Todd Rogers banned from Twin Galaxies and records removed

[MaximRecoil]

All of this because some tired person working at an office desk in the early 80s; looked at a blurry photo of a CRT TV screen and misread a number as 5.51, wrote it down, sent off a badge and then just moved on with their lives and forgot about it.

 

Three times you mean. Two other people were credited with a 5.51 in Activision's newsletter before Todd Rogers was.

[accousticguitar]

 

Three times you mean. Two other people were credited with a 5.51 in Activision's newsletter before Todd Rogers was.

Three different people getting the same score would seem to indicate that the score is possible.

[Bryan]

Three different people getting the same score would seem to indicate that the score is possible.

Or that Activision was easily fooled. Remember that they then removed the other scores and listed only Todd's and we don't know why. If I were going to cheat, I'd just use a home computer (the A8 has the same resolutions and colors available) and draw whatever game screen I wanted. Dragster isn't very colorful, after all.

[Keatah]

That.

[MaximRecoil]

 

I was so awesome too!

 

Your non sequitur is dismissed.

 

Or that Activision was easily fooled. Remember that they then removed the other scores and listed only Todd's and we don't know why. If I were going to cheat, I'd just use a home computer (the A8 has the same resolutions and colors available) and draw whatever game screen I wanted. Dragster isn't very colorful, after all.

 

Cheating on the part of those other two guys is a rather far-fetched theory. See post #138 and #140.

[Bryan]

Cheating on the part of those other two guys is a rather far-fetched theory. See post #138 and #140.

 

I believe something isn't right, though. No one has been able to prove 5.51 possible, and seeing your name in print is a compelling reason to fake a photo. It's also possible that there was a clerical error at Activision which has never been 'fessed up to. Has anyone ever produced a 5.51 certificate? The one Todd produced was shown (TG forum) to be a modified version of one of his others with a certificate number that is out of sequence with other known certificates.

 

EDIT: Here's what I think probably happened. In any large contest there are cheaters. It makes total sense that Activision would get faked photos and someone had to sort them out. David Crane has stated that they did some internal tests to try to determine the best possible time. They determined it was 5.54, one tick off of the 5.57 given by code analysis. Why did they do this? Probably in response to some sketchy photos. They got 3 showing 5.51 which was a possible increment (I'm sure they probably got some with illogical times too), and after Crane's analysis they probably took Dragster off the charts completely.

 

Today, David stands by the Todd's time but it's probably more to avoid being in the middle of a sh!t storm over a 35 year old game.

[NE146]

What the.. swear to god I just did this!

[Gabriel]

What the.. swear to god I just did this!

IMAGE.png

 

Looks legit.

[MaximRecoil]

 

I believe something isn't right, though. No one has been able to prove 5.51 possible, and seeing your name in print is a compelling reason to fake a photo. It's also possible that there was a clerical error at Activision which has never been 'fessed up to. Has anyone ever produced a 5.51 certificate? The one Todd produced was shown (TG forum) to be a modified version of one of his others with a certificate number that is out of sequence with other known certificates.

 

I don't know what actually happened. I think it's a very strange case, and I don't find any of the theories explaining those other two 5.51 scores, which were earlier than Todd's, compelling at all. Two guys from two different states both happened to decide to fabricate a picture of a 5.51 at the same time, and both of them did it well enough to dupe Activision? Then in the next issue, another guy from yet another state does the same thing?

 

Also, why did those first two guys both happen to choose 5.51 as the time to fabricate? The progression of the records published in the newsletter had been 5.61 and then 5.57, so if you're going to fabricate a time, why 5.51? For that matter, how did either of them even know that 5.51 was a time that could actually be displayed? Omnigamer examined the code so he knows which times can be displayed and 5.51 (as well as 5.54) can be. The times that can be displayed are in increments of something like 0.0337 seconds (I can't remember if that's the exact number or not, but it was somewhere around that), but without knowing the inner workings of the game, you can only observe that the increments are either 0.03 or 0.04 seconds. If you saw that the record had been 5.61 and then 5.57 (a difference of 0.04 seconds), wouldn't you expect the next times to be 5.53 and then 5.49? Of course, that would be wrong; the next two displayable times are 5.54 and 5.51, and then it goes back to a 0.04 second increment for the next displayable time (5.47). Whether it is a 0.03 or 0.04 second increment depends on how the more precise internal increment gets rounded for the displayed time.

 

One clerical error is plausible; the same clerical error happening three times in two different issues isn't so plausible.

 

I can at least see a motive for Rogers to cheat. He, along with some other people, had been credited with the record at 5.57, then he opened the latest issue and saw that two guys got a 5.51, but I don't know of a motive for those other two guys, especially since the one guy that's been found and asked about it didn't even seem to care about it, didn't remember much about it, and was even under the mistaken impression that faster times than his had been achieved.

 

And yes, Todd Rogers' image of a supposed 5.51 certificate was clearly forged; it is the exact same image as a certificate that has a slower time, slightly modified.

Edited February 9, 2018 by MaximRecoil

[Gabriel]

 

Also, why did those first two guys both happen to choose 5.51 as the time to fabricate? The progression of the records published in the newsletter had been 5.61 and then 5.57, so if you're going to fabricate a time, why 5.51? For that matter, how did either of them even know that 5.51 was a time that could actually be displayed? Omnigamer examined the code so he knows which times can be displayed and 5.51 (as well as 5.54) can be. The times that can be displayed are in increments of something like 0.0337 seconds (I can't remember if that's the exact number or not, but it was somewhere around that), but without knowing the inner workings of the game, you can only observe that the increments are either 0.03 or 0.04 seconds. If you saw that the record had been 5.61 and then 5.57 (a difference of 0.04 seconds), wouldn't you expect the next times to be 5.53 and then 5.49? Of course, that would be wrong; the next two displayable times are 5.54 and 5.51, and then it goes back to a 0.04 second increment for the next displayable time (5.47). Whether it is a 0.03 or 0.04 second increment depends on how the more precise internal increment gets rounded for the displayed time.

 

All times end with a 1, 4, or 7. This is easily observable. There's no need to work out increments.

[+Nathan Strum]

All times end with a 1, 4, or 7. This is easily observable. There's no need to work out increments.

 

Not true.

 

[Shawn]

All of this because some tired person working at an office desk in the early 80s; looked at a blurry photo of a CRT TV screen and misread a number as 5.51, wrote it down, sent off a badge and then just moved on with their lives and forgot about it.

 

I've thought along the same lines. They read 5.57 as 5.51 or made a mistype in some manor and then it snowballed from there. The player(s) liked the credit for the better score and went along with it out of ego. 30 years goes by and that ego boost comes back calling looking to be paid for in full. Ying and then Yang.

[Bryan]

Also, why did those first two guys both happen to choose 5.51 as the time to fabricate? The progression of the records published in the newsletter had been 5.61 and then 5.57, so if you're going to fabricate a time, why 5.51?

 

You're assuming these were the only fabricated results they ever got. They probably knew how the scoring went so they could throw away impossible scores. From what was left, there were a few 5.51's that looked like they might be legit, so they published them and Crane went back to the code to try and calculate a perfect score.

[Gabriel]

 

Not true.

 

Dragster (1980) (Activision).png

 

Point taken. How about all times in the time range in question end with 1, 4, or 7?

[Nukey Shay]

Put Stella on a low framerate and count them yourself. BITD they could have been observed on a slow mo video recording.

[MaximRecoil]

 

You're assuming these were the only fabricated results they ever got. They probably knew how the scoring went so they could throw away impossible scores. From what was left, there were a few 5.51's that looked like they might be legit, so they published them and Crane went back to the code to try and calculate a perfect score.

 

The idea that it was common for people to fabricate scores is even more far-fetched. Neither of the methods (doctoring a photograph or creating the digital image from scratch to display on a TV) were easy in the early '80s. Suppose you had a home computer with RF output, you would also need a paint program for it as well drawing ability, or programming skills plus drawing ability if you want to generate the raster via writing graphics code.

 

And what are those other two guys doing today? Playing the long con? To what end? When Omnigamer talked to one of them recently, why didn't he 'fess up? What does he have to lose? He's not like Todd Rogers who made a career out of it. If he faked a score, that was when he was a kid ~35 years ago.

 

And I suppose no one thought to fabricate Dragster scores prior to the first issue that published those two 5.51s, and I guess 5.51 was satisfactory for every score fabricator and would-be score fabricator out there, because, no 5.47, 5.44, etc. Not even in the subsequent ~35 years has there been anyone claiming anything lower than 5.57, except for Todd here on AtariAge claiming a 5.54 in 2002 for which he posted a screen shot. I guess that national wave of motivation to fake Dragster scores was just a brief one in 1982.

[Cobra Kai]

I think it's plausible to believe that in the 1980's, a picture of a tube tv with a Dragster screen displaying a score of 5.57, could easily be blurry enough for someone to think it's 5.51. That's what I think. We're talking about low resolution, and probably a poorly lit room, and a small tv. A 'big' TV back then was 27". No cheating necessary. Is the original picture online to look at? probably not.

[Bryan]

The idea that it was common for people to fabricate scores is even more far-fetched. Neither of the methods (doctoring a photograph or creating the digital image from scratch to display on a TV) were easy in the early '80s. Suppose you had a home computer with RF output, you would also need a paint program for it as well drawing ability, or programming skills plus drawing ability if you want to generate the raster via writing graphics code.

Never said it was common, I'm just saying that I bet they got some laughable fakes and a few good ones. I had an 800 at that time. I could have made anyone a perfect looking Dragster photo.

[MaximRecoil]

I had an 800 at that time. I could have made anyone a perfect looking Dragster photo.

 

What would you have used for reference? It isn't as if you could have just grabbed a raw pixel-dump from Stella. Looking at the Atari 2600's RF output on your TV screen wouldn't have been a very good reference. You may have gotten close, maybe close enough to fool Activision, but I highly doubt you could have gotten it pixel-perfect.

[Bryan]

What would you have used for reference? It isn't as if you could have just grabbed a raw pixel-dump from Stella. Looking at the Atari 2600's RF output on your TV screen wouldn't have been a very good reference. You may have gotten close, maybe close enough to fool Activision, but I highly doubt you could have gotten it pixel-perfect.

We're talking 160 resolution max and some clever kids trying to cheat. Fire up a paint program and put a ruler on the screen. Flip back and forth until you can't tell which is which.

[Andromeda Stardust]

All of this because some tired person working at an office desk in the early 80s; looked at a blurry photo of a CRT TV screen and misread a number as 5.51, wrote it down, sent off a badge and then just moved on with their lives and forgot about it.

He deliberately smudged the screen to make that 7 look like a 1. Assuming the "7" was one scanline in height against a dark background, it would have been easy to do. And the poorly exposed polaroid is just low tech enough to hide the flaws. Really shameful. He could have had the fame of being the first recorded instance of a legit perfect game (5.57), but he had to cheat instead.

[MaximRecoil]

We're talking 160 resolution max and some clever kids trying to cheat. Fire up a paint program and put a ruler on the screen. Flip back and forth until you can't tell which is which.

 

Again, you may have gotten close, maybe close enough to fool Activision, but I highly doubt you could have gotten it pixel-perfect. Also, give me an example of a paint program you would use for this, and what image format did it save in? When you opened the saved file, what program opened it, and did it open in "full-screen" with no element of the program that opened it being visible? I assume it would at least have to be saved and then opened, because while working on it in the paint program it seems that there would need to be some element of the program visible, else how would you select drawing tools and colors? And to save you would need to have a separate storage device (e.g., a floppy drive), because the Atari 8-bit computers didn't include one by default, and neither did other affordable home computers of the time, such as the Commodore 64, TI-99, VIC-20, and so on (not that I know of anyway).

 

I don't see this as a likely scenario for even one person to have done, let alone three. Home computers were less common than Atari 2600s in '82, and many, if not most, of them that were out there were used only for video games; paint programs weren't exactly "killer apps" for them. Storage devices were even less common than the computers themselves, and I figure that kids who were hell bent on fabricating a Dragster score were rather uncommon too, as were kids who were computer savvy. So you have a lot of uncommons that had to come together three independent times in the span of time between the fall and winter of 1982.

 

He deliberately smudged the screen to make that 7 look like a 1. Assuming the "7" was one scanline in height against a dark background, it would have been easy to do. And the poorly exposed polaroid is just low tech enough to hide the flaws. Really shameful. He could have had the fame of being the first recorded instance of a legit perfect game (5.57), but he had to cheat instead.

 

Dragster's 1 and 7 look nothing alike:

 

 

Edited February 10, 2018 by MaximRecoil

[+Random Terrain]

I had some kind of paint program on my Commodore 64 in the late 1980s that would let you see your "art" full screen. Don't know how far back in time when that feature became available on computer paint programs, but it seems like a necessary one that should have been there from the start. I'm sleepy, so I hope those sentences make sense.

[Bryan]

 

Again, you may have gotten close, maybe close enough to fool Activision, but I highly doubt you could have gotten it pixel-perfect. Also, give me an example of a paint program you would use for this, and what image format did it save in? When you opened the saved file, what program opened it, and did it open in "full-screen" with no element of the program that opened it being visible? I assume it would at least have to be saved and then opened, because while working on it in the paint program it seems that there would need to be some element of the program visible, else how would you select drawing tools and colors? And to save you would need to have a separate storage device (e.g., a floppy drive), because the Atari 8-bit computers didn't include one by default, and neither did other affordable home computers of the time, such as the Commodore 64, TI-99, VIC-20, and so on (not that I know of anyway).

 

I don't see this as a likely scenario for even one person to have done, let alone three. Home computers were less common than Atari 2600s in '82, and many, if not most, of them that were out there were used only for video games; paint programs weren't exactly "killer apps" for them. Storage devices were even less common than the computers themselves, and I figure that kids who were hell bent on fabricating a Dragster score were rather uncommon too, as were kids who were computer savvy. So you have a lot of uncommons that had to come together three independent times in the span of time between the fall and winter of 1982.

I'm not going to address all of this other than to say I had a personal computer then and I had friends who did too. Most of us had lots of copied software and getting our hands on an art program was no big deal. I knew even more people with 2600's. We could have easily pulled this off. You keep saying it's too uncommon, but I never said Activision was getting a fake photo from every kid in America. I was just saying it would have been possible for me and some others I knew, and I also knew kids who were always lying/bragging who would have done it if they could. It was never meant to be more than a theory.

[CPUWIZ]

Fist fight!