 times. It is enough to build 6 refineries to produce enough materials to launch a rocket. What for do you need so many refineries and even speed them up? Wha is that for?
This is a very late game factory, and the material is used in mass.
launching the rocket is far from the actual goal. Sure, you "win" by doing it, but it also unlocks a new science to research. The end goal of factorio is to have enough of a factory that your computer can't handle it any more.
@@Bokusa mid game, sure but very late game this won't work since it use bot, 1k spm already required over 20 refinery
I’m not sure this system truly addresses the issue of one oil type not being consumed as much as the others, resulting in the dreaded ‘Output Full’ scenario. I think this system is interesting and I’d like to see if a little circuit controlled balancing of cracking couldn’t be added to the system too…
I guess you never claimed it did. You could literally be taking the barrels to the cracking area I suppose. Sorry for stupid comment
I wouldnt call it a stupid comment :o the setup definitely doesnt do that. Its not a mathematically correct version of it. I personally almost never do math with factorio. I also dont do blueprints so its an inexact science on this channel
For me personally, the pipe-way of doing it is still superior. Your version is better than your early example, but that was pretty poorly optimized. Using so many speed buffs on only one pipe input was the problem. Just use some pumps or a maximum refinery per input pipe, and it should solve itself.
The barrel stuff does take up more space overall, and you still have the problem with one overfilling. One wire from tank to pump that goes to cracking, is so simple to set up.
I think you made it too complex. You can solve it just using pumps in riht place.
Interesting design.
Unfortunately, 2.0 will be here in two months, which completely resolved this issue
Ik and im so excited!!!!
what is being added in 2.0 that will resolve this?
Don't agree with this solution.
Use prod modules, so it uses less oil.
And if you are pipe constrained, you will probably not have enough oil to feed it anyway.
Nyoom
Nice voice, but this is a really expensive system and does not work properly.
It works just fine actually
@@Bokusa Use a red cable.
Interesting approach but the video is all over the place: repetition, clicking frantically everywhere while really not doing anything, and for the times you do make something you don't allow us to look at it for more than 0.5 sec -- very frustrating. I'd recommend announcing (even in the big lines) what you're going for and then bring us there slowly. It was hard to follow, not because of the complexity, but because you're scattered around like if you were just put on the spot. I see big potential but you need to script your videos and trim the fat (narrative + video edits). I forced myself to get to the end of the video and see your build, but you totally overlooked the problem you tried to tackle. Input/output flow rates -- what are the numbers? What about pumps? Did your robot stuff improved on the pipes?
I agree. This video takes 65 million years to get to the point. And repeats just as much.
Dont let Destiny see this video
Is destiny a different creator? Im not aware of who or what youre refering to
@@Bokusaa political/video game streamer that no-lifed space exploration for like 8 months straight until he finished it, driving viewers insane.
But he’d probably let OP pump his wife to say thanks.
Yikes
Good content but you repeat yourself many many times.
You can say that again.
Come on, dude... What did you actually explain?
I think of this more is a demonstration rather than an explanation. I make tutorials separately that I'll teach people and explain how certain mechanics work. Most of the factorio community just looks up blueprints and how to do stuff when they don't know what to do.. So ultimately making an independent explanation as to why this function's better than using regular pipes is not necessary
@@Bokusa My dude, with all respect, you can solve the issue with three red cables and a pump.
This solves a lot more issues by conpacting this and not requiring outside piping at all
the best part of Factorio is.. there is no one right way to play. This setup looks awesome and I will be using it in my next game.
@@procheeseburger_2 It's my way, my way or the highway. Buy!
cool robot design but like… pumps?
robots dont need pumps :o they pumped all their iron in their training