The current state of Engineering in Elite Dangerous offers players a huge boost to ship potential for those willing to invest the time. However, randomness in material costs, availability and primarily the Material trader's maths create a brittle system which encourages players to farm high grade materials using, mostly, immersion breaking techniques.
What may seem like a far reaching problem can be largely mitigated with simple tweaks to the material trader and engineer blueprints. These changes have the net effect of allowing changes in the BGS to be more easily compensated for within the trading system.
The State of things
There currently exists 108 material types split into 3 categories: Raw (gained from surface prospecting and mining), Manufactured (missions and ship wrecks) and Encoded (missions rewards and Scanning ships) and 23 sub categories. 80 have 5 grades and 28 have 4 grades. Each pickup of a material provides 3 units. They appear in varying quantities throughout many aspects of the BGS and gameplay, some in very specific niches (pharmaceutical isolators), others more broadly (Focus crystals). The spawn rates are largely unknown though the spawn circumstances are fairly well documented.
Material Traders exchange materials within each of the three categories. Trades can be done horizontally within a sub category or vertically between sub categories. The trading ratios between low grade to high grade, high grade to low grade and between sub categories are fixed at 1:6, 3:1 and 1:6 which
preferentially benefits gaining G5 materials and trading across and down.
The dominant farming strategies almost all have one thing in common:
they rely on farming G5 & G4 materials in large batches using relogging. This, combined with a lack of gameplay differentiation between regular material acquisition methods contributes to the feeling of 'grind'. Importantly,
these dominant strategies don't rely on farming what you need, but what is easiest to get.
And here is why:
Result
G5 material farming is nearly 1000x more efficient than farming G1. The staggering disparity means that any solution which does not address this would lead to counter-intuitively disparate numbers for things such as mission material rewards, surface prospecting and materials dropped after ship destruction. Even if we compressed the table using the smallest whole number ratio we still end up with a ratio of over 15x.
The second major flaw of the table is that ratios are linked. Changing the ratio between G4 and G3 also affects how all other grades are traded. This means that
the table cannot respond to individual materials being abnormally scarce/abundant without unbalancing the table as a whole.
This reality forces players seeking efficiency and those who make guides down a narrow set of repetitive, immersion breaking gameplay tasks which are simply not the most enjoyable way to play for most people.
Solution
This problem can be solved simply:
Remove horizontal trading and make each member of the row a discrete value used by the blueprints. Make blueprints require a total of these values, rather than a specific number of material within the row.
E.G. Blueprint Requires 27 units of Focus crystals. Crystal shards supply 2 units, Refined focus crystals supply 9. You can give her 14 units of Crystal shards, 3 units of Refined Crystals or any combination. The player can manually chose or let the system choose.
This would have the effect of allowing the tabled to be flattened far more than the current maths allow and each material value to be tweaked without severe knock on effects. It also means that commanders only need to focus on the 23 categories rather than the 108 types and encourages looking at the values more broadly.
You could preserve having experimental effects needing specific materials as farming a specific type in rare circumstances would be an exception to normal gameplay activities, rather than the rule.
RNG in the Engineering process
The current system inherits random roll effects from it's predecessor while removing the random multidimensional effect and replacing it with -fairly- linear progress towards a goal. While the rolling can be enjoyable on a level similar to playing a slot machine it means anyone setting about in a determined fashion to engineer their ship has to plan for this uncertainty which drives them to conservative and overcompensatory play i.e. better get much more than I need 'JUST IN CASE'. Making the farming problems worse.
As a given blueprint has a fairly straight line from G1 to G5 we can consider this a a progress bar from 0 to 100% engineered and it should probably simply be displayed as this with the player selecting their desired percentage and the materials being calculated for them. Further engineering on the module picks up from the previous percentage until the player has reached their desired amount or 100%
The language could be changed to support this, with the initial module being engineered and then the bar called something like turning or optimisation. So you have an FSD engineered with increased range with an optimization of between 0 and 100%