Only 100 mass storage crates, really?

[NathanHoltBeader]

Recently when I looked at the stats on a mass storage crate on my small setup, it included a statement that there were more than 100 crates in the system and that I should consider breaking it up. Am I really supposed to make do with mass storage systems with a measly 15K capacity when there are so many recipes that require thousands of metals? My plan is to make a 128x128 room for mass storage so that I can supply all my machines with the materials they need.

What sort of setup do you suggest as an alternative? I can't think of a way that wouldn't be really tedious.

[DjArcas]

Nothing's stopping you making it as big as you like, but the inverse square law means that as you grow it, it'll get exponentially slower. I can't change geometry, I'm afraid.

I generally set up tiers for separated mass storage - one for T1 materials, one for T2, then one for tertiary crafted materials. (Obviously you'd never store bars - you'd always upcraft them!)

You're suggesting a single mass storage with over 2 million capacity? Might work. I'm sure other people will comment with better setups than mine.

[steveman0]

Putting everything into a single storage is very inefficient. It takes a long time to load and unload items from it using conventional means and it's entirely unnecesary. Although costs add up to thousands of raw ore or bars you never benefit by storing all of that at once.

You should actually aim to have the smallest storage possible to maintain your production line at the highest throughput. Capacity stored is never important but the max throughput is and you'll generally be better off in this regard by maintaining smaller, quick storages. Only make them as big as you need to hook up the relevant inputs and outputs.

If you get beyond 50 crates you should probably begin thinking about how you can split it up into separate, specialized units as they'll each be quicker independently than one hulking mass. The key is working out a streamlined flow of resources to get needed materials to each unit cell. The freight system coming in the Adventures Pack is designed to make this process much easier so you can maintain even the tinest production unit cell.

[Compugasm]

Originally posted by NathanHoltBeader:

I can't think of a way that wouldn't be really tedious.

The 128x128 plan is actually tedious. Each item you put into storage is going to need 1-5 inputs to keep the line from jamming. Then, to get that item out of storage, you'll need 1-5 outputs. Worst case, that means 70 ports to make, just to get ores in/out of your storage. That's a lot of things to make by hand.

Instead, try storing finished products. It takes 64 ore to make a Research Pod. Storing the pod means building less ports, but the storage itself can be smaller, like 5x1x5 and be very fast.

[NathanHoltBeader]

Throughput instead of reserves could be in interesting strategy. I do tend to play while watching videos with my brother, so some waiting for machines to process things isn't too hard for me. I could argue right off that you'd want to store some coper and tin for making tier 1 processing tools. Also, there's a point at which there are enoug kinds of upcrafting that it becomes easier to just store bars and upcraft as needed. For instance various recipes need tin plates, lightweight machine housings, and primary upgrade modules, in additon to simplified research pods for the laboratory.

[Siro]

Aim at throughput. If you have a hopper at an intermediary step, you'll have the intermediary products available when the end product backs up.

[DjArcas]

Originally posted by NathanHoltBeader:

Throughput instead of reserves could be in interesting strategy. I do tend to play while watching videos with my brother, so some waiting for machines to process things isn't too hard for me. I could argue right off that you'd want to store some coper and tin for making tier 1 processing tools. Also, there's a point at which there are enoug kinds of upcrafting that it becomes easier to just store bars and upcraft as needed. For instance various recipes need tin plates, lightweight machine housings, and primary upgrade modules, in additon to simplified research pods for the laboratory.

In this case, I'd have one MS for Tin Bars, then one for Tin Pates, LWMH, and Primary Upgrade Modules, and then just a hopper as a target for the Research pods.

[DjArcas]

Note : There's nothing stopping you having a mass storage of 128x128 - (16384 blocks). The game's just warning you that it might be a bit slow!

[Myrianne]

Originally posted by DjArcas:

Obviously you'd never store bars - you'd always upcraft them!

To be fair, the new freight system works great with mass storage, so I use it extensively for bar distribution purposes. There are still some recipes that take bars after all (like conveyor belts).

[Compugasm]

Originally posted by Siro:

Aim at throughput. If you have a hopper at an intermediary step, you'll have the intermediary products available when the end product backs up.

Here is what siro is talking about:
https://steamcommunity.com/sharedfiles/filedetails/?id=1483168599
And if the conveyors off the hopper are motorize, the line from the storage is saturated.

[steveman0]

I tend to build small unit production facilities. 20-30 crates or so is all you need to build an outline for ports to connect to while running a bunch of manufacturing machines in the middle. An induction charger under it for things that need power. Freight connects systems together to provide components across each unit. Eventually they'll each get roomed up as a standalone unit.

https://steamcommunity.com/sharedfiles/filedetails/?id=1491791504
https://steamcommunity.com/sharedfiles/filedetails/?id=1491791719
https://steamcommunity.com/sharedfiles/filedetails/?id=1491791473

[NathanHoltBeader]

I've had some interesting thoughts about this.

First, I probably wouldn't have even used mass storage systems if I were going for throughput. I'd have used a maze of convayors and storage hoppers distrubute materials to production stations. If I used 10x10 mass storage systems, I'd stil need to do it. I once built a base at least a hundred meters accross and it was jam packed with production stations. It was only worth doing because I was able to make a mass storage system that covered the whole area so I was able to tap the resources stations needed where they were needed.

Second, unless input drones are lobotomized, a huge mass storage system can still be fairly efficient thanks to stocking inputs. If a storage drone needs to look for an empty crate, it doesn't pick one at random. It picks the closest empty crate. That way, if the limit is 1500 items, it just stores them in 10 crates clustered around the input. If outputs are placed nearby, they'll work efficiently. If the storage system extends for hundreds of meters around, the crates that matter will still be just a few meters away.

Third, tiny mass storage systems go agains an important principle: If it's worth doing, it's worth overdoing!

That's why I've dicided I'm still going to overdo it!

[steveman0]

Originally posted by NathanHoltBeader:

I
Second, unless input drones are lobotomized, a huge mass storage system can still be fairly efficient thanks to stocking inputs. If a storage drone needs to look for an empty crate, it doesn't pick one at random. It picks the closest empty crate. That way, if the limit is 1500 items, it just stores them in 10 crates clustered around the input. If outputs are placed nearby, they'll work efficiently. If the storage system extends for hundreds of meters around, the crates that matter will still be just a few meters away.

But if you're doing this... it's really no different than operating with smaller storage systems. Aside from the case where a drone a long distance away locks the crate which limits access to local drones making them less efficient. In the end it doesn't really benefit you beyond convenience whereas a well designed system can provide far higher throughput keeping it useful well into the late game. Sure you can overdo it as an alternative but you'll need to shove it all in a room somewhere to keep performance up.