Feedback on item generation in Trunk

[DuxOrbis]

Recently a friend and I have been playing in Trunk, and while it could be tweaked in any number of ways to amplify aspects of the game like item generation for rapid testing, I wanted to clarify changes to the item generation behavior in the potential Stable release.

So far I've seen early quickblades, ornate decks, greater numbers of randarts, especially early randart amulets, and shops with more randarts, perhaps more shops in general, a manual of dodging in the sewer, a triple sword in Orc, crystal ball of energy on d9, most of this in the same game. Honestly I kind of like what I've seen, it puts an interesting mix of high level gear into lower level player's hands, without making them absurdly powerful, just more interesting and perhaps more flexible, or rather giving them more choices at a time when it matters more what they specialize in. I've played very few games in Trunk to know if this is a string of flukes or not, but the chaos dial of the dungeon seems be cranked up to 12, appropriately enough.

Thoughts? Feedback on item generation in Trunk?

[DuxOrbis]

Said friend reports +2 Ice Dragon armour on d2.

[DuxOrbis]

Feedback on item generation in Trunk?

it hasn't changed

I was afraid of that. That either means something odd happens to the RNG for a time after installing or something is broken. It happens that we find very different, generally very rare items much earlier now, on several machines. I'll post with more data when I get a chance, to rule out large number bias. If you notice anything strange I'd appreciate a heads up.

[Grimm]

That either means something odd happens to the RNG for a time after installing or something is broken.

Or clustering illusion.

[XuaXua]

Clustering illusion is a transparent excuse.

[DuxOrbis]

A rod of demonology on d3. 2 wands of healing d2. There's more but you get the point.

[galehar]

I don't think there's anything preventing the generation of good items on shallow floors. They are rare, but there isn't any limit to how good they can be.

2 wands of healing d2.

Wow. This one is really amazing given how rare healing wands are.

[njvack]

A bunch of scrolls have minimum depths -- acquirement, EW3, unholy creation, silence, and vulnerability won't spawn before D4, and vorpailse, torment, and holy word before D7.

I'm not quite sure how wand generation works -- from what I can tell, a wand normally has a 1.1% chance to be HW, but this is somewhat higher on early dungeon floors as it rerolls high-tier destructive wands instead of downgrading them.

At least, I think this is what's going on. I'm not totally clear on the difference between item_level and depth_mod.

[ebarrett]

That either means something odd happens to the RNG for a time after installing or something is broken.

Or clustering illusion.

A rod of demonology on d3. 2 wands of healing d2. There's more but you get the point.

No, you don't get the point.

[mumra]

Maybe we should be compiling daily stats of different types of item, monsters, and also uniques generated in D:1-10 in online games. Then present these stats as a big line graph showing that given a large enough sample domain the weightings have remained approximately the same since the stats began. Then whenever a thread like this pops up we can just link to the chart and lock the thread

[ebarrett]

That kind of people will often ignore the facts and claim that the game is somehow broken anyway, because the conclusion is made beforehand, and because their "sample" "proves" their "point".

[KoboldLord]

There's roughly a one-in-one-thousand chance that your first two wands of the game will both be heal wounds, but there's also more than a thousand games being played every month. Somebody's going to get those two wands sooner or later. Congratulations on being the player in game #1000.

[Grimm]

That kind of people will often ignore the facts and claim that the game is somehow broken anyway, because the conclusion is made beforehand, and because their "sample" "proves" their "point".

There's no need to be a dickhead about it ebarrett. We all make errors. That's how we learn.

DuxOrbis: as you can see the general opinion is that there is most likely not a bug. In fact I think I'll lock the thread now; please pm me when you have some data to bring to the table.