There's been a few headlines over the weekend about Todd Roger's score of 5.51 in the Atari 2600 game Dragster being declared impossible and the score wiped from the scoreboards, while I hadn't heard of any of this it's very significant as it's the longest standing gaming record that's not been beaten which is listed in the Guinness book of world records and Todd Rogers is seen as one of the early pro gamers. The fact alone the record hadn't even been tied all this time was suspicious on its own and last year another Speedrunner who was unable to match the time, looked into the code and found a 5.51 time was impossible with 5.57 being the fastest possible time which he replicated with a tool assisted run:
This kicked off a massive 271 page on the TwinGalaxies forum to dispute the score:
http://www.twingalaxies.com/showthread. ... core-05-51
What's interesting about it is how can a score be disproved? Even though the evidence from the code looked convincing, it wasn't accepted because Todd could have used some sort of glitch or technique to be faster even though he had detailed the supposed steps he did which are not possible. Eventually a second Atari specialist was brought in as around a decade ago he'd looked into it and believed a faster time was possible but when he examined his code again, he found he'd made two errors and indeed 5.57 was the fastest possible time. That wasn't enough though with the calls coming in to use actual hardware so someone set up an automated input to an actual console running the cartridge to effectively do a tool assisted run on the physical console but that still only managed 5.51.
The TG admin were claiming that the only way to disprove the record would be to attempt every possible control sequence into physical hardware to prove there was no way to be faster, meanwhile Todd Rogers was going to do a live event where he would perform the 5.51 again and he was in training and had already done a 5.54 but no evidence for that despite the fact it would disprove those claiming 5.57 was optimal. Ben Heck was brought in to produce a custom interface to the ram on an actual 2600 and working with Todd, could verify whether a faster time was possible. Ben eventually agreed that it's likely the 5.57 time was the fastest possible and didn't pursue it further. Surprisingly people were still defending Todd and were now claiming that now people should stress test the game repeatedly and try successive runs with at least 3000 up to 20,000 even though the code is simple and the game wipes all variables on each run.
TwinGalaxies had claimed they were still looking into it and would announce the result on March but they then announced they were removing the Dragster score and every other one.
What was odd though that the reason the thread was a monster 271 pages was because Todd was such a legendary figure but it turns out that at the time, there were numerous other disputes with him and with scores that were much more blatantly fake because they either were the wrong multiple or the game had a maximum score that the fake score was higher than:
There were still people defending him though with some truly hilarious excuses, one was because the score was misread due to a coffee stain, one was due to a typo entering into the database, some were because he was on a prototype cartridge and some were on a variant cartridge (which there was no evidence existed).
Discuss PC and console gaming
1 post • Page 1 of 1
Phalanx - liquid cooled i7 3930K @ 4.2Ghz, 32GB ram, Nvidia GTX 1070 8GB, 2TB SSD/28TB HDD, dual blu-ray, Dell U2711/HTC Vive
1 post • Page 1 of 1
Who is online
Users browsing this forum: No registered users and 1 guest