TASVideos

Tool-assisted game movies
When human skills are just not enough

Submission #1693: Baxter's NES Arkanoid in 12:26.8

Console: Nintendo Entertainment System
Game name: Arkanoid
Game version: USA
ROM filename: Arkanoid (U).nes
Branch:
Emulator: (unknown)
Movie length: 12:26.8
FrameCount: 44808
Re-record count: 79117
Author's real name: Lennart W.
Author's nickname: Baxter
Submitter: Baxter
Submitted at: 2007-09-11 15:29:21
Text last edited at: 2011-07-15 19:55:05
Text last edited by: Zeupar
Download: Download (3601 bytes)
Status: published
Click to view the actual publication
Submission instructions
Discuss this submission (also rating / voting)
List all submissions by this submitter
List pages on this site that refer to this submission
View submission text history
Back to the submission list
Author's comments and explanations:

Goals

In contrary to the "Warped" Arkanoid run, this run does attempt to be an improvement to Genisto's full Arkanoid movie. The goals are not exactly the same though. The warped Arkanoid run's only aim is to complete the game as fast as possible. Genisto's movie aims mainly for entertainment. If you look at his run, you'd say the goals of his run are using no warps, and losing no balls (and of course doing this as fast as possible). It is however clear that many of his levels can be improved by using the Laser powerup. The Laser powerup was however only used in three levels, since clearing levels when juggling aroud three balls looks a lot cooler. When using the Laser powerup in a TAS, it doesn't look very different than using it when you play at 100% speed. It would be easy to 'improve' Genisto's movie by just using the Laser powerup in a lot of levels. This would however take away nearly everything that made the previous movie cool (in fact, I even doubt whether such a movie would be accepted to obsolete Genisto's movie). Luke, who had a wip which completed about 13 levels also realized this, and only used the Laser in the levels where Genisto had also used the Laser. This is however a completely arbitrary restriction. A better restiction is not using the Laser at all, and only using the 3-ball powerup. This is a clear, logical goal, which will keep the aspect that made Genisto's movie entertaining. When I proposed this at the forum, only DrJones disagreed with this (although I have to admit, there was not a lot of response). The goal of not losing any balls was also preferred... so the goal of this movie is to complete Arkanoid as fast as possible, only using the 3-ball powerup, and not losing any balls (and obviously not using the level-skip code).

This goal is a more stricter goal than Genisto's, since he using Laser three times. In two of the three levels where he used laser, he is faster than me... on one I am faster with using 3 balls (although this doesn't mean that using laser couldn't have been faster here). In every other level however, I saved time, and sometimes quite a lot. In the end, this movie turned out 14620 frames faster, or nearly 4 minutes and 4 seconds faster than Genisto's movie.

  • Aims for fastest time
  • Uses the 3-ball powerup only
  • Uses no warps
  • Manipulates luck
  • Genre: Action

TASing Arkanoid

TASing Arkanoid is quite different from TASing any other game. There are several things that would make it seem like it's an easy game to TAS. There is no lag. You only have to use the left and right buttons, and the A button once when shooting the ball at the start of the level. Sometimes you don't have to do anything for quite some time, if all balls are at the top of the screen. Then there are also the parts in between levels where no input whatsoever is needed (and since there are 36 levels, there are 35 of these 'cutscenes'. Even with all of this, the game is extremely difficult to TAS.

The main reason for its difficulty is the amount of options you have. You have over 100 different positions at the start where the ball can be fired from. You obviously want to get the 3-ball powerup somewhat quickly, which decreases the options by a huge amount... but there are always still quite a few good options left from where to fire the ball, and firing it one frame more the to right or left will create a completely different game.

If there are enemies at the level (nearly all levels have enemies), the moment at which you fire from this position will also affect the game, since the enemy will have a different location if you wait a little to fire.

Each time a ball hits the paddle (and this happens quite a lot, considering there are 3 balls in the game), you have 6 basic angles in which you can shoot the ball (depending on what position of the paddle it lands). If you 'run into' the ball from the side though, it will have one of the six angles, but will be shot from a slightly different location, which also produces something different. This means that there are over 6 options each time you shoot the ball.

Then the spot where the enemies appear can be influenced. If you are at the left of the screen, they will appear at the left, if you are at the right of the screen, they will appear at the right.

You can look at this as if there are many possibly roads, that all lead to a certain completion time, and you have to find the fastest. When you shoot the ball, you will be at a large intersection with many possible routes. Each time you bounce off a ball, you will come at an intersection with over 6 routes to choose from. Each time an enemy appears, there is another intersection with 2 options. In the end, you will end up with an extreme amount of options, which makes it pretty hard.

Since you can obviously not test all options, some of them must be discarted. You can often see that bouncing it off at a certain angle will make you unable to keep all three balls in the game. If it's impossible to save a ball, this option can be ignored right away. Then there are also angles which will just send the ball in a pointless direction. Mostly these angles can be ignored also for that bounce, but you must be careful with this. Making a useless bounce might enable you to shoot from a really good position the second time it comes back, so that must always be taken into account. You must notice that each ball you bounce will affect all other balls. A ball literally changes the level, and if one ball makes a good sequence, you must avoid another ball messing that up. Just like messing it up for another ball, a ball can also improve the route for the other ball. Each new choice that is made will affect all previous choices that were made.

It is obviously good to clear a lot of blocks very fast (:P), and this can be done most of the time by getting a ball 'trapped'. It is convenient if a level has a small opening somewhere, and if a level doesn't, then it's good to focus on one particular spot, in order to form a cavity in the level, where a ball can be trapped in.

This obviously hard to do, since, even though there are an extreme amount of possibilities, you can still 'only' bounce the ball in 6 directions. While this creates lots of possibilities it also limits you a lot to certain angles. You can't shoot a ball exactly where you want. At the start, you can decide the position where you shoot from, but later on, and aspecially at the end it becomes quite appearant. At the start, shooting in any direction might clear something, but at the end, there aren't that much targets left. With only a few angles to choose from, it can easily happen that you are just unable to hit the last block, even though you cleared the rest of the level very fast. This is sometimes where Genisto's movie suffered. In this movie, I tried to plan as much ahead as possible, and make sure to in the end clear ALL blocks as fast as possible. This might result in a few strange shots at the end, but I think that in most of the levels, the ending comes pretty fluently.

I want to add that there are a total of 36 levels, which is a LOT. I don't know if I would have started the run (about 2 years ago) if I knew how tedious it was. I often took very long breaks from it, but I'm really glad it's finished now.

Levels

This table will list at what time the screen turned black after each level. It are rough estimates, and the real number can be a second higher of lower. It should still give a good image of at what levels the time was saved.

Level Genisto's time My time Level difference Total difference
01 00:55 00:31 -00:24 -00:24
02 01:31 00:54 -00:13 -00:37
03 01:56 01:10 -00:09 -00:46
04 02:23 01:29 -00:08 -00:54
05 02:47 01:56 +00:03* -00:51
06 03:09 02:11 -00:08 -00:59
07 03:39 02:30 -00:10 -01:09
08 03:50 02:39 -00:02 -01:11
09 04:10 02:54 -00:05 -01:16
10 04:30 03:12 -00:02 -01:18
11 05:21 03:46 -00:17 -01:35
12 05:38 04:02 -00:01 -01:36
13 05:53 04:15 -00:02 -01:38
14 06:21 04:34 -00:09 -01:47
15 06:58 05:08 -00:03* -01:50
16 07:27 05:24 -00:13 -02:03
17 08:03 05:43 -00:17 -02:20
18 08:24 05:59 -00:05 -02:25
19 08:42 06:16 -00:01 -02:26
20 08:59 06:30 -00:03 -02:29
21 09:16 06:42 -00:05 -02:34
22 09:36 06:58 -00:04 -02:38
23 10:06 07:25 -00:03 -02:41
24 10:30 07:46 -00:03 -02:44
25 11:04 08:11 -00:09 -02:53
26 11:38 08:28 -00:17 -03:10
27 12:32 09:00 -00:22 -03:32
28 12:55 09:20 -00:03 -03:35
29 13:18 09:52 +00:09* -03:26
30 13:57 10:21 -00:10 -03:36
31 14:28 10:44 -00:08 -03:44
32 14:52 11:02 -00:06 -03:50
33 15:27 11:31 -00:06 -03:56
34 15:52 11:52 -00:04 -04:00
35 16:15 12:11 -00:04 -04:04
36 16:30 12:27 00:00 -04:04**

* At these levels, Genisto used the laser ** You might notice that the difference between 16:30 and 12:27 isn't 04:04 but 04:03, but Genisto's time was rounded down, and my time rounded up, so the difference is closest to 04:04.

Some comments on individual levels:

Level 01: I saved the more time on this particular level. The main reason for it being faster is that I focussed on getting to the top of the level at the start, and I focussed on getting all blocks at the end, without having to do bounces that weren't needed. I saved time at lots of other levels by doing these things.

Level 03: I am extremely pleased with how this level turned out. Every ball has its own purpose the entire time, and it just seems like this level was made to be cleared like this.

Level 05: Genisto used the laser on this level. I ended up only being 3 second slower, but the cool thing about not using laser here is that at the end (since the level is pretty long), the balls move pretty fast :). It was not really possible to create a big cavity in the level.

Level 07: It was possible to create a cavity in this level pretty good. That's why everything is cleared as fast all of a sudden.

Level 09: A very hard level, both because of the blocks at the top, which can't be cleared fast, and the blocks in the middle, where no cavity can be formed.

Level 11: This is an interesting level, since there is no non-grey block. This means no powerups can be collected, so the level needs to be cleared with one ball. I think this level turned out VERY well.

Level 14: This level is cool, since the enemies were used at the end.

Level 15: Genisto used laser, but in the end, this version turned out 3 seconds faster. The level is cool, siunce because of its length, the balls travel really fast at the end.

Level 23: Again a very long level. At the end, the score counter keeps counting a little. This happened since a few grey blocks were cleared at the end, which give many points.

Level 26: This level is the prime example of using enemies. I wait a little till the right frame to shoot the ball. After shooting it, I never touch it anymore. The enemies keep the ball at the top of the screen the entire time!

Level 27: This level was perfect for trapping balls in the center. The blocks are cleared so extremely fast, (not that the grey blocks in this level need to be hit 5 times), it's hard to even understand. The level changes all of a sudden, and during the course of the entire level the score counter is running like crazy. Evem after all blocks are cleared, it keeps running for some time.

Level 29: Genisto used laser at this level and gained 9 second on me. This is due to the fact that the level is perfect for using laser. There is nearly no spot to create a cavity, and the blocks at the top are annoying too. Even if a ball clears a lot at the top, or even everything, you want your ball back if it's done, but sometime it remains at the top, leaving you with 2 balls. I tried to avoid these things obviously, and dealt with the level as good as I could.

Level 33: It's somehow a hard level, since the grey blocks need to be hit a lot, and they always remain at the end. It's hard to clear a room if only a few blocks, that need to be hit often, remain.

Level 36: To be different than the warped version, I went for input time. This meant a similar strategy as Genisto. When you can stop the input is determened by when the boss fires its last shot. After you avoid being hit by that shot, input is ended.

Ending

I ended the movie at the first frame ending input was possible while still completing the movie. This is different from my warped movie of this game, which aims for the fastest AVI time. I chose going for input time now for three reasons. Genisto also did this, it looks cooler, and it shows something different than the warped version. It also suits this kind of run better probably.

I'd like to thank Genisto for his previous run, and also everybody who showed interest in the run, it really helped me get motivation to continue working on it.

Baxter


NesVideoAgent: Hi! I am a robot. I took a few screenshots of this movie and placed them here. Here goes! Feel free to clean up the list.

Truncated: Accepting. I think the goals of this movie are well thought-out and that this should accept our current all-level famtasia movie.


Bisqwit: ProcessiƋ...

Similar submissions (by title and categories where applicable):