Ability to craft lower quality with higher quality

[terradus]

Dear Factorio Developers,

I would like to share a suggestion for improving the item rarity/quality system in the Space Age expansion.

As a player, I appreciate the depth and complexity of Factorio, but the current rarity/quality system sometimes feels overly restrictive. The need to switch between recipes depending on the quality of the ingredients creates inconvenience and can be frustrating.

Suggestion:

Allow crafting using the same recipe, regardless of the rarity/quality of the ingredients used. Instead of dividing recipes into many separate entries, group them into a single, hidden recipe that accounts for the quality of the ingredients.

Calculation Mechanism:

I suggest pre-calculating the probabilities of obtaining items of different qualities at the output, based on the quality of the input ingredients. This would avoid overloading the real-time computation system and maintain performance.

Liquids: Introduce a quality system for liquids and use the existing logic for mixing liquids of different temperatures as a basis for mixing liquids of different qualities.
Ore Smelting:
When smelting a single ore with quality modules, the behavior remains unchanged.
When smelting multiple ores of different qualities, the quality of the output product is determined by a weighted average. For example:
1 legendary iron + 1 common calcite: probability of each quality (common, uncommon, rare, epic, legendary) = 20%.
10 legendary iron + 1 common calcite: probability of common quality = 9%, uncommon = 9%, rare = 9%, epic = 9%, legendary = 64%. The formula can be adjusted to achieve optimal balance.

Benefits:

Player Convenience: Simplifying crafting, eliminating the need to switch between recipes.
Flexibility: Ability to use any available ingredients.
Attractiveness: Reduction of the "annoyance" factor, attracting new players.

Assessment of Resource Impact:

I understand that you are concerned about the potential load on the system associated with calculating probabilities. However, I am confident that pre-calculating all possible combinations and storing them in the game files is a viable solution.

Recipe File Size: Even considering the large number of recipes and five quality levels, I estimate that the file size will not exceed 400 MB. This is an acceptable price to pay for increased convenience and flexibility. Roughly speaking, each recipe is a data structure containing probabilities; the total number of such structures equals the number of current recipes in the game. Each data structure, each recipe, will include 5^N elements, where N is the maximum number of materials in the recipe.

Calculation of the Number of Crafting Variations:

For example, consider a recipe with 6 components. Each component can have 5 quality levels. Thus, the number of possible combinations (and, accordingly, internal recipes that need to be hidden behind one external one) will be:

5 (quality of the first component) * 5 (quality of the second component) * 5 (quality of the third component) * 5 (quality of the fourth component) * 5 (quality of the fifth component) * 5 (quality of the sixth component) = 5^6 = 15625 variations.

This number may seem large, but, as I said, these calculations can be done in advance and saved in the game file. Most importantly, the variety of recipes will remain hidden from the user, and in the end, they will only interact with 1 visible recipe.

Thank you for your time and attention.

P.S. The calculation mechanism that I proposed is just an example solution. There are various ways to optimize and improve this system. For instance, a more efficient mathematical formula could be developed to calculate probabilities in real-time, avoiding the need to pre-calculate all combinations.

Alternatively, the implementation of a quality system with the ability to mix ingredients of different qualities could be optional. Players who enjoy the current system could disable this feature in the game settings. This approach would satisfy the needs of different player groups.

[Koub]

[Koub] Merged into an older thread with a similar suggestion : variations on adding a way to craft a given quality recipe with different quality inputs.