bumping this post. seems to only be a problem in DPP with the toggle not actually changing the damage output. additionally, the base toggle w/o metronome (never, once, twice, etc.) also doesn't seem to have any affect on the calc's damage output
-
Follow our Instagram!
-
Curious about what we're currently doing about bots? Check out the latest updates from staff in this thread.
-
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.
PS! Pokémon Showdown Damage Calculator
- Thread starter Austin
- Start date
Found a bug; not sure if it's a bug with showdown itself or the calculator. In the champions calculator, a Garchomp with Jolly and 32 points in speed; but lowered by one stage, is 111. In practice, on the showdown ladder, it's actually 112.
Since "[Champions] NatDex Draft" has been released now as a format on Showdown, could the Champions-Calculator allow Pokémon outside of the Champions-Roster please?
Hi i noticed Nightmare isnt listed on showdown calculator https://calc.pokemonshowdown.com/index.html?gen=2 i think it would be good to implement considering how often nightmare can play in a gsc game since sleep and rest talk is good in this . is there any way to implement nightmare damage on the calc? considering it acts at the end of the turn and deals 1/4 of the opponent pokemon max hp. And if bad dreams can work on the calc why nightmare cant
Last edited:
Briefly looking into the data that we use for the random battles sets for the damage calculator, I noticed that all of the sets for every Pokemon are the same. Not just in our damage calculator, but in the data we pull from. I wrote a Python script (whose source I have written in this post) and tested whether the sets in our data entirely match between the two formats.
Python:
import json
format1 = 'gen8randombattle.json'
format2 = 'gen8randomdoublesbattle.json'
print(f'Comparing {format1} with {format2}')
setequality = []
with open(format1) as f1:
sets1: dict[str, dict] = json.load(f1)
with open(format2) as f2:
sets2: dict[str, dict] = json.load(f2)
for mon in sets1.keys():
if mon in sets2:
equal = sets1[mon] == sets2[mon]
# print(f'{mon:20} {equal}')
setequality.append(equal)
print(f'Are all sets equal: {all(setequality)}')Running this script on the Gen 9 formats gets us...
...the expected result.
Running this script on the Gen 8 formats gets us...
...something that validates your post.
Let's look at how sets are generated for Gen 9 random battles formats:
It looks like there are sets for the random battle and the random doubles battle formats.
What about for Gen 8 random battles?
It looks like it doesn't have sets for Gen 8 random doubles battle formats. Maybe what happens is that for Gen 8 random doubles battles, it just uses the same sets as for Gen 8 random battles. Maybe A Cake Wearing A Hat could correct anything about my post if it's incorrect.
Gen 8 Random Doubles is...not supposed to exist right now. It's supposed to temporarily be unplayable until we're done revamping it. We'll have to fix that.
It seems that the -1 speed calculations in the Champions section are incorrect. For example, it shows that a 32+ speed Mega Alakazam's -1 speed is 146 but it's actually 148 in battle. It's also 148 if you try calcing at level 50 in the SV section (The actual 2/3 x 222 calculation is also 148 btw). I've checked with multiple Pokemon like Scarf Excadrill -1 should be 153 but it shows 151 in the Champions section. Max Speed Garchomp -1 should be 112 but it shows 111.
A user reported a large discrepancy in the randbats Gen 1 calculator:
Randbats calc:
Lvl 70 Articuno Blizzard vs. Lvl 90 Ekans: 179-211 (72.7 - 85.7%) -- guaranteed 2HKO
Regular calc:
Lvl 70 Articuno Blizzard vs. Lvl 90 Ekans: 142-168 (57.7 - 68.2%) -- guaranteed 2HKO
The regular calc is what's right (or at least that's what matches Showdown). The difference lies in the special stat on Ekans; randbats calc has it much lower, at 104. No idea why it's different, did not try to debug myself.
Randbats calc:
Lvl 70 Articuno Blizzard vs. Lvl 90 Ekans: 179-211 (72.7 - 85.7%) -- guaranteed 2HKO
Regular calc:
Lvl 70 Articuno Blizzard vs. Lvl 90 Ekans: 142-168 (57.7 - 68.2%) -- guaranteed 2HKO
The regular calc is what's right (or at least that's what matches Showdown). The difference lies in the special stat on Ekans; randbats calc has it much lower, at 104. No idea why it's different, did not try to debug myself.
There's a discrepancy between PS! and the damage calculator for old gens LC where some attacks show the wrong amount of damage. Tested in ADV LC and it seems to affect DPP as well, but maybe not BW onwards. It's been difficult to reproduce due to only affecting specific damage ranges, so I'm not entirely sure why it's happening. I've tried to test with dual types, multihits, and EVs, but none of them seem to be consistently causing problems.
Examples of calcs where two targets have the same defensive stats, but the one with only neutral defensive types is shown to take more damage:
I thought it was just multihits before this, but HP Fighting has the issue here instead of Double Kick. It also does different amounts of damage to Swablu and Anorith, so maybe it's related to the EVs being applied??
Water Spout apparently doing more damage to 0 EVs Skitty than 0 EVs Barboach with the same special bulk, but then doing the same amount to both when I change their bulk to match a Lileep:
Teams and replays for testing Bonemerang:
https://pokepast.es/f54f87e1460878d6
https://pokepast.es/b31e2f4be05f9f22
https://replay.pokemonshowdown.com/gen3lc-2649569303-79lag2trkjfgej1r8wnrst8envp8700pw
https://replay.pokemonshowdown.com/gen3lc-2649547136-tfji3rg4vjyk9x5oqjquw7sb1tic63lpw
Notably, the Lileep takes 13 damage from one hit of Bonemerang, when the calc says it should only be taking 12 at most. (First screenshot in the spoilers)
Examples of calcs where two targets have the same defensive stats, but the one with only neutral defensive types is shown to take more damage:
Teams and replays for testing Bonemerang:
https://pokepast.es/f54f87e1460878d6
https://pokepast.es/b31e2f4be05f9f22
https://replay.pokemonshowdown.com/gen3lc-2649569303-79lag2trkjfgej1r8wnrst8envp8700pw
https://replay.pokemonshowdown.com/gen3lc-2649547136-tfji3rg4vjyk9x5oqjquw7sb1tic63lpw
Notably, the Lileep takes 13 damage from one hit of Bonemerang, when the calc says it should only be taking 12 at most. (First screenshot in the spoilers)
its a problem with showdown, not the calc. its related to a pokemon's typing, see lileep in your calcs which has a rock typing to be weak to grass but then a grass typing to resist grass. The Damage Calc shows you the correct output, to get what showdown is doing change the damaged pokemon's typing to be single type neutral (i.e make lileep a pure normal).
Hey, the set for Mega Blastoise on the calc set for SM RU has Rain Dish as its ability even if you select the mega. It should be Mega Launcher. Actually, the normal one has the wrong ability too, though less important, it's supposed to be Rain Dish in accordance to its analysis but it's not.
Dragonize boosts every single move including non-Normal ones.
My brother has lost a tournament game because of this please do something; he is very sad atp.
Fixed
Hi, don't know if this is the place for this post but I think the run and bun calc uses a smogon sprite library and they are not correctly loading at the moment, thanks :)
There is a bug in the damage calc that I assume has to do with rounding somewhere. In this screenshot you can see the calc say that there is a 6.3% chance to OHKO after rocks, but Diglett-A has 18 HP, meaning that it takes 1 damage from rocks and would be 17 HP for the purposes of this calc.
The correct display should say "50% chance to OHKO after rocks", not a 6.3% chance. My assumption is that this is somehow related to the calc rounding the stealth rocks damage since diglett-a is a 4x resist but I'm not entirely sure.
The correct display should say "50% chance to OHKO after rocks", not a 6.3% chance. My assumption is that this is somehow related to the calc rounding the stealth rocks damage since diglett-a is a 4x resist but I'm not entirely sure.
