Based on just 3 lap timing and best lap times, i've configured the best 1st lap time and the best timing for a lap on sunny beach.

From what i saw it is mathematically possible to reach 1m 2s 429ms.

From observing the speed ratio of best lap timings, and after 131 km/h was reached, most best laps never reach 129km/h, most reaching 127km/h but have better avg km/h timings.

After an average timing from turning around corners, if the turns maintain a 85+km/h power while reaching a 131km/h terminal speed on its 1st straightaway then a sub 1 m 2seconds is also possible. Considering the runs, and turns it is also possible for the best lap to be sub 20s according to just math. Though unlikely.

The limiting factor is waves and their cycles - you're correct these kinds of times are quite possible, but unless new tech's found, only on "calm" wave settings. On standard waves, you lose too much time from airtime and readjustment to get anything close to this, and typically, one of your first or second lap being fast comes at the expense of the other lap being made much slower.

I wrote a bit more back in a discussion of the Will Lam probably fake time thread on SB human limits discussing the question from two angles, "sum of best lap times from top runners" and "sum of best known recorded lap times".

The 102.6 is indeed impossible unless something new's discovered or play significantly improves. If you look at top times and best laps from the top 15 or so best members of the leaderboard seen in a record, they were 22.338 by shibbypod in a lap 1 pb, and the 20.412, and 20.177 in Shibby's record. They sum to 01:02.927.

That lap 1 time is borderline optimal. A 20.3 is roughly the current optimal lap 2, and lap 3's record is a 20.096 on a different difficulty setting. On that 1:02.927 there's therefore MAYBE another theoretical 200ms to save - but saving this without new tricks would first not only require one of the fattest WRs this community had ever seen to shave 186ms on our most competitive time, but also a theoretically perfect race where the player achieved the best lap ever done in this community, on literally all 3 laps of the same race, despite the map typically only letting 2 out of 3 laps be fast - and then you STILL would need to find another impossible to account for 30-40ms to get to their time.

Pretty much unless they found a glitch that gains speed or a shortcut, that time's fake.

Correct, while it is possible the chances anyone gets to those optimal settings is near considered impossible. My hopes is too establish a bounds of known possibility. From only mathematical scenarios even while running on calm myself i have a established a bounds here, though i've ran faster times lap wise this only certifies a sub 1:3.000 is possible. (Click Me) I was not able to break a 129 km/h barrier. I had noticed the waves provide a very crucially small speed boost when riding the waves as close as nearly vectorially possible, provided that the rider was able to do this, i've noticed shibby was able to get to a 131 km/h whilst riding a wave. Now with further study in order to beat shibby's record a rider needs a few factors into consideration. The rider needs to get to a speed of 127 km/h then on the first turn maintaining at least 85 km/h around turns, after the first turn 129 km/h around the very first straightaway against three waves, the rest of the straights need to be maintained around 129 km/h. According to speed times, and wave vectors a simple B press needs to occur within a .025-.075 time frame. to simplistically level the rider closest to the wave on straights.

To note past records from magazines where only a picture was needed for submission. It could have been possible that the wr was broken, could the riders have been using calm settings before to obtain those times? Now im aware of Misuken (I think the spelling is right) being actually real along with other notable times that seem feasible... Now there is an even faster time belonging to a Chris Murphy, of Manchester who had a validated NTSC record time at a 1:02.694

To cite the credibility of this one record would be unknown even a sub 1:3.000 has never happened yet. Though even in this article that was posted under this title COULD SB 1:03.755 BY "WILLIAM LAM" FROM 1998 HAVE BEEN REAL? I can safely say that any of the prior times such as william lam and below could actually be real. It would account for a natural bell graph within statistics, even the times below are more realistically what we have seen is possible even for the best of players. The frequency of them are relatively low.

In addition to the above previously mentioned i have proven that through my best lines a on SB(Sunny Beach) first lap best time being a 22.361, and fastest lap times (though not official from shibbypod) being a 20.077 if im not mistaken brings the theorectical barrier to 1.2.515, though i had previously broken the unofficial record by .006, making the new single lap best 20.071. This brings the barrier down to 1.2.503.

I can with confidence say that these times are possible. I will be happy to give a hint (I did not press B as often as one might think...rarely infact). Also i was able to void the booeys. Though in a completely absurd way. Though! This was just a test, and contributed nothing to this new theorectical time and represents a far more lower time. For now thats just an experiment until i can sufficiently re-create a time below 1.0.000 A sub (0.59.000) could soon be possible.

I just want to clarify that these will remain unofficial times. Until I have replicated them under the same conditions.

Also... I broke the 131 km/h barrier. I got to a 136 km/h. Convincingly enough for only a few frames.

A small clerical error that for some reason is not letting me fix above.

~My Best first lap tested time = 22.361

○With shibbypods best laps added consecutively... a 3 lap race would conclude to a 1.2.515

○ With my tested time beating shibbys by .006ms the threshold would now be 1.2.503

●Mention to the time being re-created i meant to state (1.0.000) not 1.2.000. A sub 59s 3 lap is in the works.

●Also when refering to speed i was trying to mention km/h.

Just for kicks, I figured I'd try out calm water. First attempt was a 1'01"860 in case it helps with your analysis. Just my two cents but I could see a lower 1'01" (maybe even sub 1'01"5... but I doubt a 59" 3 lap is in the cards (though I'd be tickled to see it).

That said, I don't think calm water really tells us anything about what might be possible with default water (which is what the vast majority discuss and compete on). I think we'd need the game to be decompiled to make any significant headway on that.

Love that you're trying though. I also tried my hand at some Sunny Beach analysis but it ended up being kind of pointless in the end. I thought that maybe I could analyze runs by writing a script that: (1) broke the video into individual frames, (2) used OCR to read the elapsed time and speed shown in the frame, and (3) use that information to reconstruct a model of the course and which segments had the highest potential time saves. I actually wrote working code for (1) and (2). Then I realized that I had multiple things preventing me from actually analyzing the run. For example, the last two digits in the on-screen time are repeated in a predictable cycle that doesn't correlate with real time. This is probably a short cut to save memory space but it means that you probably aren't getting accurate enough data from analyzing the frames. So I abandoned that project and started looking into how decompiling a 64 game works (oof)... but trying was fun and I did learn some things about the game by trying. Long story short: do your thing! I'm curious to see your conclusion.

That's some crazy fast speed. The whole reason i needed to establish boundaries for a control. Calm being the control group. Though it doesn't help if your numbers are incorrectly respondent to the information...ie: insufficient amount of data. Likewise, you've peaked my interest with that lower time.

Also, way ahead of you on that. I do have some experience with reverse engineering (not an expert) but I will be more than happy providing any new information regarding the game. I have the game files and low level memory booted, specifically the physics base is quite nominal. Interesting to say the least. Even the raw files from a comp error I manage to de-script a sizeable amount of code, but currently working brute force to solve the key pattern... Fascinating enough that injecting new values are possible, i think I've found the nessesary values but as of yet still need time to process. Anything I upload regarding code will be handled with care, considering legality and for the sake of integrity of the community.

Further more, nothing in the base game has been altered or changed. I'm for now continuing to test information with a physical copy or a digital (Switch), I do have several emulations as well. Eventually ill move to change values to see interactions on an emulation and eventually to test that theory injecting code fixes to a physical N64 copy. I'll be testing the pal version first and will leave complete video documentation regarding the topic.

As for my personal endevors, ill still be submitting authentic times with video. Aiming to beat that WR or my PBs.

I could only imagine... 60fps Original Wave race...4k quality. Same graphics though.

Also, I guess I should spill the 🫘 the issues regarding booeys involve "booey jumping", using the waves to jump over the the physical barrier. I was able to gain considerable height, and I think it's possible to "bounce" across the lap median.

Just hit 145 km/h on sunny beach but lost control and hit a booey on first lap. This was on Normal difficulty, time trials.