This time is not possible on this game. It says the verification method is referee but there is zero evidence this score ever occurred except for the word of the gamer. No TG referees ever witnessed this score and there is no video. But there is video proof in the form of a TAS that this time is not possible. http://tasvideos.org/5517S.html

That dispute form needs some work.

Well this is one that people have been waiting for. I'll be interested to see how this goes.

Much of the conversation surrounding this score has already happened on the Impossible Scores thread, beginning with a post from @Tompa on page 29: https://www.twingalaxies.com/showthre...reboard/page29
@Omnigamer - your considerable effort into the investigation of this score has been invaluable so far, and hopefully you can continue to be a big part of this discussion on this new thread...thank you to @D.B. Cooper for making the dispute official.

Didn't Todd say recently that he would be able to do this publicly if it was a sponsored event? It would take him over a thousand attempts.

Originally Posted by datagod

Didn't Todd say recently that he would be able to do this publicly if it was a sponsored event? It would take him over a thousand attempts.

Jace has stated that he is in the process of organizing the event and that Todd has agreed to participate.

Please remember folks that here is the only place that submitted evidence cannot be edited or deleted. Links to other sites or even internal links will not be permanent. We don't want to make Todd go through this again in another thirty five years so please gather everything here.

Thanks

Originally Posted by Dave Hawksett

Please remember folks that here is the only place that submitted evidence cannot be edited or deleted. Links to other sites or even internal links will not be permanent. We don't want to make Todd go through this again in another thirty five years so please gather everything here.

Thanks

I should have noted when I posted the Youtube video that it was for explanatory purposes only, I take no responsibility for any opinions expressed by the Youtuber, and it was not intended as "evidence" in this Dispute.

Sorry for any misunderstanding.

I applaud Twin Galaxies for its current position that score disputes, whether successful or not, are to be conducted transparently and permanently documented for historical posterity. Just as Twin Galaxies is criticized for continuing to recognize questionable scores, it has been equally problematic in the past that scores have been removed, possibly without legitimate reason, and certainly without transparency or proper documentation. For example, in the Robotron marathon score dispute, a newspaper in 1983 reported that the score being broken was 325 million. No such score appears in the database. We will probably never know when or why it was removed at this point. If Jeffrey Yee or Les Martin's Pac-man scores were ever recognized by Twin Galaxies, there should be a record of when and why they were removed. Same way with Steve Sanders' 3 million on Donkey Kong.

Wait a second...how did this soapbox get underneath me?

Crossposted Omnigamer's analysis from reddit.com/r/speedrun

Link: https://www.reddit.com/r/speedrun/co...2600_dragster/

Original Post in thread above:

"Some threads have popped up recently that touched on Todd Rogers's record for the Atari 2600 game Dragster.
As the story goes, Rogers claimed to achieve a time of 5.51 in Dragster on 1982. This surprised the developers, who had simulated a maximum fastest of 5.54. Supposedly he was flown out and performed it in front of them as well, and even once more later for TG for a total of 3 times in that year.


35 years later, nobody has come close to touching that time. Several people have achieved a 5.61, but that's nowhere close. Even TASes have only managed 5.57; still short of even the supposed simulated time of 5.54 by the developers.
This bugged the crap out of me. Partly because that's just plain suspicious, but also because I had written a section in my book about this event, and if it was faked/cheated/etc, then I don't want to perpetuate the myth further. So I decided to dig in and see what Dragster is really capable of.


So over the course of this weekend, I disassembled and analyzed every line of code for the game. It's fairly straightforward, but with some nuances in the growth of speed limits and how the frame counter affects time that were not easily understood. I took all of this and recreated the exact speed system in Dragster in a spreadsheet. You can check it out and try it yourself. The linked document is read-only to preserve it; please make a local copy for yourself to see what time you can get. I have confirmed that the values and inputs match up exactly to what happens during gameplay.


What I tested and learned through this process:

The best possible time for normal play seems to be 5.57. I believe the "simulated" best time from Activision likely had a flaw.
Getting a 5.57 is not possible in all cases; you are restricted by the frame rule cycles based on the frame you start.
The complete memory is wiped on every reset; there's no way to "dirty" the memory between attempts.
P1 and P2 are handled entirely separately on opposing frames; what P2 does will never affect P1 and vice versa. I also have not found any way whatsoever to generate lag.
P1 and P2 both go through the exact same code path; there is no difference between them.
The PAL version and an Asian version that I found have only very minor differences from the original US release, and nothing that affects gameplay values or otherwise. Purely technical stuff for different video standards.


Based on all this, I'm going to make this assertion: 5.51 is not achievable in normal play Even with some types of cheating, 5.51 is still very difficult to hit. You can try for yourself using the spreadsheet; try having it start in gear 2, or with an initial speed of 2.


It seems like there are multiple witnesses and otherwise for this particular record, but based on the code of the game I do not see how it's possible to hit 5.54, let alone 5.51. According to Rogers himself, he "starts" in 2nd gear. This might barely allow a 5.51 if true, but going through the game code I don't see any means to advance past 1st gear during the countdown. You will be flagged as "EARLY" no matter when you do it. However, an early start doesn't stop you from continuing motion, as a blow-out does. I have a theory that, if others did witness it, the Early flag was set and they were instead going off of the other player's timer when the dragster car reaches a stop at the end. This would allow speed to start before the timer actually increments. Starting in 2nd gear by itself would have no long-term benefit, and in fact only helps if you cut out the clutch transition period entirely.
So that's my take on this mystery. I do not think the record is legitimate if played according to normal rules. That said, 5.57 is definitely achievable, and 5.61 seems to be the best record with video evidence. It's just waiting for somebody to come along and snatch it.


I welcome all of you to fiddle around with the spreadsheet and see what kind of times you can come up with. If you hit 5.57, you can tell how well you did by looking at your total distance at that point. The best I've managed is a 97.26; see if you can get higher than that.


~Omnigamer
"