Submission Text Full Submission Page
This is a Tool-Assisted Speedrun of Puzzle De Pon! R! for the Neo Geo. It is intended to beat the game as fast as possible, while retaining an acceptable degree of entertainment.

Game objectives

  • Emulator used: FBA-RR v0.0.3.01 (alpha version) built on Jun 2 2009
  • Plays all the 39 levels
  • Gets to the "The End" screen as fast as possible
  • Some stylistic choices might be labeled as Speed/Entertainment Tradeoffs (see below)
  • Manipulates luck
  • Plays at hardest difficulty (level 8, set by entering Service Mode at the beginning of the movie)
  • Genre: Puzzle

Comments

Puzzle De Pon! R! is similar to Puzzle Bobble (Bust A Move) in many aspects. In fact, Taito is involved in both of those games. The difference here is that the goal is to free a shape from the bubbles; every three levels, it's a Zodiac sign. Since there are 39 levels and only 12 Zodiac signs, the 13th sign is the Solar Symbol; once that level is beaten, the player becomes the sun god. Luck manipulation plays a very important role in this TAS. The rerecord count of 9241 (in before someone screams an over 9000 joke) for a 8:47 movie is not a world record, I'm aware of that, but I hope it denotes how many attempts (up to 100 sometimes) I did before shooting every bubble in order to get a convenient bubble sequence, I tried my best. The RNG here is tricky. While it only depends on the current frame and not on the keys pressed, most of the times it just doesn't want to give you bubbles of a specific color unless you fulfill a specific condition. To "unlock" some colors, sometimes you need to make a certain number of shots, sometimes you have to delete a specific set of bubbles, sometimes you have to delete all the bubbles of a specific color. Sometimes it's just faster to waste a shot than to wait for the first natural occurrance of a color. As a general rule of thumb, when I couldn't get a color I needed, I attempted to delay the shot up to ~50÷60 frames; after that it would mean I'd have to stand still on the screen for more than a second in real time and I thought it would have affected the entertainment level. Call this Speed/Entertainment tradeoff, but it's not quite that. In the end it's just faster to shoot a couple of random bubbles most of the times. Except in a couple of levels where I obviously delay a shot for a long time in order to get a star or a bomb (they can't appear in all the levels though). In some levels, notably level 3 (Aries), I'm known to take a seemingly slower sequence. While that might be true, it helps me on a long term luck manipulation. I'm not blindly saying that all my choices are optimal, and I'm sure there might be other possible sequences. I tried to pursue alternate/unexpected combinations to keep the viewer entertained, as long as those sequences weren't too much time consuming, in fact, most levels are faster when done in a way that wasn't intended. The result is a speedrun which is tightly optimised and (I hope!) pleasant to watch. The last input is used to get a faster curtain fall on level 39. That's not necessary, the last input could have been the last shot since the curtain falls anyway. But, I only add a second or so to the movie file in this way, and, more important, I get to the "The End" screen ultimately faster. Since the "The End" screen appears before than the name entry for the high score table, a name isn't entered. That would require the movie file to be 2+ minutes longer because of the unskippable ending credits. Also there is no "game over" nor "congratulation" screen after you enter the name, so the "The End" screen is a reasonable place to use as a landmark.
This is by far my most serious attempt at a TAS. I worked up to 16 hours per day on this and I tried to do the best that I could. I put a lot of effort in this movie, but ultimately it was fun to make this. Which is why I uploaded it to my Youtube channel. Even if it won't be approved here, it'll have some spotlight anyway. I hope this time I can make it, though.

Other comments

  • Explain here things the audience would probably like to see: I love Puzzle games, do you?
  • Explain also things that could be improved in your movie: You can sacrifice more speed for more entertaininment. But those tradeoffs caused me troubles in the past, so this time I went for pure speed as much as I could. Say that this is 98% pure speed an 2% tradeoffs. Of course one may say that even 2% is too much, but I stand by my choices.


TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15629
Location: 127.0.0.1
This topic is for the purpose of discussing #2292: nineko's Arcade Puzzle De Pon! R! in 08:47.25
Joined: 8/27/2007
Posts: 4
Location: Pinellas Park, FL
I watched this run the whole way through, and I'm pretty impressed! It was a great deal of fun to watch, and from what I could see, the run was flawless! It tore through all the puzzles as quickly as I could imagine possible, given the nature of the game, and looked pretty damn cool when doing so. Great job, nineko! I'd definitely love to see this one published. You get a yes from me. :)
Experienced player (961)
Joined: 12/3/2008
Posts: 940
Location: Castle Keep
Voted meh, because the movie bored me to death, again a real bad game choice, but less terrible than previous ones, so ill abstain from voting no this time Edit: technicaly its probably good tought, hum, i dont want to see/know if theres errors or possible improvement, im not realy interrested, that game is realy as much un-entertaining as its possible to be Also, 9000 rerecord for 8min is nothing so "special", its small even...
Joined: 6/4/2009
Posts: 893
watched too, i think you missed a orange ball shot in the taurus puzzle exept this, the run look good even if it's not very entertaining
Skilled player (1444)
Joined: 7/15/2007
Posts: 1468
Location: Sweden
I think the two main problems with this movie is that it is about 6 minutes too long, and that there is no sense of increasing difficulty in the levels, so it gets very repetitive (probably not the only run on the site guilty of that though). Still, I did manage to watch through the whole movie, and I would at least call it a decent watch. The optimization looked ok as well. Weak yes vote.
Agare Bagare Kopparslagare
Joined: 10/3/2005
Posts: 1332
I liked Josh L./Superhappy's (canceled) Puzzle Bobble TAS from awhile back, and this looks to be a more condensed version, with a few extra twists. No obvious problems with execution, which the rerecord ratio corroborates. Looks like a solid submission. Yes vote.
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
Nicos wrote:
watched too, i think you missed a orange ball shot in the taurus puzzle
Wait, there are no orange balls in that level, only red, blue, and gray. Unless we're thinking about two different levels.
Gone.
Joined: 2/15/2005
Posts: 246
Location: Torquay, England
Yes vote from me. I watched this all the way through, and the only off thing I noticed was not linking up the orange balls immediately at ~5:54, but I'm sure nineko has a reason for doing that.
there is no sense of increasing difficulty in the levels
Not only could you probably say that about any TAS that you haven't played the game for yourself, but also, I thought there was some obvious sense of this, as later on the levels got pretty full up.
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
Chatlog 1 ikarikeiji: 5:04 ikarikeiji: why did you put the blues where you did? nineko84: One moment, let me see nineko84: Because red balls can't be manipulated before ikarikeiji: ohh nineko84: I explained that in the description, sometimes the rng prevents you from getting some colors unless you fulfill a certain condition ikarikeiji: game must have a complicated ball-generation algorithm nineko84: They're just locking out some colors nineko84: Unless you: -do N shots -delete all the bubbles of a color (2 shots in advance) -delete specific bubbles (2 shots in advance) nineko84: I had to wait 2 more shots in that level, so I thought to do a "one million shot" Chatlog 2 ikarikeiji: Also ikarikeiji: 5:54 ikarikeiji: why didn't you just link up the oranges straight away? nineko84: Let me see nineko84: Same deal nineko84: Two more shots required to get the green balls ikarikeiji: Ah right nineko84: The green balls can't appear until the blue balls are all deleted nineko84: And since the rng works on the next next ball, two shots of lag are needed
Gone.
Player (210)
Joined: 7/7/2006
Posts: 798
Location: US
Watched the youtube video, I gotta point out "Cancer Get" at 2:32. This appears well executed, though I think use of memory watch and analyzation of game mechanics would improve your orb selection process from what you describe. This doesn't really detract from the viewing though. The run is a bit repetitive, but not too long really. Is this a good version to TAS compared to other games in the Bust a Move series? Weak yes vote.
Joined: 6/4/2009
Posts: 893
nineko wrote:
Nicos wrote:
watched too, i think you missed a orange ball shot in the taurus puzzle
Wait, there are no orange balls in that level, only red, blue, and gray. Unless we're thinking about two different levels.
you right it's not taurus, it's @ 5:50 in your youtube video, you shoot 2 orange to the right then clear the blue, shoot another orange to the left. instead of shooting the orange in the middle to shut the gap and clear all the oranges, you shoot the right orange and shoot 2 orange again to clear the last one (that's a lot of orange Oo)
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
Ah, I see. It's the same thing asked by bobxp, I couldn't manipulate a green ball any sooner so I decided to have some fun with the orange balls to entertain the viewer. Unfortunately the rng isn't friendly sometimes. I'll double check it now in case I missed something, though. Stay tuned, I might edit this post later. edit: confirmed. I just spent ~200 rerecords on this, the green ball can't appear any sooner than that. It is actually possible to get it one shot sooner, but the delays introduced make the level slower in the end. Unfortunately all I can do there is to waste a few shots.
Gone.
Joined: 6/26/2007
Posts: 28
Location: Canada
Just wasn't interesting enough ..
Joined: 6/4/2009
Posts: 893
thanks.
Mitjitsu
He/Him
Banned User
Joined: 4/24/2006
Posts: 2997
arukAdo wrote:
Also, 9000 rerecord for 8min is nothing so "special", its small even...
After reading this I decided to compare this to the average rerecord rate you would expect for a movie of this length. After taking a trip to the movie statistics page and pulling out these statistics. Average movie length: 24:21 Average rerecord count: 27412 I converted the movie length into frames and divided it by the rerecord rate and got the magic number of 3.1978. I then divided it by the frame count of this movie and got 9892.738 (to avoid speculation of rounding bias). So it seems that 9241 is well below average. For it to be hard or an incredible amount of effort put in you would need to be 2.5 to 3x above the average. So you would need to take at least 25 to 30k rerecords to justify that statement.
Player (121)
Joined: 2/11/2007
Posts: 1522
Dromiceius wrote:
I liked Josh L./Superhappy's (canceled) Puzzle Bobble TAS from awhile back, and this looks to be a more condensed version, with a few extra twists. No obvious problems with execution, which the rerecord ratio corroborates. Looks like a solid submission. Yes vote.
Yeah, that.
I make a comic with no image files and you should read it. While there is a lower class, I am in it, and while there is a criminal element I am of it, and while there is a soul in prison, I am not free. -Eugene Debs
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
arkuAdo and Mitjitsu: I reworded a bit that passage in the submission text. I didn't mean that I wanted to extabilish a new world record with the rerecords, and sorry if I sounded presemptuous, I just wanted to say that I put some thoughts before each shot, as I spent up to 50 or even 100 rerecords for some of them since the rng was so tricky, not to mention the times I had to rollback to the beginning of the level to see if I could find a faster strategy. I had: a savestate at the end of the previous level, a savestate at the beginning of the current level, a savestate at the end of the previous move, a savestate with the current position in the current move. In the end I think the number of rerecords is appropriate for a movie of this kind.
Gone.
Joined: 10/3/2005
Posts: 1332
nineko wrote:
In the end I think the number of rerecords is appropriate for a movie of this kind.
It is. On the MovieStatistics page, the rerecord ratio is calculated as rerecords/movie-length-in-seconds. The ratio for this movie is about 17, and the average ratio Mitjitsu posted is about 19. Being neither a platformer with crazy zip physics nor an RPG requiring buttloads of manipulation, I don't see what the issue is, here.
Joined: 7/2/2007
Posts: 3960
Seemed enjoyable enough. It'd be nice if we knew more about how the RNG works, since there's a lot of apparently-suboptimal play because it won't give you the colors you need. As it is, we basically just have to take it on faith that you put the time in to find the best possible setups (which is, I think, what all the arguing over rerecord counts is for). One thing I noticed: Here it would've been slightly faster to move the orange balls down one row; they'd have less distance to fly. However, I don't think this is significant enough to block publication.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Joined: 5/2/2009
Posts: 656
I LOVE this game. But there's a problem about it: There's, like, a million games using that engine, with different tweaks each game. It's worse than Tetris. Yeah, I vote yes for this one, I like it and stuff. But how the site will react if someone try another version?
My first language is not English, so please excuse myself if I write something wrong. I'll do my best do write as cleary as I can, so cope with me here =) (ノಥ益ಥ)ノ
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
Derakon wrote:
Here it would've been slightly faster to move the orange balls down one row; they'd have less distance to fly.
Yes. In fact in many levels I try to place balls at the lowest possible row for this very reason. Sometimes, however, I gave priority to reduce the rotation of the cursor, which is mandatory for me as it's the worst time consumer. Good point though, maybe in this specific occasion I did a wrong choice, nice find.
Gone.
Joined: 9/29/2005
Posts: 46
Location: Denver, CO
Yes vote from me. Other people might complain about multiple same-type or same-game runs, but in my opinion, TASvids should grow to become an archive of EVERY game annihilated in the best possible or most entertaining way... regardless, there are a couple things I didn't understand in the run, but I'll chalk it up to unfamiliarity with the game and engine. Regardless, a very solid run that kept me entertained throughout.
"If any man is not free, then I too, am a small pie made of chicken"
Joined: 6/4/2009
Posts: 893
OmegaWatcher wrote:
I LOVE this game. But there's a problem about it: There's, like, a million games using that engine, with different tweaks each game. It's worse than Tetris. Yeah, I vote yes for this one, I like it and stuff. But how the site will react if someone try another version?
the same way we dealt with the genesis version of pitfall or the same way we treat hacks... we should accept just ONE "puzzle bobble" game type and reject the others exept if the gameplay realy change in comparaison to the original... in the same way, i think that this submition if published should be obeseleted by the "original puzzle bobble" and that the movie would be only judged on it's entertainement value (as it is incredibly hard to make the game entertaining for those who watch and not only those who play)
Joined: 12/1/2006
Posts: 26
Yes vote. This is f*** awesome.
Former player
Joined: 3/23/2004
Posts: 95
Yes Vote. I love this game and the video was pretty amazing. Only thing I noticed...it seemed to lag at some parts on your input (like you'd shoot a ball...and then wait a second to shoot the next one when you were going to fire it in a different direction anways). Was this due to the game or you?