> Mruczek says he's worried that the handling of the Wiebe rec-ord has set a dangerous precedent that could set back the community to the '80s, when people would claim records that were impossible to achieve.
I'm not sure if this quote has aged like wine or milk. Possibly both.
For context, we now know the Billy Mitchell score at issue was almost certainly made on an inaccurate version of MAME. It has emulation artifacts, and no sound, which MAME couldn't emulate well back then. Twin Galaxies does accept emulation scores, but they have to be verified in a different way, and they go on a different leaderboard. The way Billy recorded it without the additional emulator verification would have made it trivially easy to play an input file and claim a just-slightly-higher score for pure ego reasons.
For legal reasons, I'm not saying Billy Mitchell actually did this, but I am saying he sued Twin Galaxies when they removed the score at issue.
There's a few other particularly suspect high scores that were purged from Twin Galaxies; notably Todd Rodgers' literally impossible time on 2600 Dragster. That score and Todd's explanation for it is now a speedrunning in-joke.
jmcgough · 4h ago
> Gardikis is famous for being one of only three people to achieve the so-called "Holy Grail" of gaming records: a perfect speed run on the original Nintendo Super Mario Bros., which means that he finished the game and saved the princess in 5 minutes and 8 seconds.
For those who are curious, there are now seven runners with 4m54s records. Andrewg had a long reign but isn't very active these days and has fallen to like 47th. Nifski is generally considered the strongest runner now.
WalterGR · 4h ago
I presume this is on original hardware?
toast0 · 3h ago
I think twin galaxies is firmly in the on hardware camp (although, the record setting in this example wasn't original hardware, the super gameboy isn't quite the same as an original gameboy).
For most records, there's three categories: tool assisted, emulated, and original hardware. TAS techniques are often hard to replicate with live human inputs, but emulator and original hardware records tend to converge but usually there's movement on the emulator record first.
ljf · 5h ago
A brilliant article - very much enjoyed reading that.
I'm not sure if this quote has aged like wine or milk. Possibly both.
For context, we now know the Billy Mitchell score at issue was almost certainly made on an inaccurate version of MAME. It has emulation artifacts, and no sound, which MAME couldn't emulate well back then. Twin Galaxies does accept emulation scores, but they have to be verified in a different way, and they go on a different leaderboard. The way Billy recorded it without the additional emulator verification would have made it trivially easy to play an input file and claim a just-slightly-higher score for pure ego reasons.
For legal reasons, I'm not saying Billy Mitchell actually did this, but I am saying he sued Twin Galaxies when they removed the score at issue.
There's a few other particularly suspect high scores that were purged from Twin Galaxies; notably Todd Rodgers' literally impossible time on 2600 Dragster. That score and Todd's explanation for it is now a speedrunning in-joke.
For those who are curious, there are now seven runners with 4m54s records. Andrewg had a long reign but isn't very active these days and has fallen to like 47th. Nifski is generally considered the strongest runner now.
For most records, there's three categories: tool assisted, emulated, and original hardware. TAS techniques are often hard to replicate with live human inputs, but emulator and original hardware records tend to converge but usually there's movement on the emulator record first.
Some of the previous discussion is here https://news.ycombinator.com/item?id=8142269