How to Study Shmups, Dissect SuperPlays, and Create Efficient Routes

A Guide to High-Level Play by Juju Kenobi


Edited and Localized by Mark MSX

Foreword by Mark

After collaborating with Juju on his Shmup Spotlight article for the website and digesting many aspects of the article to try and incorporate into my own play (especially in light of my continual frustration with the Dodonpachi 2-all), I found myself wanting to know more about the specifics of how he goes about learning a game as strict as Dodonpachi DaiOuJou. Being the pushy guy that I am, I asked if Juju would be interested in doing a follow up article about the specifics of his learning methods and, being the homie that he is, Juju agreed and followed up with this article. I really do think it is very valuable advice for players wanting to push themselves further, learn faster, and play more efficiently.



To begin, I would like to say that this is not a practice regimen that I insist everyone has to follow, but since Mark asked for details, I will try to describe, as best as I can, how I have enjoyed playing shmups during the last year.

This write-up will mostly cover my learning/playing of Dodonpachi DaiOuJou Black Label and a little bit of me learning Ketsui. These are examples of games that I played with scoring in mind. With that said, by no means, do I think everybody should play and enjoy shmups the same way that I do. I’ll just try to present an honest and accurate description of my way of doing things.

You may want to start scoring a shmup for different reasons. You may really like a shmup and want to learn more about it, it may be a (friendly) competition, or maybe you just like playing for score.

I won’t spend much time explaining why playing a shmup for score is really interesting. Prometheus has already written a great deal about this in his guide “Full Extent of the Jam.”

*Editor’s note: Full Extent of the Jam was a huge influence on how I got into the genre, so it is absolutely a recommended reading, please do check it out. It’ll be linked at the bottom of the article. I’ll also link my interview with Prometheus about this guide as well.*

I will just say that in some games, ignoring the score will make you miss a good part of what the game has to offer. For example, chaining in Dodonpachi DaiOuJou really gives the game a great rhythm and flow, and so I never find the first stages boring or annoying because of the scoring system.

Before diving into scoring, I usually play the game quite a bit. I try to at least get a good feel for how it plays (I won’t spend hundreds of hours on a game I don’t like ^^). This way, I know which ship I prefer, how the enemies behave, how the bosses act, and get a general impression of the different patterns.


Part 1: How to Start

First, I try to gather as much information as I can about the game, I’m quite the studious person. I don’t mind reading a lot before diving into the game. If the game is fascinating to me, I enjoy every second of this process. This may seem like homework to some people, but if you want to have a really good understanding of what is happening in the superplay videos, this is quite necessary. And even after that, you will still be amazed at how much you can learn.

For Dodonpachi DaiOuJou, Pazzy’s annotated videos were a true blessing. Other than that, I watched the Prometheus Special Demonstration video on Icarus’s channel and the STG weekly episode about DOJ.

*Note: all of these videos can be found on the DOJ BL & WL video index pages, which are linked at the bottom of the article.*

I then asked Blackisto (a strong Dodonpachi player) about shmupmame, as I saw him using it on his stream. I wanted to know how to use save states properly.  I decided to play Black Label because it seemed more fun according to Prometheus’s posts on and Feedbacker was playing this version so I hoped he could answer my questions. Then I took several famous replays: WY 3.58bil, LYX 3.2bil, both of Pazzy’s WL annotated replays, Fufufu’s world record on White Label, the Prometheus Special Demonstration, and VioletHatPurple’s 1.6 billion run. My idea was to learn the best strategies, if possible, but when I couldn’t understand or do it (because I’m not good enough), I could try other alternative strategies.

I have been playing the piano since I was six, and from that experience I learned a way of doing things that I enjoy reproducing with shmups. When you want to learn a new piece of music, you first learn how to play each hand alone, determining what finger to put on each key. Next, you memorize the partition for each hand alone, then you try to put the two hands together, small section after small section. Then, every time after learning a small section, you try to play everything you have learned until you finally learn the whole piece. And of course, many times during my learning process, I also listen to the piece interpreted by better musicians. Sometimes, when I can’t figure out which finger to use, I try to watch what other people do, sometimes I do things my own way because I’m more comfortable with my own approach.

So I did the same with Dodonpachi DaiOuJou. The good thing with DaiOuJou, or other Dodonpachi games, is that it is very easy to judge whether or not what you are doing is good enough. As long as you can keep your chain and make the hypers drop at the same time, it’s good enough for a start. This differs from a game like Ketsui or Futari, where you will have to be your own judge, deciding which amount of chips/gems is good enough.

The first thing I tried to do was very basic. I wanted to chain until I activate my first hyper on top of the tower of Stage 1, as you can see in most superplays. This looks really easy at first. But you quickly realize some difficulties. In order for this to happen, you need to kill the enemies in a specific way (with your shot or your laser), there are some timings that have to be respected ….

Usually, I told myself: “I need to execute this small section three times in a row without any mistakes before learning the next section.”

Very soon, I found myself trying to find visual cues in the superplays. Where should I stand on the screen to kill only these enemies and leave the ones I need alive? At what exact moment do I need to start moving?

Before starting a new stage, I would re-read all Pazzy’s annotations about the stage, then I would start learning it.

Usually, the best way to practice for me was while at home. I would determine the small section I needed to learn, usually 10, 20, or 30 seconds depending on the complexity of the section. Then, I would watch the superplay, try to determine a visual cue for every movement or action, and then try to reproduce it. Easier said than done ^^.

In addition to at-home practice,  I was watching superplays during my lunch break at work. Since I couldn’t really practice there, I would rather try to find as many visual cues as possible, so this way, when I get back home, I would not need to analyze the replays too much. By doing this, I ended up with many written pages of routes and visual cues. The funny thing is that you can know which routes I studied at work because when I was practicing at home, I didn’t need to write anything as I was practicing the sections immediately. My Ketsui routes are all written down for the first loop :p


Juju’s handwritten Ketsui route notes (in French).


I proceeded like this until I finished learning a stage, before the boss. At that point, I would work on doing the whole stage without any major mistakes. In the case of DOJ, this would mean no chain breaks, no deaths, and no missed hypers. Usually, you will find that when you are trying to put all the sections together other difficulties/small problems will arise. Sometimes, the execution between two sections is a little bit messy, or sometimes the savestates give you too much comfort and create bad habits.

However, if you need more than ten tries to do it, this probably means you need to rethink your approach and learn some of the sections a little bit better.

For me, the boss fight was the reward after learning the whole stage. Boss fights are typically less about execution and more about dodging, which I’m more familiar with, thanks to Touhou. Boss fights are also a nice break from learning the stage.

With bosses, you need to experiment a lot. You should try to determine what is static, what you can influence, and what is random. If something is really difficult, you can watch the superplays and try to determine if there is some logic behind every small movement. You will be surprised by how much you can discover once you are familiar with the game.

When I finally became satisfied with my knowledge of a stage and its boss, I would move to the next stage and repeat. I really love this specific period of playing because you are learning a lot and you feel like you are always getting better at the game. Everything is new and everything feels fresh. You are rewarded for being active and trying to understand as much as you can.

It’s not the time to start grinding yet.

After I finish learning the first big milestone (usually the first loop), I then go ahead and start doing a few runs. I typically really enjoy these first few runs. You probably don’t have a real PB in the game and you will likely improve it with every run that you do. But very soon, you will start to see the problematic sections in your runs.

From this point is when I start to push for more consistency.


Part 2: Grinding and Consistency

Congratulations, you’ve reached the point where you have started to know the game very well! You have memorized every section one-by-one and practiced the execution of each one of them quite a lot. But still, you will likely find yourself dying at several points during your runs and this is where you really need to push further.

I have a real appreciation for tight execution and timings. When I worked on my routes, I try my best to find “THE ANSWER” to every section. If I die to something, especially during the stage sections, I have to know why.

Investigating my performance is really important to me. If I died because I was really off my timing on a certain section, it’s alright, I will just check after the run to ensure that I can do that section several times in a row. But if I did everything I was supposed to do and still died, this means that my knowledge about the game is not good enough.

During practice, you need to be quite judgmental about what you are doing. If you break your chain you need to know why. You can’t just try to do the same thing over and over and be satisfied because it works half of the time. If something doesn’t work, then you need to adjust your visual/audio cues. As I like to say, these games are already difficult enough — and you have to deal with the pressure of your own mistakes — you don’t want luck to be a factor. If luck is unavoidable, then you want to reduce the impact of luck as much as possible.

For example, in Dodonpachi DaiOuJou, if you want to score, you need to be able to no-miss no-bomb every section of the stage without breaking your chain. This is completely doable, the game is well designed. So every time I died during the various stage sections, I knew it was my own fault. Either I made a movement/positioning mistake or I tried to “luck out” a section.

The best way to correct movement mistakes is grinding. By doing the sections of the stage over and over, after I am convinced that if I respect everything I know about the stage, then everything just comes down to memorizing the right timing, like notes on a piano.

If I died from something unexpected or from something other than mistiming, then it’s probably because I was lazy and should have learned a section better.

Moral of the story, don’t be lazy, please improve your routes. Then, when you feel confident about them, grind them.

It is a feeling I am familiar with when learning a piece for the piano. There are often some troublesome sections. You can feel that when you play them, they are a little bit messy, imperfect, but they are alright. Someone who is not into music probably wouldn’t notice them. But, as a musician, you need to be honest with yourself and improve these sections if you really want to master what you are playing.

At least, in shmups, the game tells you directly that you are wrong either by destroying your ship or holding back points. It’s an impartial judge, most of the time.


Juju hitting the midboss link


Moving forward, after I feel confident enough with my grinding and adjustments, I start doing runs until I put up a score that reflects what I have learned. After that point, I then learn the rest of the game the same way: lather, rinse, and repeat. Do some runs, see what’s wrong, make some adjustments and grind again.

In the end. I am pretty sure that I spent 80% of my time learning and practicing and only 20% of my time doing runs.

*Prometheus reported a similar practice to run ratio in my interview with him.*

It is only when I want to get a new PB that I start doing a lot of runs, because runs are also very important. As players, I do feel like we need to build some sort of endurance, or at least learn how to maintain focus for a long period of time under pressure. The first time that I have a good run, I know that I will probably crumble under pressure. But after you get in the same situation 10 times, it gets better. And if you are able to do it once, you will do it again, maybe even better if you keep pushing.

*I’ve had a very hands on experience with my really bad nerves while doing DDP 2-all attempts, the way I’ve gotten better is through doing full runs and repeated exposure to the sections that would make me nervous before.*

During the period when I start doing a lot of runs, this is my routine: I warm up by doing some technical/difficult sections, the full chain of 2-5, the 1-3 midboss link, the 2-4 railguns, etc. Sometimes, when I need a palate cleanser,  I’ll train on some Touhou spell cards. After that, I start doing full runs. Ideally, you can record yourself.

*Somewhat tangential to Juju’s viewpoint, but please do figure out a way to record your runs and highscores. The shmup community needs more videos of accessible gameplay, rather than just a handful of really impressive superplays. As Juju has demonstrated in his writing, replays that utilize different strategies and approaches are very useful. Can we please move past the days of our records just being a name and a string of numbers?*

Some days, if I feel that I am playing really poorly, or if I feel too frustrated, I switch back to training and practicing.

At the end of my session of runs, I try to remember everything that went wrong during my attempts: every death, every chain break, everything I could improve on (this is why recording your runs is helpful). If I remember clearly doing a movement mistake, it’s alright, I just practice the section a few times and ensure that my knowledge about the section is sufficient. But if I die in training, this means I need to change a few things.

In the end, I think it depends on the individual player to find the perfect balance between training and runs. Some players are able to understand and adjust after every run. This is not something that I am able to do, so I end up training a lot.



“Full Extent of the Jam,” a guide to learning the genre written by Prometheus:


My interview with Prometheus about Full Extent of the Jam, as well as DDP and DOJ:


Dodonpachi DaiOuJou White Label Video Index:


Dodonpachi DaiOuJou Black Label Video Index:

Back to Top