I've always thought about this. I think the only game in the series where I didn't feel like I was being overloaded with useless trash was Ayesha. The Arland games and Ryza 1 (haven't played 2 yet) just feel like the give you far too much trash to sort through when doing a synthesis. I've been playing around with designing a synthesis system like you'd find in Atelier, and really wondering how to best balance the need for a lot of materials vs making sure those materials always feel meaningful. It's hard without accidentally forcing the player to grind.
Oh, I have some thoughts about this!
One thing that separates Arland from more recent games (that I think was brought back with either Ryza or Mysterious but I'd have to double-check) is that in Arland, the same ingredient can have both different qualities AND different traits. So for example, if we're talking about blue puniballs, I can get blue puniballs that are maybe quality 20-30 at the beginning of the game and have really weak traits like, I dunno, Puni Repel, but in later dungeons find the same blue puniballs with higher quality and better traits like Attach (Elemental).
Dusk doesn't really have this in the same way. I think item quality can still differ, but the traits are fixed; if you're trying to pass on traits in recipes and don't care about quality (or have other ways to raise it instead of relying on ingredient quality), then literally any blue puniball will do because they all pass on the same traits.
On the one hand, it makes things quite a bit simpler; you don't have to cross your fingers and hope you'll get an ingredient with a specific trait for a cool weapon or bomb you want to build, or if you DO have an ingredient with a specific trait, you don't have to be precious about it and hoard it until you can build the right thing. It also means you're less incentivized to keep around 400 puniballs because maybe one or two of them have a trait you might need in the future. But on the other hand, making every ingredient have the same traits takes a bit of the fun out of collecting; ingredient quality doesn't matter that much except in certain scenarios, so you don't get that fun prize of (say) seeing that blue puniball with a ridiculous attack trait or whatever.
I don't think one system is better than the other, it just leads to different things to prioritize in synthesis. For example, I find that in all the games where traits aren't consistent from ingredient to ingredient, I try to "save" specific traits I like by synthesizing low-level recipes using that ingredient, that can then be used in other recipes (or recycled into even more of the same low-level recipe, ex. synthesizer --> zettel --> synthesizer that uses paper ingredients --> zettel --> etc.) Whereas in the Dusk games I don't think I did as much of that, and instead focused on how to minmax the various synthesis boost options you had in Ayesha and later to, say, carry over more traits or make really expensive ingredients cheaper.
One of the things that can help with this is whether the game also has the concept of combining traits. Ryza has a simplified version where some traits can have levels, so ex. an ingredient with Defense Level 2 can combine with an ingredient with Defense Level 5 to create Defense Level 7. But of course this just leads to you hoarding all the ingredients with Defense so you can combine them later. Arland and I think Dusk had a more advanced version of this where you could combine ATK/DEF or ATK/SPD or DEF/SPD traits into a combo trait, and then combine two combo traits for All Stats Increase/Boost/etc, or other similar combinations that weren't always obvious and had to be found through experimentation (or careful wiki/compendium reading). Again, encourages you to keep ingredients with lesser traits because you can sometimes combine them.
store.steampowered.com
