Re: Does a beacon consume too much energy?
Posted: Sat Jul 09, 2016 2:06 pm
Nah I think I understand it and I get your point. BTW I've tested my 0.12 base and I'll send you the map and modpack now.
I guess with megabase you mean something like launching a rocket every other minute...? So basically stuff that is running 24/7 and therefore becomes predictible, which allows to use perfect ratios?Qon wrote:For megabases with beacons you want to have one big block of beacons and assembly machines producing everything. Serparation increases energy wasted by beacons that can't cover 8 machines. And then you can't shut down beacons for individual machines any more without also affecting the production of other machines producing something else while sharing beacons. It's also superior for making expansions.
On the other hand if you do it at the correct ratio you either have all machines running or none, so you can turn off the whole megabase with an energy switch instead.
Even with looping belts back the items would eventually build up if you don't have enough consumers but too many producers. Brings the loop to a grinding halt eventually (which may be pretty bad if the loop is carrying more than 1 item type on a single belt side, because starting up from there may be impossible once all assemblers are deadlocked with "no resources" because the belt didn't move to bring the lacking one in range of an inserter).Qon wrote:If you have enough production then items will build up on the belt enough for all inserters to pick up items as long as you aren't looping the belts back.
Yes, that is what I meant. If you don't have a resource sink that is predictable things change of course. But it's usually rockets. I should specify that.MeduSalem wrote: I guess with megabase you mean something like launching a rocket every other minute...? So basically stuff that is running 24/7 and therefore becomes predictible, which allows to use perfect ratios?
Because some recipes may run only once in a while (like crafting items for expanding your base, etc), which is unpredictible but you still may want to use beacons on the intermediate items to save resources or whatever. There exact ratios may depend on circumstances like how many of these unpredictible recipes run in concurrence and if one is sharing intermediate resources between them to save footprint (like having one spot to produce ECs to share for all these unpredictible recipes because it might not make much sense to produce everything locally). There it is better if the beacon+assembler arrays would be seperated and switched on/off on demand.
Yeah I meant that a looping belt that would be kept sparse by not inserting more than necessary or items looping far back enough for assemblers of other types to use the items or similar case. Which is of course not always true for all looping belts. The point was that you could build an endlessly looping sparse belt to prevent inserters from picking up the items, not that no looping setup would ever halt or fill up. Production would halt then if it isn't allowed to flow on to the belt.MeduSalem wrote:Even with looping belts back the items would eventually build up if you don't have enough consumers but too many producers. Brings the loop to a grinding halt eventually (which may be pretty bad if the loop is carrying more than 1 item type on a single belt side, because starting up from there may be impossible once all assemblers are deadlocked with "no resources" because the belt didn't move to bring the lacking one in range of an inserter).Qon wrote:If you have enough production then items will build up on the belt enough for all inserters to pick up items as long as you aren't looping the belts back.
I think in most of these cases Prod3+Beacon is kind of wasteful, because these machines need to run for a long time until they've amortized the cost of the modules (if Prod modules are even possible, e.g. with robot frames and engines).Qon wrote:Agree with separation being a good idea. But having locally produced intermediates is also a good alternative in some cases so it's not always clear which is best. A bit is up to personal taste here. Recipies for expanding your base are extremely cheap for the most part. Roboports, blue belts, robots and module 3 are quite demanding and are required in big amounts though.
Not really sure if you are criticising my statement or even responding to what I said. Your statements are only vaguely related to my statements. I'll try to respond...siggboy wrote:I think in most of these cases Prod3+Beacon is kind of wasteful, because these machines need to run for a long time until they've amortized the cost of the modules (if Prod modules are even possible, e.g. with robot frames and engines).Qon wrote:Agree with separation being a good idea. But having locally produced intermediates is also a good alternative in some cases so it's not always clear which is best. A bit is up to personal taste here. Recipies for expanding your base are extremely cheap for the most part. Roboports, blue belts, robots and module 3 are quite demanding and are required in big amounts though.
So I guess if you actually mass produce stuff that requires green or red circuits, it's better to get those from the place where you make the circuits for the rocket factory.
If you're min-maxing to the point where you want to disable individual beacons to save energy, then it's probably very ill-advised to put a lot of resources into modules that you will never be able to recoup.
I said you need a lot of them. In a productivity focuses factory you do. The costs are regained in a few hours or minutes depending on what you use them for. In a rocket silo it's regained immediatly, otherwise AntiElitez would not use them in his sub 2h speedrun. I don't consider tens of hours a long time for a megafactory so modules are worth it for me.I think in most of these cases Prod3+Beacon is kind of wasteful, because these machines need to run for a long time until they've amortized the cost of the modules (if Prod modules are even possible, e.g. with robot frames and engines).
I said it was up to personal taste and it really depends on many factors. Now you assert that it's always definitly the best way to do it by taking the circuits from the rocket factory. But then you might possibly have to rebuild your rocket factory to accomodate the logistics and sacrifice your rocket production speed and so on. You can't just say that. Better for what exactly? Energy? Modules used? Yes limiting your production is great for limiting your energy used/second or modules required. I don't want to limit my production.So I guess if you actually mass produce stuff that requires green or red circuits, it's better to get those from the place where you make the circuits for the rocket factory.
Why would I never be able to recoup the resurces? Energy isn't the only resource. Why can't I min-max my energy and my productivity if I want to make something as good as possible?If you're min-maxing to the point where you want to disable individual beacons to save energy, then it's probably very ill-advised to put a lot of resources into modules that you will never be able to recoup.
Optimization, the source of our enjoyment in Factorio q:siggboy wrote: Conversations like this one, that deal with "efficiency" in an infinite sandbox world where nothing really matters, only make sense in the context of min-maxing, or optimization for its own sake, if you will.
Did you calculate that crafting time affects recoup time negativly? I don't find it obvious. Show me how you proved that please?siggboy wrote: However if you want a lean, or efficient, or min-maxed or "nice" solution, then putting Prod3 into a machine that makes Flying Robot Frames is ugly, because those two modules have to amortize their cost in that machine you've put them into, and that's barely possible (because Flying Robot Frames has a long manufacturing time, so that machine will probably never have cycled often enough to recoup the investment in the Prod3 modules).
Well I only disagree with people who are wrong xDsiggboy wrote:Maybe you agree with that, maybe you disagree, it was not really my goal to disagree with you here, not everything I write is in order to disagree with you .
People smarter than myself have done it two years ago already: Productivity Module MathQon wrote:Did you calculate that crafting time affects recoup time negativly? I don't find it obvious. Show me how you proved that please?
Ahh that thread. Haven't seen it in a long time... It's definitely one of the classics and probably the source of all the PM3/SM3 craziness.siggboy wrote:People smarter than myself have done it two years ago already: Productivity Module Math
Well I agree that crafting time is a factor, but not the only factor. My point is that amount of ingredients and their worth is should also be considered since the resources per second can be quite high even for slow recipies. And that is key, resources per second. Or maybe "value" per second (and investment cost) is even better if you want since you save more than just raw resources. And flying robot frame is fairly high on that list by DaveMcW. Engines are very low though.siggboy wrote:People smarter than myself have done it two years ago already: Productivity Module MathQon wrote:Did you calculate that crafting time affects recoup time negativly? I don't find it obvious. Show me how you proved that please?
(But if you want to see me being wrong about this point and then eventually conceding then you should go here )
There's nothing "crazy" about spamming PM3+SM3 if you're making a rocket factory, it's simply the most effective way if you're min-maxing (and if you make more than only a couple of rockets).MeduSalem wrote:probably the source of all the PM3/SM3 craziness.
Well once you have all the robot frames you need you can take out the modules and use them in the factory you needed the robots for if you aren't constantly requiring more of them. You don't lose the modules.siggboy wrote:Robot Frames are high, [...] but usually one doesn't make enough of them to amortize the cost of the Productivity modules; and then it's just a net negative to use them.
It wasn't meant like that... and more like a expression that I know that min/maxing stuff can get out of control quite fast because how I'm a min/max guy myself... and because of how everyone has a different oppinion what are the most important characteristics to min/max around.siggboy wrote:There's nothing "crazy" about spamming PM3+SM3 if you're making a rocket factory, it's simply the most effective way if you're min-maxing (and if you make more than only a couple of rockets).MeduSalem wrote:probably the source of all the PM3/SM3 craziness.
That's of course true, and usually what I end up doing. It's just that plugging PM3 into everything mindlessly (even if the product is "valuable") and then not think about it again is a costly mistake. The modules should ideally always be in a machine that is cycling/producing something.Qon wrote:Well once you have all the robot frames you need you can take out the modules and use them in the factory you needed the robots for if you aren't constantly requiring more of them. You don't lose the modules.
What would you guys say if I told you that by sheer accident of forgetting to set an output limit in one of my maps I have 750 PM3s, 500 SM3s and 500 EM3s sitting around in storage chests doing nothing but mock me?siggboy wrote:That's of course true, and usually what I end up doing. It's just that plugging PM3 into everything mindlessly (even if the product is "valuable") and then not think about it again is a costly mistake. The modules should ideally always be in a machine that is cycling/producing something.Qon wrote:Well once you have all the robot frames you need you can take out the modules and use them in the factory you needed the robots for if you aren't constantly requiring more of them. You don't lose the modules.
Well all you have to do is to use them to not waste the resources q:MeduSalem wrote: What would you guys say if I told you that by sheer accident of forgetting to set an output limit in one of my maps I have 750 PM3s, 500 SM3s and 500 EM3s sitting around in storage chests doing nothing but mock me?