DP Research Thread #6 (Even newer)

[poccil]

Accuracy begins as a number from 0 through 100. For example, normally, the accuracy of Thunder is 70. However, if it's sunny the accuracy is 50 instead, and if it's raining the attack "cannot be evaded" meaning it ignores accuracy and evasiveness modifications. Effects that multiply accuracy by "X/Y" in reality multiply the result by X and divide the result by Y. Accuracy and evasiveness stat stages are just one part of the accuracy calculation process.

The value for accuracy is always an integer. Whenever division occurs, the result is rounded down. This occurs, for example, when multiplying by "X/Y" as described above.

[callforjudgement]

There are some rumours on another forum (that I don't trust as a reliable source) that battle mechanics are different for AI-controlled Pokémon than they are for human-controlled Pokémon; the example given was that the secondary effect of Metal Claw was more likely to trigger when used by a computer player than by the human. Nobody on SQSA seemed to know whether this was true or not (at least, I got no answers), so it may well be worth testing here. (This is competitively relevant at the Battle Frontier, at least, especially if it affects Quick Claw hax; some of the streak-endings seen in the Battle Frontier thread are ludicrously unlikely, so it's possible that we just have the probabilities wrong.)

[0n1l1nk.]

Its well known than in the battle frontier the AI cheats, all the more in EMERALD.

The AI must have some kind of system predicting what the outcome will be, and so decide using that attack or not.

[callforjudgement]

0n1l1nk: so you think that what's happening is that the AI checks to see what effect each of the four moves will have and uses the best one, and does some unintentional inbattle RNG abuse as a side effect? If so, I wonder how you could prove it?

Also, it's the "well known" that I'm suspicious about; nobody seems to have an actual source for what specifically is happening, which is why I'm asking in this thread. (Some "well known" Pokémon facts have turned out to be false in the past, after all.)

[Pink]

It seems very unlikely that this is the case. This would mean either the AI is changing the seeds to make the RNG work in their favor (or completely ignoring it) or the game waits to give CHs or Quick Claws to the AI, which would mean the game scripts for when the player is going to die (basically). Very unlikely indeed.

[david stone]

Or it means that a move used by the AI that should have 85% accuracy has 90%, for instance. Not likely, but in theory possible.

[Zystral]

Originally Posted by Fat obi

If a Pokemon is immune to U-turn and the opponent uses U-turn, does the user still get to switch out after U-turn fails to do anything?

No damage is done so the user does not switch out. It simply counts as a 'Doesn't affect <Pokemon>' and the move fails completely.

Originally Posted by Fat obi

If a frozen Pokemon uses Flame Wheel / Sacred Fire, but is also confused, can it still hurt itself in its confusion? If it does, does it remain frozen or does it thaw? Same for a frozen Pokemon with Truant, which might be easier to test but could theoretically give different results.

Using SoulSilver vs. Diamond, I managed to freeze a Flame Wheel Slaking, and I can say that if a Truant mon is Frozen, it places Freezing at a higher priority. Rather than say "SLAKING is goofing off!", it says that it is frozen solid. I think it's because you don't choose a move on the second turn, you simply click 'Fight' and Truant activates automatically.

As for Frozen and Confusion...
It places Confusion on a higher level. Frozen Shuckle has Confuse Ray used on it. As I tried to use Flame Wheel, it said "Shuckle is confused" first, then Shuckle hit itself in confusion, and is still Frozen. The result is the opposite of TruantFreeze. Hope that answers your question, Obi.



I also have some news on Conversion2, in regards to the OP. Tested on SS.
- It changes the user's type completely, even if dual-typed, so something Dark/Ghost would become simply mono-Water if hit by a Fire move.
- It fails if there is no other type to change to (e.g. Steel type hit by dragon move).
- It will take all available types that resist or are immune to. (e.g. opponent used Thunderbolt. Conversion2 can change you to either Dragon, Electric Grass or Ground.)
- Each type available for selection has an even chance of being selected, even though at random. (e.g. if hit with a Dragon-type move, 100% chance of being Steel-typed, if hit with a Steel-type move, 25% chance of being Electric, Fire Steel or Water.) Also, if user is one of the available types that type is not counted on the selection (e.g. Grass-type is hit by Water-type move and uses Conversion2 has 50% chance of becoming either Dragon or Water.)
Not sure if there's anything else about Conversion2.
In a double battle, if two foes hit you with different attacks, then it obviously works off the most recent one, but I think that's obvious. It works through a Sub, but Smogon's Dex has that already.

In terms of Protect/Detect:
if the user is faster than the opponent and the user has protect.
1st turn: ninjask - Protect / chimchar - Ember (fails)
2nd turn: ninjask - Conversion2 - FAILS, no prior attack to be resistant to.


AND

1st turn: ninjask - Conversion2 - FAILS / Mon2 - Ember
2nd turn: ninjask - Protect / chimchar - Ember (fails)
3rd turn: ninjask - Conversion2 - FAILS. (I assume no move was used

AND
1st turn: ninjask - Conversion2 - FAILS / chimchar - Earthquake (doesn't affect ninjask)
2nd turn: ninjask - Conversion2 - FAILS

These experiments prove this: Conversion2 works off the the most recent attack the foe used, not the most recent attack that connected.
So if the move the foe used right before Conversion2 activates did not do damage, then the Conversion2 fails.

[orius747]

Something that may be useful to note, though I'm not sure this is the place for it, is that the IV checker fellow in Heart Gold appears to cycle through the pokemon's max IVs. For example, I bred a Gible, and in an effort to determine whether it inherited speed from its parents, I was spamming A on the IV man. After six or seven clicks I noticed he listed Attack as his highest, then defense, and then special defense. I cycled through my Gible's IVs for ten more cycles (thirty clicks in all) and he did this consistently. Afterwards, I checked the Gible's IVs and those three were the only three max IVs. Sadly, the Gible's speed sucked, but I think this can be quite useful for checking what IVs were inherited.

Also, like the Everstone in DPPt, the power items will not guarantee an IV will be inherited if the parents are from different countries.

Hopefully I'm in the right place. If not, please let me know if there's a HG/SS research thread.

[Airfoil]

I can't try this, but has anyone gotten a Porygon w/ Mimic, Arail Ace, and Conversion from Firered/Leafgreen and attempted to use Roost while Porygon is a pure flying type in a double battle? Or will it just fail?

Turn 1:
Porygon uses Conversion, Porygon became a flying type (because of Arial Ace)!
Porygon's partner used Roost!

Turn 2:
Porygon uses Mimic! Porygon learned Roost!
Porygon's partner does something (it would need to use roost again if it outruns Porygon)

Turn 3:
Flying type Porygon used Roost!

[RBG*]

It becomes the ??? type which has no weaknesses or resistances.

[Xel]

If your Pokémon uses a pinch berry (Leppa Berry for example), then Tricks it away, will it be Recycled if the foe's Pokemon tries to ? If not, will your Pokemon still be able to Recycle the Berry ?
And more, if you trick a berry that is then consumed by the foe's Pokemon, then re-Trick it, will Recycle work ? From the description on Smogon's page I understand that it should work, but it's still unclear.

[Zystral]

If you have any berry and it is consumed, when you Trick, you will obtain the foe's item and the foe will get nothing. If you try to Recycle after that it fails since you are already holding an item.
If they have no item when you use Trick and then you use Recycle, then you get the Berry back.
If you use Trick AGAIN so they get a berry and you then have nothing and use Recycle again, you will have two of the same berry.

I sort of don't understand what you're saying, you're implying that we're Tricking the Berry AFTER the berry is eaten, which makes no sense :S

[Scepticallistic]

Originally Posted by Fat orius747

Something that may be useful to note, though I'm not sure this is the place for it, is that the IV checker fellow in Heart Gold appears to cycle through the pokemon's max IVs. For example, I bred a Gible, and in an effort to determine whether it inherited speed from its parents, I was spamming A on the IV man. After six or seven clicks I noticed he listed Attack as his highest, then defense, and then special defense. I cycled through my Gible's IVs for ten more cycles (thirty clicks in all) and he did this consistently. Afterwards, I checked the Gible's IVs and those three were the only three max IVs. Sadly, the Gible's speed sucked, but I think this can be quite useful for checking what IVs were inherited.

Also, like the Everstone in DPPt, the power items will not guarantee an IV will be inherited if the parents are from different countries.

Hopefully I'm in the right place. If not, please let me know if there's a HG/SS research thread.

No, I've had him tell me two times in a row the max IV was in Special Attack and then tell me Speed. So I guess that disproves the cycling theory (your case may have been a fun coincidence).

[eric the espeon]

Two things I need confirming/testing before editing the on site pages:

Does Magic Guard block the damage from Ghost Curse?

How does Trick interact with Unburden exactly, people on #C&C could tell me the answers for basic situations (Trick when both have items gives no boost, Trick when foe has no item and you do gives a boost), but what happens when:
Unburden user has an item, foe does not, both use trick on the same turn.
Unburden user has an item, foe does not, Trick is used by one Pokemon two turns consecutively so the items ends up where it started (check the speed boost, does it have +2 or something else?).
Unburden user has an item, foe does not, Trick is used by one Pokemon three turns consecutively so the items ends up being passed away from the Unburden user twice (check the speed boost, does it have +2 or something else?).

[poccil]

I haven't tested these interactions but this is what I currently have on my descriptions:

Magic Guard: "Effects, other than damage, can't reduce bearer's HP, with these exceptions: ... Curse when used ... "

Unburden: "Bearer's Speed is doubled ... if an item that the bearer held was consumed or lost ... while bearer had this ability and since bearer became active." My belief is that Trick has no interaction with Unburden, since it doesn't cause an item to be "consumed or lost." Unburden's effect doesn't stack if an item that the bearer holds is "consumed or lost" multiple times.

I will test both as soon as possible.

EDIT:

[The Curse/Magic Guard behavior was mistaken. --Poccil, Edited Sep. 25, 2012]

Also, apparently, using Trick when the bearer of Unburden has an item causes Unburden's effect to "activate", whether or not Trick's user has an item or not. Moreover, through further testing, Unburden's effect works even if the bearer's item is lost through its use of Fling and Natural Gift. Thus, I've updated the ability description to read as follows:

Bearer's Speed is doubled if an item that the bearer held was consumed or removed from the bearer while bearer had this ability and since bearer became active.

[Zystral]

My experiments show that a ghost using Curse on a Magic Guard user, the Magic Guard holder will not take Ghost Curse damage.

Also this is my result of the Unburden/Trick result:
experiment for same-turn tricking
the unburden holder tricks their item away and gets a speed boost. the speed boost is retained when the foe tricks the item back.

experiment for 2 consecutive turn tricking
the same thing happens in the above scenario. they lose their item and get a speed boost. when the trick the item back the next turn, speed boost is retained.

experiment for 3 consecutive turn tricking
1st trick - speed boost obtained through lost item
2nd trick - speed boost retained when item recovered
3rd trick - speed boost obtained through lost item, total boosts: +2

So I think poccil is right, it should be changed to if the item is consumed or removed from the bearer.
Also mention how these boosts stay in place even if another item is obtained.

[Wichu]

Originally Posted by Fat poccil

I would assume the trick works because the game recalculates the Pokemon's stats as it reduces its effort values. The same thing happens with leveling up and using "vitamins" (which I call stat enhancers.) Since a level 100 Pokemon can't level up and stat enhancers fail to work at a certain point, the only recourse is to use stat-reducing Berries to recalculate the stats.

This is quite old, but I disagree. Placing a Pokémon in a PC box also recalculates the Pokémon's stats; when stored in a PC box, the game discards the Pokémon's current stats to save memory, and instead displays the stats calculated from its IVs and EVs. It's been proven that this doesn't work since G/S/C (where EVs didn't exist); this strongly suggests that a level 100 Pokémon cannot gain EVs (if it could, the game would have to store a copy of its EVs in order to ensure its stats don't change when it is stored in the PC).
The most likely explanation is that EVs are given when a Pokémon gains experience; as level 100 Pokémon do not receive experience, they cannot receive EVs either.
The person who originally posted the 'trick' must have been mistaken or lying. I've tested this in Pearl, and it didn't work; I don't see why it would have been changed in Platinum (I don't have any level 100 Pokémon in Platinum at the moment, so I can't test it).

Also, although Jimbo may actually be fat, the 'Fat' shown when he is quoted isn't part of his username; it's a small joke in the forum settings. Check the quote in this post :P

Originally Posted by Fat ZystraL

experiment for 3 consecutive turn tricking
1st trick - speed boost obtained through lost item
2nd trick - speed boost retained when item recovered
3rd trick - speed boost obtained through lost item, total boosts: +2

So I think poccil is right, it should be changed to if the item is consumed or removed from the bearer.
Also mention how these boosts stay in place even if another item is obtained.

I don't think this is quite right. The Unburden boost is not a direct stat boost (like the Speed Boost ability); it's an extra boost on top of that (like Tailwind). As such, I'm pretty sure it can't be stacked; i.e. on the 3rd trick, although it may tell you it got another boost, it will still have the same Speed as with one boost. This is just inferred, so obviously, someone should make sure.

[Naxte]

Originally Posted by Fat Wichu

This is quite old, but I disagree. Placing a Pokémon in a PC box also recalculates the Pokémon's stats; when stored in a PC box, the game discards the Pokémon's current stats to save memory, and instead displays the stats calculated from its IVs and EVs. It's been proven that this doesn't work since G/S/C (where EVs didn't exist); this strongly suggests that a level 100 Pokémon cannot gain EVs (if it could, the game would have to store a copy of its EVs in order to ensure its stats don't change when it is stored in the PC).
The most likely explanation is that EVs are given when a Pokémon gains experience; as level 100 Pokémon do not receive experience, they cannot receive EVs either.
The person who originally posted the 'trick' must have been mistaken or lying. I've tested this in Pearl, and it didn't work; I don't see why it would have been changed in Platinum (I don't have any level 100 Pokémon in Platinum at the moment, so I can't test it).

Not sure about HG/SS (the games it was requested to be tested on), but it was tested in Platinum, and found to not work there either:
http://www.smogon.com/forums/showpos...postcount=3039

[Xel]

Originally Posted by Fat ZystraL

If you have any berry and it is consumed, when you Trick, you will obtain the foe's item and the foe will get nothing. If you try to Recycle after that it fails since you are already holding an item.
If they have no item when you use Trick and then you use Recycle, then you get the Berry back.
If you use Trick AGAIN so they get a berry and you then have nothing and use Recycle again, you will have two of the same berry.

I sort of don't understand what you're saying, you're implying that we're Tricking the Berry AFTER the berry is eaten, which makes no sense :S

Well, that's what I'm saying. The fact that you can Recycle a consumed berry means that the game keeps track of the fact you had an item at the beginning of the battle after you've eaten it, and what I'd like to know is if this "ghost item" is affected by Trick or not. And if it is tricked, will the opponent be able to Recycle it or is the original holder the only one who can do this ? So, to test it, someone should take a Pokemon knowing both Recycle and Trick and another one knowing only Recycle, go in a double battle, make this Pokemon eat his berry and then trick it to the other Poke knowing Recycle. If he can Recycle the berry, the "ghost item" is affected by Trick. If he can't Recycle the berry, try to Recycle with the original bearer. If it works for him, it means that the "ghost berry" is sticked to its consumer and is therefore not affected by Trick. If Recycle doesn't work for both Poke, it means that the "ghost item" is lost when you use Trick, but it may also mean that only the original holder/consumer can recycle the item, so I guess you could re-Trick and try to Recycle the berry with the original holder to be sure of that.

Another question would be if a consumed item can be Recycled after the foe's used Knock Off (when you had already consumed your item, of course), but I guess it's already been covered.

[DDRMaster]

The "ghost item" does seem plausible considering that after you use Knock Off on a Pokemon, you can't Trick them because the game remembers that they're supposed to have an item. It still could work differently since you don't get the item back at the end of the battle like you do with Knock Off, though. I'd test this myself, but I recently restarted my HG without deleting the old save file so I have to keep my DS on all the time, since I can't save.

[Zystral]

Okay so hang on.

Pokemon Z has an Item. It consumes the item.
Pokemon Y tricks Pokemon Z so Pokemon Z is holding an item. If Pokemon Y uses Recycly then Recycle fails since Pokemon Y never consumed its item.
If Pokemon Y tricked again so Pokemon Z was not holding an Item and Pokemon Z uses Recycle, it gets the consumed item.
Knock-Off + Recycle fails since Recycle only works for consumed items.

[Xel]

So that would mean that the ghost item is sticked to its consumer, isn't it ?
And about Knock Off, i meant you had to Knock Off a Pokemon who had already consumed its item and then try to Recycle, to see if the ghost item is Knocked Off or not.

[Zystral]

Yes, the ghost item is respective to its original holder. Also, Knock Off has no effect on a consumed item - Knock Off turns the held into into a Ghost Item, and so Recycle gives you back your consumed item.

[poccil]

My attack description for Recycle reads "If user isn't holding an item, user receives the last item that was held by a Pokemon at user's position this battle and was consumed this battle. That item is reset." Recycle can thus recover even items held by previous Pokemon at the user's position. Also, Knock Off, Trick, and Thief don't reset or change the item being restored, since in these cases the item is not "consumed." (My description for Knock Off reads "If this attack is successful, opponent drops its held item even if it has zero HP. (That item is not consumed.)")

[Zystral]

No it's not, but Knock Off still gives the illusion of a ghost item
It's technically no item but the effect is still the same since it's no item. So even if you used Knock Off when they had already consumed their item, Recycle still restores that item

Not sure what your problem is.