RNG Fodder

[Ascaron]

On the Adom Wiki, there's been discussion of late about whether or not the practice of using "RNG Fodder" actually works. This thread is going to be a discussion about this practice which will be referenced throughout the wiki.

So, what are your experiences with using items as RNG Fodder? Does it work? Has anyone done any proper testing in the past on this? Is it really worth picking up that brass amulet just in case it'll protect your Climbing Set in the rift?

[Soirana]

the only thing i know was part of ToEF testing. What does not work in tower, can't say for other locations.

[JellySlayer]

Preliminary testing indicates RNG fodder may work in some circumstances:

Test 1: ADOMBotted myself a character with 10 SoCR, 10 PoGA, 10 AoLS and nothing else. Repeatedly wished for acid vortices (25 tests in total). SoCR were destroyed in about 1/3 of explosions. No PoGA or AoLS were ever destroyed, curiously.

Test 2: Ran same character to Animated Forest, wished for a beeswax and a banshee. Collected about 13000s of useless junk. After 25 tests, SoCR were only damaged twice. PoGA and AoLS were never damaged. Importantly, acid vortices only seem to damage 3-7 different stacks of items if they actually damage items--having more stacks of items seems to reduce damage to useful items.

[Al-Khwarizmi]

No PoGA or AoLS were ever destroyed, curiously.

I think acid never destroys potions. Perhaps it doesn't destroy amulets either, I don't know.

From experience I can say that each element or effect in the game destroys certain kinds of items, for example fire bolt destroys many more items than frost bolt (and also more kinds of items). It would be good to have an exact list of the items that each bolt destroys. It's mostly common sense (for example cold cannot destroy a metal weapon but heat can melt it) but anyway there are always non-obvious cases.

From experience I'm also sure that RNG fodder works. I haven't made any studies, but I know that I'm a packrat and I practically never lose anything valuable in my pack. I usually don't bother to leave stuff before going to the Rift except for extremely important items like SoCR's, and I almost never lose anything valuable. But well, it's good to see well-founded studies about it.

[Soirana]

From experience I'm also sure that RNG fodder works. I haven't made any studies, but I...

I thought we had gentlement agreement that all rumors go were they belong to. Aka to theme "Rumorbook work".

[furious_styles]

Okay. I got together a test char, L50, with a few stacks of standard 'good loot':

Potions ('!')
   A - heap of 10 uncursed potions of gain attributes                    [40s]
   B - heap of 10 uncursed potions of ultra healing                      [20s]
Scrolls ('?')
   C - heap of 10 uncursed scrolls of chaos resistance                   [20s]

My intention was to throw him down the Rift a bunch of times, make a note as to exactly how many of each item were destroyed over the course of the whole affair, then add in a bunch of other crappy items and repeat, again noting the rate of destruction of the three key item stacks. However, the research kind of veered off in another direction when I noticed an interesting behaviour.

So I hurl this unfortunate soul down the Rift twenty times and recorded every destroyed item each trial:

1.  5 PoGA
2.  5 PoGA
3.  2 PoGA
4.  Nothing
5.  8 PoGA
6.  2 PoGA
7.  Nothing
8.  6 PoGA
9.  8 PoGA
10. Nothing
11. 9 PoGA
12. 10 PoGA
13. 6 PoGA
14. 3 PoGA
15. Nothing
16. 5 PoGA
17. Nothing
18. 8 PoGA
19. 2 PoGA
20. 7 PoGA

Okay, that's weird. I wonder what happens if we discard the PoGA? Let's try another twenty with the SoCR and PoUH only.

1.  Nothing
2.  4 PoUH
3.  2 PoUH
4.  Nothing
5.  2 PoUH
6.  Nothing
7.  4 PoUH
8.  Nothing
9.  8 PoUH
10. 2 PoUH
11. 7 PoUH
12. 4 PoUH
13. 10 PoUH
14. 7 PoUH
15. 5 PoUH
16. Nothing
17. Nothing
18. 6 PoUH
19. 3 PoUH
20. Nothing

The PoGA in the first trial seemed to be effectively 'protecting' the stack of PoUH, i.e. if the roll was failed such that damage was applied to an item stack, it would always be the PoGA that took the hit.

Well, this is pretty intriguing, I think to myself. It may not be that PoGA always protect PoUH or similar, but that depending on the order in which stacks are arranged in the character's inventory (which is dependent on the order in which they are discovered IIRC?) that certain stacks may always 'cover' other stacks in this fashion. This is speculation founded on the observation that the RNG, when it rolls to destroy items, is always destroying from the 'top stack' in the PC's inventory.

Let's try with just the SoCR. Are the PoGA and PoUH protecting them in turn?

1.  Nothing
2.  Nothing
3.  Nothing
4.  Nothing
5.  Nothing
6.  Nothing
7.  Nothing
8.  Nothing
9.  Nothing
10. Nothing
11. Nothing
12. Nothing
13. Nothing
14. Nothing
15. Nothing
16. Nothing
17. Nothing
18. Nothing
19. Nothing
20. Nothing

SoCR are rift-immune? I never knew that. A little bit more testing with a SoID stack appears to confirm that all scrolls are immune to Rift destruction, which stands to reason really as a scroll would be rather difficult to smash. The 'cover stack' effect is far and away the most curious thing here though.


Let's try some more testing, this time only with items we know to be destructible.

Potions ('!')
   A - heap of 10 uncursed potions of ultra healing                      [20s]
   B - heap of 10 uncursed potions of water                              [40s]
   C - heap of 10 uncursed potions of poison                             [40s]
   D - heap of 10 uncursed potions of strength                           [40s]

1.  1 PoPsn
2.  6 PoUH
3.  10 PoUH, 5 PoPsn
4.  2 PoUH, 5 PoPsn
5.  Nothing
6.  3 PoPsn
7.  10 PoUH, 2 PoPsn
8.  6 PoPsn
9.  4 PoUH, 5 PoPsn
10. 8 PoPsn
11. 5 PoUH, 3 PoPsn
12. 10 PoUH, 4 PoPsn
13. 7 PoPsn
14. 5 PoUH
15. 2 PoUH
16. 2 PoUH, 4 PoPsn
17. 8 PoUH, 1 PoPsn
18. 7 PoUH, 3 PoPsn
19. 8 PoUH
20. 1 PoUH

Well, that pretty much proves that some stacks 'cover' others. The stacks of PoWater and PoStr are completely invulnerable in this inventory arrangement. Now then, let's take away the potions of ultra healing, and see what happens - will the vulnerable stack now be poison only, or will the arrangement change?

1.  1 PoStr
2.  6 PoWater
3.  Nothing
4.  10 PoWater, 7 PoStr
5.  Nothing
6.  9 PoWater, 9 PoStr
7.  10 PoStr
8.  4 PoWater, 8 PoStr
9.  9 PoStr
10. 7 PoWater
11. Nothing
12. 6 PoWater
13. 7 PoWater, 4 PoStr
14. Nothing
15. 9 PoWater, 10 PoStr
16. 8 PoWater, 3 PoStr
17. Nothing
18. 2 PoWater
19. 4 PoWater
20. 9 PoStr

This is pretty fascinating. Now the poison is invulnerable. It appears to be that the RNG only makes a destruction roll on every second stack (starting from stack #1) in the PC's inventory from the top down.

One last trial to confirm.

Potions ('!')
   A - heap of 10 uncursed potions of water                              [40s]
   B - heap of 10 uncursed potions of poison                             [40s]
   C - heap of 10 uncursed potions of dexterity                          [40s]
   D - heap of 10 uncursed potions of confusion                          [20s]
   E - heap of 10 uncursed potions of strength                           [40s]
   F - heap of 10 uncursed potions of gain attributes                    [40s]
   G - heap of 10 uncursed potions of boost mana                         [20s]
   H - heap of 10 uncursed potions of troll blood                        [20s]
   I - heap of 10 uncursed potions of cure corruption                    [20s]
   J - heap of 10 uncursed potions of boost speed                        [20s]

1.  8 PoWater, 5 PoDx, 7 PoSt, 9 PoBoostMa, 8 PoCC
2.  2 PoWater, 10 PoDx, 2 PoCC
3.  9 PoWater, 2 PoCC
4.  4 PoSt, 2 PoBoostMa, 5 PoCC
5.  6 PoWater, 4 PoDx, 10 PoSt, 5 PoBoostMa, 6 PoCC
6.  9 PoDx, 7 PoBoostMa, 2 PoCC
7.  7 PoWater, 9 PoDx, 1 PoSt
8.  7 PoDx, 8 PoSt, 6 PoBoostMa, 2 PoCC
9.  9 PoWater, 4 PoDx, 8 PoBoostMa, 6 PoCC
10. 7 PoWater, 3 PoDx, 9 PoBoostMa, 2 PoCC

Well, there you have it. Alternating stacks from the top down. This is pretty awesome actually. Obviously this would need some testing in a practical setting, but the implication of this is that with intelligent inventory management, one could pick and choose exactly which 'disposable' items one wished to sacrifice to the RNG when climbing the Rift.

[furious_styles]

Something else I'll just add in a separate post, as the one above is convoluted enough already:

Assuming this holds true for all destructible item types, we can conclude that 'RNG fodder' is of no practical benefit for the case of the Rift (**this does not appear to be true, see edit), as a destruction roll will be made on 50% of all item stacks, no matter the total number of stacks. So, that stash of PoCC you may want to protect has a 50% chance of being rolled on, regardless of how many cursed fragile arrows and gnomish candies you take down the Rift with you.

This requires more extensive testing with other item types, which I hope to get to as time allows.



** Edit: the RNG fodder assumption I made above appears to be false. Having just thrown a backup save of a character-in-progress down the Rift and noted the destruction, two things were apparent:

1. for potions at least, the vulnerable stacks are indeed still every second stack starting from the top stack (not a single potion from an 'even-numbered' stack was destroyed), and

2. not nearly as many stacks had destroyed potions as you would expect. Glancing at the controlled test that used ten potion stacks, about 66% of the 'vulnerable' stacks experienced item destruction. Having thrown a char down the Rift who was carrying close to 100 potion stacks, he only experienced destruction to six of these stacks. It seems that, every-other-stack theory aside, sheer number of item stacks can indeed spread the destruction out amongst them.

[JellySlayer]

Furious, have you checked with stacks of different item types? For example, does having a bunch of arrows or gems in your inventory have any effect on your potion stacks?

[furious_styles]

Furious, have you checked with stacks of different item types? For example, does having a bunch of arrows or gems in your inventory have any effect on your potion stacks?

Doesn't seem to. The first time I threw him down he had the kind of inventory you'd expect from a lategame packrat with SoA, and he experienced destruction to 6 out of 91 stacks. I then ditched everything from his inventory except potions (and artifacts) and tried two more times, and got destruction to 6 out of 91, then 9 out of 91.

[JellySlayer]

Hmm... maybe I'll redo my acid vortex with this in mind and see if I come up with something similar. Your tests seem to suggest that RNG fodder works, to some degree or other, but not at all in the way that it was expected to.