When I first heard about trinkets I was intrigued: they sounded like a fun way to inject some extra variation and challenge into a run and make it feel different from other runs with the same class. Now having played with them a bit they feel a lot more situational than I thought.

In many cases they seem like I’m just spending resources to make the game more challenging and the rewards from it aren’t commensurate. Since my mindset shifts into “survival mode” after I leave the character select screen and start the game, I generally avoid even creating most of the trinkets.

However I have seen a few cases now where beginners go into trinkets with gusto and it ends up costing them the run. This is leading me to suspect that trinkets may have a “beginner trap” effect where the lure of additional rewards is not being properly offset by an informed assessment of the risks. Of course, my view of this is only anecdotal!

So I have a question for everyone: how do you see trinkets fitting with your experience in the game?

I think one danger for any roguelike — when developed over a long period of time with a stable long term community — is for development to lean too far in a direction that favours providing new challenges to experienced players. Perhaps the most infamous example of that is NetHack, a game with a sheer cliff of a learning curve. I don’t think SPD is in much danger of that any time soon. Having said that, I do still worry about beginners because of their role in growing and maintaining the health of the community for the game.

Thoughts, anyone? Evan: can you share any insights from your analytics? I am particularly concerned about mimic tooth, wondrous resin, and chaotic censer. Do beginners use these trinkets differently from experienced players? Do they impact beginners’ success rate differently from experienced players?

Currently Unstable Spellbook draws random scrolls from a list of 10 eligible scrolls with replacement. My suggestion is to change this so that scrolls are drawn without replacement.

This idea came to me after someone on Reddit claimed to have drawn a bunch of strings (a string of 4 and a string of 6) of the same scroll in a row, all within the same game. Generally when this happens it gets people out of the game and has them thinking there’s something wrong with how scrolls are chosen.

My suggestion, to draw the scrolls without replacement, would make longer strings of duplicates like this impossible. It would also make the Unstable Spellbook more strategic in its use because you could keep track of which scrolls you get and then be able to make plans for potential upcoming scrolls. To make this less tedious, you might consider allowing the player to see some of the potential upcoming scrolls, similar to how some versions of Tetris show you the upcoming pieces (though not necessarily in exact order like Tetris).

Some further notes and thoughts:

• Identify, remove curse, and magic mapping are all half as common as the other scrolls. This could be handled by having a deck of 17 scrolls, with 7 duplicates for the more common types but only 1 copy of each of the 3 above.
• If you do go with a deck type system, maybe the player could keep adding more scrolls (beyond the needed for each upgrade) to bias the deck in their favour. This would make the Unstable Spellbook into a kind of deck-builder minigame, like Slay the Spire!
• Another idea might be to remove the popup choice for upgrading scrolls you draw, in favour of allowing the player to add both regular and exotic scrolls separately, giving them separate distributions within the deck. This loss of control would represent a small tactical nerf to the usage of the book which would partially offset the strategic buff caused by letting the player know and have more control over the distribution of scrolls they get from the artifact.

Anyway, thoughts, opinions, suggestions? I personally love the Unstable Spellbook in its current form but I have talked to others who don’t like it at all. My thoughts around this suggestion are to attempt to bridge this gap and make the item feel less random while still preserving its random flavour. The tradeoff is that this suggestion would make the item a bit more complex, though I don’t see think it’s an unreasonable amount of added complexity.

Alchemy is quite a complex system in the game and many players don’t engage with it at all. Even at the most tricked-out “deck builder” version of this suggestion, it’s still quite a lot less complex than alchemy because the choices are much more straightforward: want to see more of a scroll? Add another copy to the spellbook!

I love the variety and strategy trinkets are bringing to the game in 2.4! They do add to early game inventory pressure, which for me is the most frustrating part of the game (juggling a full inventory, throwing stuff down pits, running back and forth).

If trinkets were stored in the velvet pouch instead of the main inventory it would at least keep inventory pressure the same as it is now, without adding to it.