More in depth production

In Space Engineers 1 production was super linear, it always went Survival kit>basic assembler>Basic refinery>assembler>refinery>whatever modules you want to use and then more recently the prototech blocks I guess. This served its purpose well enough by acting as roadblocks from progressing to more advanced blocks before you had progressed far enough in the game to produce these production blocks. The issue was that typically by the time you had a refinery and an assembler, you were barely into the game and you were done improving your production. At most, you might go for additional refineries and assemblers to improve throughput, but that isn't a new or interesting challenge to overcome, it's the same challenge you've already overcome a second time!

With that in mind, I would like to suggest the following production mechanics!

Initially, you start with suit crafting! Suit crafting is can be done anywhere, but has some serious downsides. First of all, it will be incredibly slow, on the scale of 1 item every 5 seconds and only runs while you're in your menu, and it would burn through a quarter of your suit's electricity per item while it did so, but for this suit energy will slowly recover over time if it's not being actively used, and the engineer is in sunlight. The goal of this is to allow someone with just their suit the ability to build up from nothing, rather than having to start with a drop pod. Later on it would also help with spot repairs where you just need 1 specific item where you can disassemble the item and use it to make what you need, as a QOL feature.

Stage 2 would be the survival kit, which would be able to do more than in Space Engineers 1, being able to properly refine ores and assemble a greater variety of items. The downside of the survival kit though would be that it has the yield inefficiency of the basic refinery in SE1.

At this point we have a split since 2 different sets of production items can be built here! First we have the micro refinery and micro assembler. These would be blocks built around assembling tools, and getting the required resources to make those tools. As such these blocks are way smaller, sized to be small enough that you could realistically fit them on a ship the size of a small grid ship in SE1, and also way slower than even suit crafting, but they're very power efficient, and can technically build any item, though you wouldn't want to try to use them on an industrial scale. Theyre great for small ships or for quickly producing a specific crafting item you need anywhere. Think star trek's replicators but slower.

The other set of production items is the regular assembler and refinery, which works the same as they do in SE1, with enhancement modules included. This does mean we skip the basic assembler and refinery, but I think thats fine with the micro-production blocks and the next set of production blocks.

The final stage of the tech tree is modular production blocks. At this point weve learnt about all of the production mechanics in SE1, so we move onto new mechanics for SE2. The entire idea of these production blocks is that theyre smaller than the regular production blocks but you cant build just one, you need to build an entire factory! Tools dont really get produced here, theyll generally be handled by the micro assembler, but everything else here is multistep processes.

For the refinery, I was thinking we could break it into 4 steps.

- The first is crushing, which is relatively low power and processes ore into ore power, which has a lower volume while remaining the same weight. You would have a crusher block which can have speed and efficiency modules but not yield. This step would likely take quite long but not the longest of every step.

- You would then have a separator block which takes in ore dust and outputs unrefined metal dust and gangue (the gangue could have a special reprocessor block that turns it into rock) at this point actual rock would change from crushed rock to small amounts of unrefined metal dust. This block wouldn't take a lot of time or energy, and can take all 3 types of modules, with yield working by improving the ratio of metal dust to gangue. Most of the reduction of weight also takes place here. (If the game has molten lava using its water system, then this machine would also be able to separate that into different molten metals.)

-The third step would be smelting. This step takes the raw metal dust and turns it into liquid metal The smelter would be able to take all 3 modules, but yield modules would have a reduced impact. This step would also have the highest energy cost of every step.

-The final step would be the injection mold, which turns molten materials into ingots. This machine would only be able to take speed and power modules, and would be the slowest of all steps.

The benefit of breaking it up like that is that different bases will be built in different ways depending on what's the most useful for them, for example, a zero gravity miner might have the crusher on board to allow it to mine more before returning to base, or a mobile base might prioritise the separator since it would reduce the weight of the outputs both as a flat value, and by turning some portion of the processed into gangue which gets launched out of an ejector!

As for assembling, I think the most interesting thing to do here would be to take a page out of the book of factoriolikes. So you would only really need one type of machine here, which is smaller but way faster and way more energy efficient than a regular assembler, but at the cost that the machine has to be configured to produce a specific item, and that the imputs for the assembler are not always just raw ingots. For example iron plates and girders would be able to be made directly from iron ingots. but then interior plates, as well as large and small tubes need to be made from plates, and construction components need to be made from girders (and maybe gravel to represent concrete?). Every recipe would be about 25% more resource-efficient than regular assemblers, take about half the electricity and use about a third of the time. The idea is to make it so that if you just need to assemble a small number of components for repairing stuff or building small ships youre better off using an assembler, but if you want to reach a truly industrial scale you need to either build a home station somewhere, or your mobile base would need to be truly massive.

As a side note, if this does get done, then I think that conveyors should be monodirectional.