Harnok & CraftyController
You ever notice how that new update redefines the physics in the mid‑level? I think there's a hidden mechanic that could let us skip the whole platform chain—want to dig into it together?
Sounds like a challenge. I'm all for efficient paths, but I need the details first. Let's review the physics changes and see if that loophole holds up.
Alright, the update tweaked the momentum curves for mid‑air jumps—gravity drops a shade at 0.3‑second intervals, and the acceleration cap has been lowered by 10%. This means the apex is now 15% higher, but the fall speed is 12% faster after the cap. If we time the second jump exactly at the 0.28‑second mark, the momentum carry‑over can be maxed out. The loophole? You can line‑up a 180° spin on the first jump, hit the platform just after the cap resets, and the physics engine treats the spin as a continuous impulse, giving you a 3‑tile lift. We just need to verify the spin angle tolerance—if it's within ±5°, the bonus sticks. Let me run the simulation and ping you with the numbers.
Okay, give me the readout and the spin‑tolerance curve. I'll check if the 5° window holds under repeated trials and see if the 3‑tile lift is consistent. If not, we’ll fall back to the usual path.We complied.Got it. Let's see the numbers.Nice, hit me with the stats.Alright, send over the data and let’s see if that spin trick survives a test run.Got the data, thanks. Let’s run the simulation and confirm the spin tolerance.I'll check the tolerance curve and run a few trials. Let's see if the spin holds up.
Here are the numbers from the last 200‑trial batch:
- Momentum carry‑over average: 1.18 × the base boost
- Spin angle tolerance curve (°) – 0° gives 0% extra lift, ±5° gives 100% of the 3‑tile bonus, ±10° drops to 75%, ±15° to 50%, and beyond 20° the bonus disappears entirely.
- Success rate for ±5° window: 94.6%
- Consistency: In 18 out of 20 runs the 3‑tile lift held after the second jump; the two failures were due to a 1‑frame jitter in the jump trigger.
So yes, the 5‑degree window is solid, but make sure your input timing is locked to the 0.28‑second mark. If you hit that and stay within ±5°, the lift should be reliable. Good luck with the tests.
Got the numbers. 94.6% is decent but not perfect—those 1‑frame jitters are a pain. I'll lock the trigger and run a few dry runs to iron out the timing. Thanks for the data. Let's see if the physics holds when we push it to the limit.