From 1st December 1999 to 14th January 2000, Reebok sponsored a challenge exclusive to the PAL release of Sonic Adventure to promote their DMX lineup of sneakers. Contestants across all of Europe would play the first half of Emerald Coast, grabbing 5 pairs of sneakers before reaching the goal in the fastest possible time. The winner was one Neil Riddaway from the UK, with a time of 46.09 seconds. His prize? £1000 worth of Reebok clothing, and a personalized Dreamcast kiosk. He also got an article about him written in Official Sega Dreamcast Magazine. He's probably been interviewed multiple times since then, but the only one that I can find was done in 2021 for SEGA Dreamcast Info. Now, here's where it gets interesting. I presented his time of 46.09 to the Sonic Adventure Speedrunning Discords. Both of them, for Dreamcast and DX. The consensus, even amongst the game's top runners, is that this time is impossible. Niczur, the current WR holder for DX, attempted to replicate Neil's run by playing the Dreamcast DLC mod in the PC port, and his time was a 45.83: Some notes to take away from this: Niczur's execution was not flawless, however he estimates he could only have saved 3/4 of a second. That puts a fully optimized run at around 45.08, or just over a second below Neil. In the video, Niczur uses a strat known as "Whale Skip" at 0:33. Whale Skip saves about 1.5 seconds off a normal run, but there is simply no way Neil would've achieved a whale skip, as it wasn't discovered until 2021. DX and DC are practically two different games, and in speedrunning Emerald Coast specifically, DC lags behind DX runs by around 8-9 seconds. And on a PAL system running at 50hz, this likely would've been even slower. The DLC timer starts before the level is ever entered. Due to the enhanced loading times of the PC version, it only takes him a second to enter the stage. However, on an actual PAL system (and in Redream) the fadeout is much slower, resulting in a three-second delay: . With all this in mind, if we're being generous, then the absolute most optimized run Neil could have achieved in 2000 without using external exploits is around 53.64. And again, that's being generous, given what the scene was like at the time. So how can we explain this discrepancy? Could he have used a cheat device? Nope. Neither Action Replay CDX nor Xploder had released yet at the time. Could he have modified his game in some other way? Nope. A 15-year-old boy possessing both the tech and the knowledge to hack a Dreamcast game in 2000 is improbable with how primitive it would've been. Could he have simply faked his time? Nope. It's not as if he just posted a photo of his time; it was saved to his VMU and then that data was uploaded to an online server. Said data was in a proprietary format that could only be read by other VMUs. Pretty hard to fake that. WitherMin, the current WR holder for the Dreamcast version, believes the time could only have been achieved if Neil used a MadCatz Dream Pad controller, which had just been released in December 1999. The controller causes analog input to go farther than intended, resulting in characters running well-beyond their top speed, which as you can imagine, can enable some crazy skips. This is disqualifying for a modern WR attempt, but it would still have been allowed under the Reebok challenge rules. However, according to Neil in the interview I linked above, while he was aware of the MadCatz exploit, he did not use that controller. Neil had no reason to lie about not using a MadCatz controller. It had been 21 years, it wouldn't have been against the rules, he has no world record to uphold, and he has no reputation in the community to maintain. There would've been no shame in admitting it. So why doesn't he? I suppose it's possible that he's telling the truth and he's just a Sonic god. However, the interview reveals other tidbits that make Neil seem rather suspicious. Not only did he somehow manage to submit a perfect challenge-winning run at the last minute, ahead of everyone but one other person by a few seconds, but his save file with the winning time was somehow deleted by his 3-year-old nephew. It sounds awfully convenient, doesn't it? In the end, I was left to conclude that either Neil has been holding onto skips and strats the speedrunning community has never heard of for 25 years...or he's not being entirely truthful. Is he like Matt Turk, a genuine legend who managed to get such ludicrous times that it took runners decades to figure out how he did it? Or is he like Todd Rogers, a con-man who managed to fool the speedrunning scene for fame and fortune, and get away with it for decades? Hero, or villain? There was only one possible avenue left to explore, and that was to ask the man himself. His twitter DMs, however, are shut down, and as of writing, he has not responded to my interview request on Facebook (which contained no accusations, before anyone makes assumptions). I noticed, however, that in his Twitter banner, he still proudly displays that Dreamcast kiosk. Perhaps he has something to protect after all: his pride.
Whilst I love this as a potential bit of drama and investigation... There is one... strange anomaly with it. In the magazine it says "Neils winning time was 46.09 just 0.05 seconds faster than 2nd place." Well that would mean the leaderboard should look like this... 1st place: 46.09 2nd place: 46.14 But if we're estimating the winning time is more likely to be around 53.64, that would mean the 2nd place is also questionable? So it would suggest there's two likely situations. 1: Neil did win, but his winning time has been reported/recorded incorrectly. 2: Both first and second place cheated... potentially more if the records could ever be found lol.
Yes, 2nd place is questionable as well. In the interview, Neil mentions that he and the 2nd place runner were ahead of everyone else by a few seconds, and that the time trial clans of the day were baffled as to how they were getting their times. 3rd place was likely much closer to the figure I gave.
Occam's razor - he had a strategy that hasn't been documented, and doesn't want to be bothered about a video game he played two decades ago.
This game is aggressively speedrun and the idea that one or two people from twenty years ago had a novel strategy that hasn't been employed yet, on top of the gains being so large over the established WR progression makes this theory outside of Occam's Razor IMO, lying about the Madcatz Controller seems more likely.
Using third party controllers to gain advantage in those competitions was seen as cheating and looked down upon by players, and at one point it was addressed by the SA1 site, telling players to refrain from using those controllers, and making them officially against the rules. Granted, that rule was established after the Reebok competition. But I would say he does have a reason to hide it if he used that.
He saw Marco ahead of him and decided to lean forward. Classic gamer move. QED. Does anyone have one of those 3rd party controllers on hand and is good enough to try the run using the strats known then (if any) and now?
I remember the only time I actually beat Devildoom on StH was with a madcatz controller. I just mapped the fire to turbo.
Why the Madcatz pad makes characters go faster is explained here, looking into more about the history of this I noticed this post from last year is there any truth to this? Maybe it’s what this guy did instead of using the dream pad.
As shown in the deadzone video in your post, the original first-party Dreamcast controller correctly limits thumbstick input to a circle; however the system itself doesn't, allowing the full square to be used. Speedrunners of Jet Set Radio and Blinx (the Xbox has the same issue) use third-party controllers as it allows for better acceleration or higher speeds when moving diagonally. Having said that, filing down the third-party controllers in order to increase their range is very much a common thing because of this.