[DiscordArchive] Did you try this? Did it work for you?
[DiscordArchive] Did you try this? Did it work for you?
Archived author: carmage • Posted: 2025-02-20T18:14:56.392000+00:00
Original source
Did you try this? Did it work for you?
Archived author: Bench • Posted: 2025-02-20T18:15:03.018000+00:00
Original source
No, I did not try it
Archived author: Shin • Posted: 2025-02-20T19:25:56.876000+00:00
Original source
I said I had a quick look and to wait for other devs feedback. Please do as the other devs request
Archived author: carmage • Posted: 2025-02-20T19:26:55.521000+00:00
Original source
I tried and it doesn't work for me.
Archived author: walkline • Posted: 2025-02-20T20:22:35.220000+00:00
Original source
Alright, I tried to investigate this issue further and test your theory using the new similar to suggested collision test approach.
I added some code to the movement opcode handler, where I'm working with the raw coordinates sent by the client.
I also wrote a naive implementation to get the maximum ground height in a circle. I believe it works similarly to the suggested collision test, right?
Here is the test output:
```
[MSG_MOVE_FALL_LAND 0x00C9 (201)]: Z: -56.393646, ground with this Z: -56.393734, actual ground at point: -56.393734, max ground in circle: -56.393646.
[MSG_MOVE_START_FORWARD 0x00B5 (181)]: Z: -56.393646, ground with this Z: -56.393734, actual ground at point: -56.393734, max ground in circle: -56.393646.
[MSG_MOVE_STOP 0x00B7 (183)]: Z: -56.378326, ground with this Z: -100000, actual ground at point: -56.37831, max ground in circle: -56.378326.
[MSG_MOVE_JUMP 0x00BB (187)]: Z: -56.378326, ground with this Z: -100000, actual ground at point: -56.37831, max ground in circle: -56.378326.
[MSG_MOVE_HEARTBEAT 0x00EE (238)]: Z: -54.81194, ground with this Z: -56.37831, actual ground at point: -56.37831, max ground in circle: -54.81194.
[MSG_MOVE_FALL_LAND 0x00C9 (201)]: Z: -56.3782, ground with this Z: -56.37831, actual ground at point: -56.37831, max ground in circle: -56.3782.
```
It seems that you are right. Even on such a flat surface, your suggested approach has an impact. Interestingly, it works perfectly, the Z position from the client matches my naive implementation of getting the maximum height in the circle.