I Used Math to Create an Infinite Pokémon Battle

A mathematician walks into a bar and orders a pint, half a pint, a quarter pint, an eighth pint... The bartender gives him two pints and tells him to split it himself.

Parents: one more battle then it’s lights out. adef: those fools!

Ah yes, Leppa Recycle Slowbro is obviously too simple, let's use Zeno's Paradox instead.

Slowbro: "Look at what they need to mimic a fraction of my power!"

And here I was assuming this was about Recycle/Heal Pulse/Block/Slack Off on a Slowbro w/ a Leppa Berry

Funbro stares blankly

Dude you don't need math to do this you just need to bring a Primeape with Rage to Lorelei's Dewgong in Gen 1

Most of my knowledge of the actual mechanics of Pokémon games comes from Pikasprey's Soft Lock Picking series. It's nice to see that there's more than one of him in the world.

Pikasprey “look what they need to mimic a fraction of my power!”

Not gonna lie, if you told me that someone set up a battle that can only be won by solving Zeno's Paradox, I would've assumed it was Pikasprey.

My personal headcanon for how this works lore-wise is that Raticate is hitting Tropius with smaller and smaller forces until it uses a unit of force so small that it somehow breaks physics and the universe collapses in on itself. All that remains is a Rat gnawing on a Tree floating in the void, much like the Nidhogg of legend gnawing on the roots of Yggdrsil

All this complicated maths falls to the simplicity of two wobbofetts unable to do more damage and take more recoil than leftover heals with struggle.

16:32 Assuming Super Fang was coded 100% correctly here, missing 4 Super Fangs in a row is a 1/10000 or 0.01% chance!

It's probably worth mentioning that these games weren't even on a console that had floating point numbers until Gen 4. Even division was a challenge- the Gameboy didn't have division at all, and the GBA had it but it was so much of a pain to use that you wouldn't use it unless you REALLY needed it. Super Fang wasn't implemented with any kind of division, it was implemented as a bit shift (which also explains why it was natural to round down). Edit: I'm not quite sure what Atsign did here, but I suspect he took the game's decompiled source code, replaced "i16" with "float" in a bunch of places, and ran a modern C compiler on it. The original GBA devkit would throw an error if you did that (because the GBA didn't support floats), but the modern compiler has a built in library that simulates IEEE floating point numbers with whatever hardware you have, and that's probably what actually is being used here.

As a software developer, my guess was that three things could happen: 1. The value would be so small game would just end up -0 infinetely as the fraction resulting of the division would be too small and return 0 2. The division would actually underflow and return a really large number of damage (I guess literally the overlfow limit?), and the attack would then underflow the HP. 3. A stretch, but maybe the CPU would just throw an exception, and if not caught by anything, the game would unceremoniously lock up, just like it does when encountering unknown opcodes or dividing by 0 when doing glitches, not a visual spectacle like atsign provided us with, but the game just freezing, including the audio. I've seen underflows happen on hardware with the pomeg berry glitch... so that was why my bet was on 1 and 2, but I'm still interested to know what actual hardware would do, as the Berry glitch is substraction underflowing, not division, as atsign explained.

Lorelei, Funbro and Wobuffet: we don't need fractions.

Ah yes, Achilles and the Squirtle

9:50 atsign the GOAT

Use /l/adef or code ADEF to get Displate's largest discount of the year! #ad

In today's episode of "adef turns anything and everything into a Pokemon video": Achilles and the tortoise