-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
-
Congrats to the winners of the 2023 Smog Awards!
Gen 2 PS Development - Post bugs here
- Thread starter Joim
- Start date
Lavos
Banned deucer.
Lavos
Banned deucer.
Marty would you mind doing these instead? Zarel seems to have a lot on his to-do list and some of these bugs are immediately relevant to SPL.
- Blissey move incompatibility bug with Present + Heal Bell
- Swagger still confusing at 999 Atk
- Belly Drum giving +6 instead of +5 when drumming at -1, -3, or -5
- Freeze Clause unimplemented
Thanks!
- Blissey move incompatibility bug with Present + Heal Bell
- Swagger still confusing at 999 Atk
- Belly Drum giving +6 instead of +5 when drumming at -1, -3, or -5
- Freeze Clause unimplemented
Thanks!
Just want to be a bit more specific so it's easier to implement:
Swagger is supposed to fail/miss if the opponent is at 999 attack or +6 attack boosts.
Stat boosting moves should not add boosts after the 999 cap is reached or when the boost is already maxed out. This means Snorlax with Curse cannot boost to +6 attack, only +5. It can however boost to +6 defense. Vaporeon can only boost to +5 with Growth, and additional Growths will not raise the stat similar to how it would at +6 for other mons. This is different from simply capping the stat at 999, which is implemented. Snorlax commonly is the target of Growl and Charm, so whether or not it's at +5 or +6 is relevant, since after a Charm it should be at +3 instead of +4. This is also very relevant for Curse on Curse Pokemon as whether or not critical hits ignore stat boosts is based entirely on who has the higher stat boost; this is implemented correctly on PS already, assuming the stat boosts are correct (which they are not).
This does not affect how many stat boosts you can Baton Pass to a Pokemon. If the Baton Passed stat boosts surpass the 999 cap, then the Pokemon will still have the same number of stat boosts as the passer, the stat will simply be capped before being used for anything.
When determining if the 999 cap is met, Thick Club, Screens, Burn and so on are not supposed to be taken into account. The only factors used in determining if the 999 cap has been met are the stat+stat boosts, nothing else. This means Marowak with Thick Club can Swords Dance twice, even if it's pointless.
Belly Drum is a bit weird. Drumming essentially applies a +2 boost until either the user reaches 999 attack or +6. This is why Snorlax reaches +6 on even numbers, but +5 on odd numbers of initial boosts. I'm fairly certain Pokemon with a higher base attack could end up having +4/+5 instead of +5/+6 if it were to use Belly Drum. In the game's code I believe it applies Swords Dance multiple times, so I think Swords Dancing might also only check for the cap before boosting, i.e. check if it can boost -> boost twice regardless of where the first +1 puts you.
edit: srs tho if you need more clarification on something hmu on discord
Swagger is supposed to fail/miss if the opponent is at 999 attack or +6 attack boosts.
Stat boosting moves should not add boosts after the 999 cap is reached or when the boost is already maxed out. This means Snorlax with Curse cannot boost to +6 attack, only +5. It can however boost to +6 defense. Vaporeon can only boost to +5 with Growth, and additional Growths will not raise the stat similar to how it would at +6 for other mons. This is different from simply capping the stat at 999, which is implemented. Snorlax commonly is the target of Growl and Charm, so whether or not it's at +5 or +6 is relevant, since after a Charm it should be at +3 instead of +4. This is also very relevant for Curse on Curse Pokemon as whether or not critical hits ignore stat boosts is based entirely on who has the higher stat boost; this is implemented correctly on PS already, assuming the stat boosts are correct (which they are not).
This does not affect how many stat boosts you can Baton Pass to a Pokemon. If the Baton Passed stat boosts surpass the 999 cap, then the Pokemon will still have the same number of stat boosts as the passer, the stat will simply be capped before being used for anything.
When determining if the 999 cap is met, Thick Club, Screens, Burn and so on are not supposed to be taken into account. The only factors used in determining if the 999 cap has been met are the stat+stat boosts, nothing else. This means Marowak with Thick Club can Swords Dance twice, even if it's pointless.
Belly Drum is a bit weird. Drumming essentially applies a +2 boost until either the user reaches 999 attack or +6. This is why Snorlax reaches +6 on even numbers, but +5 on odd numbers of initial boosts. I'm fairly certain Pokemon with a higher base attack could end up having +4/+5 instead of +5/+6 if it were to use Belly Drum. In the game's code I believe it applies Swords Dance multiple times, so I think Swords Dancing might also only check for the cap before boosting, i.e. check if it can boost -> boost twice regardless of where the first +1 puts you.
edit: srs tho if you need more clarification on something hmu on discord
Last edited:
- Blissey move incompatibility bug with Present + Heal Bell
-> not done, Zarel is the validator expert and I'll defer to his far better grasp of the subtleties involved to solve this one
- Swagger still confusing at 999 Atk-> done!
- Belly Drum giving +6 instead of +5 when drumming at -1, -3, or -5-> done half-assedly; Belly Drum's stat boosts will now work "correctly" but every other boosting move will need a complete overhaul of the stat functions and I'm not going to do that during SPL, not that I have time to test it properly anyway
- Freeze Clause unimplemented-> done!
That message appears after cancelling a switch. It does not have any actual effect/you can still cancel switches.
This is obviously just a minor bug and fixing the actual mechanical bugs should be more important than fixing this one.
Last edited:
Mr.E FriendOfMrGolem120 Fixed, will take effect in a few days.
I'm not sure if someone has posted this or if this applies here, but it seems that I can apply natures to Pokemon and the stats actually update. The teams are valid for the format. I'm not sure if the stats actually update in battle, but it does in teambuilder. I have a couple of photos. In the first one, you can see the natures and the valid text. In the second one, you can see Gengar's affected stats (Gengar only goes to 318 speed, but here it's 349).
This might be a teambuilder issue but a bug is a bug.
This also affects gen 1 for some reason.
This might be a teambuilder issue but a bug is a bug.
This also affects gen 1 for some reason.
From turns 20-21 of this game a Gengar uses Destiny Bond and falls asleep to a Jynx. It then takes a sleep turn and dies and Destiny Bond activates. It should not have as Destiny Bond wears off after any turn is spent move used or not..
https://replay.pokemonshowdown.com/gen2ou-713870525
https://replay.pokemonshowdown.com/gen2ou-713870525
Earlier I saw Thief fail despite not having an item. I thought it was bugged until the lovely Mr.378 reminded me that this is correct since Thief and a few other moves suffer from 255/256 "accuracy" on their side effects.
That being said, could we get a message in the simulator similar to when belly drum boosts +2 or when Giga Drain fails against subs? It would prevent a lot of confusion.
That being said, could we get a message in the simulator similar to when belly drum boosts +2 or when Giga Drain fails against subs? It would prevent a lot of confusion.
did damage, no item taken. It should definitely display the reason if the item was leftovers or something else visible after using thief. It's confusing as hell otherwise.
This should be applicable no matter what to things like Zap Cannon, Sacred Fire, and Dynamic Punch though.
edit: Or alternatively whenever thief fails to take an item while you have no item, put a message saying that this could be due to no item or due to a small chance for thief's effect to fail. That'd probably be the simplest implementation that doesn't reveal information. If it gets spammy it could just be a once-per-battle message when it happens.
edit2: Obviously less important, but adding a message for when you overflow your own stats would be helpful as well. Shouldn't reveal any information since all the factors that do overflow stats are visible to the player. That's probably the only remaining glitch that regularly confuses players.
This should be applicable no matter what to things like Zap Cannon, Sacred Fire, and Dynamic Punch though.
edit: Or alternatively whenever thief fails to take an item while you have no item, put a message saying that this could be due to no item or due to a small chance for thief's effect to fail. That'd probably be the simplest implementation that doesn't reveal information. If it gets spammy it could just be a once-per-battle message when it happens.
edit2: Obviously less important, but adding a message for when you overflow your own stats would be helpful as well. Shouldn't reveal any information since all the factors that do overflow stats are visible to the player. That's probably the only remaining glitch that regularly confuses players.
Last edited:
Lavos
Banned deucer.
Bug Catcher Lavos back with another small installment of Gen 2 sim glitches.
1) Recoil damage on Substitute: https://replay.pokemonshowdown.com/gen2customgame-716773035
On turn 33 of this replay you can see that Tauros used Double-Edge to attack the Smeargle's Substitute. In Gen 2, recoil damage is calculated based on how much damage was inflicted on the Pokemon that received the attack. Since Double-Edge hitting Substitute means Smeargle isn't technically taking any HP loss, the game should inflict exactly 1 HP of recoil damage to Tauros. However, as you can see in the replay, Tauros sustains 20 HP of recoil damage. This is because Smeargle's Substitutes have 78 hit points, and Showdown is incorrectly calculating recoil damage based on the damage that the Substitute took, so 78 * 0.25 = 19.5, rounding up to 20 HP of recoil. This is surprisingly relevant in the case of something like Double-Edge Snorlax attacking a SubSeed Exeggutor and taking more recoil damage than it should.
2) Additional move effects when Substitute is active: https://replay.pokemonshowdown.com/gen2customgame-716776033
In Gen 2, if a Substitute is active at the beginning of an attack targeting the Substitute, no secondary effects from that attack can trigger on the Pokemon which had a Substitute, even if the Substitute is broken during a multi-hit attack. That means if Beedrill uses Twineedle on Smeargle's Substitute, even though each hit would normally have a 20% chance to poison, it should never activate that secondary effect, even if the first hit of the attack breaks the Substitute. Unfortunately, as you can see on turn 8 of this replay, Twineedle's first hit breaks Substitute, and the second hit inflicts poison on Smeargle. Please fix this quickly, the stability of the Gen 2 Randbats metagame is at stake.
1) Recoil damage on Substitute: https://replay.pokemonshowdown.com/gen2customgame-716773035
On turn 33 of this replay you can see that Tauros used Double-Edge to attack the Smeargle's Substitute. In Gen 2, recoil damage is calculated based on how much damage was inflicted on the Pokemon that received the attack. Since Double-Edge hitting Substitute means Smeargle isn't technically taking any HP loss, the game should inflict exactly 1 HP of recoil damage to Tauros. However, as you can see in the replay, Tauros sustains 20 HP of recoil damage. This is because Smeargle's Substitutes have 78 hit points, and Showdown is incorrectly calculating recoil damage based on the damage that the Substitute took, so 78 * 0.25 = 19.5, rounding up to 20 HP of recoil. This is surprisingly relevant in the case of something like Double-Edge Snorlax attacking a SubSeed Exeggutor and taking more recoil damage than it should.
2) Additional move effects when Substitute is active: https://replay.pokemonshowdown.com/gen2customgame-716776033
In Gen 2, if a Substitute is active at the beginning of an attack targeting the Substitute, no secondary effects from that attack can trigger on the Pokemon which had a Substitute, even if the Substitute is broken during a multi-hit attack. That means if Beedrill uses Twineedle on Smeargle's Substitute, even though each hit would normally have a 20% chance to poison, it should never activate that secondary effect, even if the first hit of the attack breaks the Substitute. Unfortunately, as you can see on turn 8 of this replay, Twineedle's first hit breaks Substitute, and the second hit inflicts poison on Smeargle. Please fix this quickly, the stability of the Gen 2 Randbats metagame is at stake.
Im sorry but I searched the bulbapedia[ https://bulbapedia.bulbagarden.net/wiki/Twineedle_(move) ]and found this sentence:
Generation II
If the first strike breaks a substitute, the second strike can poison the target.
maybe we should test it on the real game
Bulbapedia is known to be inaccurate with game mechanics, especially in relation to old Gens. I'd trust Lavos on this one.
Just caught this, I wonder why this slipped under my eye for such a long time lol.
Currently you can change SpATK DV's and SpDEF DV's as well as their EV's without any probelm independent of each other, but that should not be the case. It is true that both display different stats, but the idea of "Special" as it was in Generation 1 didn't really vanish, as both stats are tied to each other by having the same DV and EV Value.
See Bulbapedia for one source, since they got it right there. I can find more of them if needed.
https://bulbapedia.bulbagarden.net/wiki/Statistic#Special
Currently you can change SpATK DV's and SpDEF DV's as well as their EV's without any probelm independent of each other, but that should not be the case. It is true that both display different stats, but the idea of "Special" as it was in Generation 1 didn't really vanish, as both stats are tied to each other by having the same DV and EV Value.
See Bulbapedia for one source, since they got it right there. I can find more of them if needed.
https://bulbapedia.bulbagarden.net/wiki/Statistic#Special
https://replay.pokemonshowdown.com/gen2ou-758776038
Towards the end of the replay Tyranitar's ability sandstream is activated, even though there are no abilities in Gen 2. The rest of the game is played out with the sandstorm condition.
Towards the end of the replay Tyranitar's ability sandstream is activated, even though there are no abilities in Gen 2. The rest of the game is played out with the sandstorm condition.
Apparently Marty fixed this.
According to the description of the item on pokemon showdown and bulbapedia, Lucky Punch is supposed to increase Chansey's critical hit rate to 25%. After getting suspicious due to a very small number of critical hits in a game I spectated I tested with a friend to get a larger sample size. Out of 480 Water Guns, only 31 of them were critical hits which is what you would expect if Chansey didn't hold any item (~6,4583%). Here is the replay.
I obviously have no proof that we both used Lucky Punch as items but you can try it out yourself and see.
I obviously have no proof that we both used Lucky Punch as items but you can try it out yourself and see.
xJoelituh
Banned deucer.
I think you were just unlucky? https://github.com/Zarel/Pokemon-Showdown/blob/master/mods/gen2/items.js#L54