Comments
Georgia, USAnymx3 years ago

Thank you for looking into this PoppaPickett. As a moderator, I respect how you approached your decision. Even though we still have suspicions, it is only right to require solid evidence.

DCG_Grinder likes this
Georgia, USAnymx3 years ago

I agree with Razorflame. The 80 round is certainly a trigger to believe the 10 and 30 round runs may have been created in the same fashion.

DCG_Grinder likes this
Georgia, USAnymx3 years ago

Ooops. The 22 frames I was talking about? I mean 44. I'm so used to having 22 inputs, separated by a blank frame, to work with. Sorry.

DCG_Grinder likes this
Georgia, USAnymx3 years ago

One more thing, there is another point that I think is worthy to note. I'm seeing a lot of fast maneuverings for pieces being navigated through tight pathways. Runners should know how dangerous that is, as it can quickly backfire. There are numerous occasion where broken pieces are navigated in a risky fashion. I don't see other runners taking that chance.

From a TASing perspective, there are 12 frames before you loose control of a piece...when it is about to land permanently on other blocks. So you really don't have that much time, as a human, in order to make this kind of move.

Razorflame, DCG_Grinder and 2 others like this
Georgia, USAnymx3 years ago

Greetings, I was thinking of holding onto this information...so I guess I'll make a surprise announcement. I am in the process of TASing this game. I have been doing so since July 2019. It didn't take long before I realized that this game needs to be BOTed. Currently, I'm about 67.5% through with the game.

...And now the details. This submission is too unbelievable. Here are my points to why this seems to be a "Savestate" run.

  1. A 42 minute improvement? This reduction is not strictly a problem, since I believe a human could potentially get very good at this game...but I haven't seen anybody make that kind of progress that others can't follow suit. Considering this is a 2 hour run, that puts it closer to half cut...which I find impossible.
  2. I didn't recognize any deaths. Not sure if I missed one, but deaths are a HIGH possibility as you progress through, and something that is expected. The previous record had plenty.
  3. RNG, for the patterns displayed, is accomplished by the end of the last accepted input on the previous round. There is a 22 frame window after the explosion of the last "Flashing" block. After that, no input will change the pattern, except for the "Left" direction for a number of frame afterwards...but not after the success message for the upcoming bonus tally. As for the piece to solve the game...RNG can be altered in two ways...delay pressing "Start" on the round, or making a frame perfect pause when the first piece is about to be dropped. Having a different RNG value from the previous round can cause different cycling, but it can be enumerated through by the previously mentioned two methods. I bring this up, because a "Savestate" can be established shortly after completion of the round and hold that pattern. This is an advantage, giving foreknowledge to formulate a strategy.
  4. Possible patterns. This game cannot produce endless patterns for every RNG combination; however, I can say that my search for fast solvable patterns yields around 800 to 1000 possibilities (for levels above 30), with some of the flashing blocks being "Grid Locked". As for the other non-usable situations, there has to be 10,000+ patterns...with a small percentage of color re-arrangement. So, this means a large number of bad patterns do exist and are highly likely to end your run prematurely. My guess is, that RNG is determined by a 16 bit value, which would mean a possibility of 65536 patterns...which would go along with my experiences so far.
  5. Mistakes in the run. I have noticed a number of places where human mistakes were made. This doesn't mean that I believe the run is legitimate, but to point out that the game's difficulty is high enough that this kind of error should be seen a lot more than it was. As the moderated pointed out, if this was intended to be a TAS...it was poorly made because of these optimization errors...which I attribute to the culmination of best efforts for each round. So, this would make it more like a "Theory TAS"...meaning the possibility of a super lucky human run with minimal mistakes. When I was TASing Super Metroid, Theory TASes were very common and done to determine if the limitations of the game were being reached. Those times are very hard to get close to. Here, it certainly seems that it the case.

Anyway, I hope this supports my reasons why this run should be re-evaluated.

Razorflame, DCG_Grinder and 2 others like this
About nymx
Joined
3 years ago
Online
2 years ago
Runs
0