Mario Kart MB - Print Page

Mario Kart MB
https://www.mariokart64.com/cgi-bin/yabb2/YaBB.pl
Mario Kart >> Mario Kart 8 >> RNG?
https://www.mariokart64.com/cgi-bin/yabb2/YaBB.pl?num=1466222676

Message started by Jazzy on 06/17/16 at 20:04:36

Title: RNG?
Post by Jazzy on 06/17/16 at 20:04:36

Today I was curious about something in MK8, so I decided to run a test. I came to the likely conclusion that there is some sort of RNG in MK8 runs, and that, given the same controller inputs, the final time can vary. What follows is a description of the test I ran and its results.

The Test

The test was performed using Morton, Mach 8, Cyber Slicks, and Gold Glider on SNES Rainbow Road, using the Wii U Pro controller. While connecting the controller, I held the left stick up and slightly to the left, and the right stick down. By moving the left stick, I simulate a constant right turn (without drift), as the controller will think it's in held slightly to the right when it is in fact in the neutral position. Similarly, as the right stick can be used in place of the A button, moving the right stick while connecting the controller ensures that the kart is always accelerating.

I went into the time trial, pressed restart, and left the controller alone. The kart, with the controller set up the way it was, can complete the course this way. After the start of the 2nd and 3rd laps, after passing the finish line, I pressed X (rearview) to keep the controller alive so it would not disconnect.

I repeated the race by pressing restart and letting the kart go on its own. This way, I can compare two runs with the same inputs.

Results

Race 1: 2:01.481 / 2:02.084 / 2:00.541 = 6:04.106
Race 2: 2:04.214 / 2:05.367 / 2:01.834 = 6:11.415

http://imgur.com/oMnuNSM
http://imgur.com/KpKGB9O

The two races had significantly different results. From watching the run, I observed, that on the straight before the forked path, in the first run, the kart did not fall off at one point, but in the second run, it did. I presume this was due to different alignment coming into that turn, possibly due to the waves.

Conclusion

Although I'm still not 100% sure, I'm quite convinced that RNG may play a role in races. This would also explain why ghosts are no longer input based. I'm not sure what to think of this all right now, but it's interesting for sure.

If you're saying that me pressing X is "not the same input", consider that I didn't press X until the second lap, and the lap 1s were different as well. I'm pretty convinced there's more going on.

I didn't make a video, because, to be honest, I didn't expect a difference in the times; and I don't have a good way to make a video anyway. But, feel free to try my experiment on your own.

Title: Re: RNG?
Post by Rhodechill on 06/17/16 at 23:39:54

Yea and the Yoshis in the backgrounds change colors too

Title: Re: RNG?
Post by WillD on 06/19/16 at 08:20:28

That's really interesting! I look forward to more investigating on this :)

Title: Re: RNG?
Post by seveirein on 08/16/16 at 14:54:43

This is an old thread, but I want to point out that the value sampled off an analog stick will drift slightly, due to a variety of factors. It will fluctuate slightly due to noise, temperature variation, etc. This will not have practically any effect over a very short period, but it will definitely cause minor variations to eventually occur, which of course can cause drastic time differences.

To truly test this, the best way would be to use hacking tools to force exactly the same input to the digital domain, bypassing analog to digital conversion entirely.

Title: Re: RNG?
Post by Jazzy on 08/16/16 at 15:43:08

Good to know. I wouldn't have known this, since I have never worked with analog devices before (only digital). But I can see how what you say makes sense. I still think this is worth testing, when it becomes possible to.

Title: Re: RNG?
Post by Mick on 08/24/16 at 01:05:48

How about doing the same test with the d-pad instead?