logic behind requester chest amount when pasted

Post all other topics which do not belong to any other category.

logic behind requester chest amount when pasted

Postby t-lor » Thu Jul 12, 2018 7:29 am

Hi

Not really a bug, but its bugging me none the less ;)

I do a lot of build factory > choose recipe > copy&Paste recipe on the requester chest next to it.
This always works, it requests the correct ingredients.

However the amount it questions varys deeply within recipes.
For instance, if u paste :
prod modules 1. (5 red and green chips) the requester chest requests 7 of each, so 1.5 times the recipe (and way to low imho)
Walls (5 brick) the requestster chest asks for 225 brick. or 45 times the recipe.

Anyone know where those figures are coming from ?
t-lor
Long Handed Inserter
Long Handed Inserter
 
Posts: 52
Joined: Tue Apr 25, 2017 9:28 pm

Re: logic behind requester chest amount when pasted

Postby Klonan » Thu Jul 12, 2018 7:40 am

t-lor wrote:Anyone know where those figures are coming from ?


Info here: https://www.factorio.com/blog/post/fff-213
It aims to keep the chest supplied with 30 seconds of crafting ingredients.
User avatar
Klonan
Factorio Staff
Factorio Staff
 
Posts: 3171
Joined: Sun Jan 11, 2015 2:09 pm

Re: logic behind requester chest amount when pasted

Postby mrvn » Thu Jul 12, 2018 9:51 am

Klonan wrote:
t-lor wrote:Anyone know where those figures are coming from ?


Info here: https://www.factorio.com/blog/post/fff-213
It aims to keep the chest supplied with 30 seconds of crafting ingredients.


I wish that time would be configurable.
mrvn
Smart Inserter
Smart Inserter
 
Posts: 2204
Joined: Mon Sep 05, 2016 9:10 am

Re: logic behind requester chest amount when pasted

Postby BlakeMW » Thu Jul 12, 2018 12:51 pm

I think it could easily stand to limit default requests to 1 stack: if the player truly needs a higher throughput they can always increase it manually. For example, a Heat Exchanger will cheerfully request 1200 copper plates and a fast underground belt is happy to request 1500 iron gears - those quantities are absurd. Limiting to 1 stack wouldn't have any serious effect on the situations where the current logic works well.

Of course this can also just be a matter of these things not having a requester_paste_multiplier or having a daft one (10 for centrifuges? what's the chance I want enough buffer to produce 10 centrifuges? I probably don't even want a total of 10 centrifuges ever because modules and beacons). Most things in the logistic and production tabs are not going to be mass-mass-produced in the same way as intermediates.
BlakeMW
Filter Inserter
Filter Inserter
 
Posts: 882
Joined: Thu Jan 21, 2016 9:29 am

Re: logic behind requester chest amount when pasted

Postby mrvn » Thu Jul 12, 2018 1:25 pm

Some flaws in your examples.

I easily want 16-32 heat exchangers and extending a nuclear reactor in a 2xN setup requires that much. But that is probably rare so it doesn't matter if they are build one after the other with huge gaps.

But fast transport belts? I easily use up 400 of them and then keep right on going using 400 more. If you build them with requester chests than a 30s buffer is totally reasonable. If not too small. I don't think the assembler would run continiously here with only 30s buffer.

Totally with you on the centrifuges. But why do they only take 4s to build? They are hugely complex machines that need high precision. Nothing that you manufacture in a rush. A build time of 30-60s would be absolutely reasonable.
mrvn
Smart Inserter
Smart Inserter
 
Posts: 2204
Joined: Mon Sep 05, 2016 9:10 am

Re: logic behind requester chest amount when pasted

Postby BlakeMW » Thu Jul 12, 2018 8:50 pm

mrvn wrote:But fast transport belts? I easily use up 400 of them and then keep right on going using 400 more. If you build them with requester chests than a 30s buffer is totally reasonable. If not too small. I don't think the assembler would run continiously here with only 30s buffer.


The thing is I don't think that production/logistic things with a build time of 0.5 have that for the sake of machine crafting, it's rather for the sake of hand crafting.

The question is "how probable is it really that an average player is going to want this thing produced non-stop on a sustained basis". Wanting 800 underneathies? I can believe that. Wanting them non-stop? To make Express Underneathies non-stop in Assembler 3 would require 18000 gears/minute, 1000spm is only ~9600 gears/minute and very few players ever get as far as making 1000spm factories.

To me it's absurd to think that the majority of players would intend an assembler to produce underneathies nonstop and thus the logic of having a large enough buffer to produce them nonstop is not applicable. And that freak of nature who is making a 10000spm factory and really does want those underneathies pumped out non-stop (because he's going to grind that UPS into the DIRT) can just go to the damn requester chest and turn the quantities up.
BlakeMW
Filter Inserter
Filter Inserter
 
Posts: 882
Joined: Thu Jan 21, 2016 9:29 am

Re: logic behind requester chest amount when pasted

Postby mrvn » Tue Jul 17, 2018 9:42 am

It's not about needing them non-stop. It's about placing a blueprint and then needing them non-stop for 5 minutes.
mrvn
Smart Inserter
Smart Inserter
 
Posts: 2204
Joined: Mon Sep 05, 2016 9:10 am

Re: logic behind requester chest amount when pasted

Postby BlakeMW » Tue Jul 17, 2018 5:12 pm

mrvn wrote:It's not about needing them non-stop. It's about placing a blueprint and then needing them non-stop for 5 minutes.


Is this how you actually play? Because it seems to me the design pattern of placing an assembler in advance and producing a stockpile of the things in a chest in advance of needing them is overwhelmingly common.

Secondly, some conclusions are strange: To sustain production of Assembling Machine 3 at 0.5/s (modified by assembler) you would need a 1:2:4 ratio of Assembling Machine 1,2,3. I guess I could imagine some players doing this for the sake of perfect ratios but I find it hard to imagine them thinking "I'm really going to need to produce these assembling machine 3's at a rate of 2.25/s, so I'd better drop down 4 assembling machines to produce assembling machine 1".
BlakeMW
Filter Inserter
Filter Inserter
 
Posts: 882
Joined: Thu Jan 21, 2016 9:29 am

Re: logic behind requester chest amount when pasted

Postby FrodoOf9Fingers » Tue Jul 17, 2018 9:50 pm

I think that it'd be cool if the copy paste mechanic from assemblers was an additive function. I often have a single chest providing to multiple assemblers that make different things, and being able to copy paste all of them onto a single chest would be nice.

It'd help this situation to, pasting the same recipe over and over should increase the amount of each item in the recipe.

This is in line with how copy paste works in a document, pressing ctrl-c multiple times concatenate (adds).
FrodoOf9Fingers
Long Handed Inserter
Long Handed Inserter
 
Posts: 67
Joined: Sat Apr 29, 2017 11:13 pm

Re: logic behind requester chest amount when pasted

Postby mrvn » Wed Jul 18, 2018 2:34 pm

BlakeMW wrote:
mrvn wrote:It's not about needing them non-stop. It's about placing a blueprint and then needing them non-stop for 5 minutes.


Is this how you actually play? Because it seems to me the design pattern of placing an assembler in advance and producing a stockpile of the things in a chest in advance of needing them is overwhelmingly common.

Secondly, some conclusions are strange: To sustain production of Assembling Machine 3 at 0.5/s (modified by assembler) you would need a 1:2:4 ratio of Assembling Machine 1,2,3. I guess I could imagine some players doing this for the sake of perfect ratios but I find it hard to imagine them thinking "I'm really going to need to produce these assembling machine 3's at a rate of 2.25/s, so I'd better drop down 4 assembling machines to produce assembling machine 1".


I stopile iron plates without limits. There is nothing else to do with iron ore (except much later and only small quantities). But if you then convert them all to belts then what do you build your oil factory out of when you need to make more oil products? So belt stockpiles are limited (although that's probably the biggest limited stockpile I keep). Assemblers are even more limited, to just one slot in a chest. For other things like roboports I even use a wire and enable the exserter to roboports == 0. Which means one in the chest, two in the assembler, ingreediences to build some more in the requester chest (if that is used). Overall I try to not stockpile things I don't need while keeping the materials to build what I need when I need it.

In some cases I go a different way. For example with Bobs+Angels you have a lot of excess crushed stone to get rid of. Crushed stone gives you landfill, stone, stone bricks, stone wall, gates, stone pipe, stone underground pipes. For those I tend to wire up inserters and program them for e.g. stone underground pipes < stone pipes. So I keep them balanced but build as much as crushed stone comes in.

Anyway. No, I don't build assembling machines 1/2/3 in the perfect ratio. But in such chains the slowest step should run non-stop when the output is needed. Otherwise it slows down all the steps. If your gear assembler isn't producing gears as 0.5s a gear then you have to wait forever for anything else to get built.
mrvn
Smart Inserter
Smart Inserter
 
Posts: 2204
Joined: Mon Sep 05, 2016 9:10 am


Return to General discussion

Who is online

Users browsing this forum: Heavylaus and 8 guests