Why I build these tools

I want the songs people love to be playable, and charted well

Everything on this site comes out of that. I only play Clone Hero drums, so that is the perspective all of it is written from.

What I want

More high-quality charts for the songs Clone Hero and YARG players want to play. Old uncharted songs, niche songs, new albums.

This started with wanting to play songs I already loved, and finding that the ones I was most excited about had never been charted. So I charted some myself. Later I started playing drums in a local band, and the same thing happened from the other direction: the songs my group wanted to play had no charts either.

There have historically been three ways to get a chart for a song. A charter charts the song they wanted to chart, you pay someone, or you chart it yourself. All three are bounded by how many hours charting takes. That is the number I am trying to move, and I want to move it without touching what counts as good.

Building lyric alignment is where this became a principle rather than a habit. I could have run it over every chart on my own drive and been finished in an afternoon, and that would have helped exactly one person. Everyone else downloading those charts still gets no lyrics and has no idea my tool exists. The version worth building was the one a charter would want to use, as a first pass, on a chart they were going to publish anyway.

Who this helps, and how

The same underlying work is worth different things to different people, and is probably worth handing to each of them in a different shape.

ChartersCharting a song takes hours, and a lot of those hours go to work that is repetitive rather than musical.If a song takes less time, maybe more songs get charted, and more of them are songs somebody was waiting for.
New chartersThere is a steep learning curve to charting tools, and many people do not see it through.If there is less to learn before you get to the musical part, maybe more people with good ears stick with it long enough to get good.
Players charting for themselvesSome people are not publishing anything. They want to play one specific song at home tonight.The same work serves them, handed over in a different shape, and nothing they make has to reach anyone else.

These tools already exist

Charting is already partly automated, and where the output is good it has been welcomed.

Charters built tools that generate the lower difficulties from an Expert chart, and they seem broadly used now. Charters looked at that output, decided the quality was good enough, and the player base got Hard, Medium, and Easy on far more songs than anyone was ever going to chart by hand.

The same thing happens further up the process, more quietly. Charters use AI to separate stems to get an isolated drum track, and use peak picking in a DAW to place snare and hi-hat hits off the waveform. That work runs on software someone had to know to find, install, and learn.

So there are two things worth improving: how good the output is, and who can get at it. Mine run in a browser, free, with nothing to install. Someone with good ears and no audio production background should be able to get to the part where their judgment matters.

What I hold to

These are the constraints I build under, and the reasons I picked them.

The bar belongs where it is

These games are fun because the community holds a real standard for what a chart has to be. I want more songs at the quality that already exists, not more songs at a lower one.

The tools should take the repetitive work, not the judgment

Placing hundreds of notes you already know are there is work a machine should do. Deciding which lane a cymbal belongs on, and what a chart should feel like to play, is the craft, and two good charters given the same song will make different calls about it.

That part should not be automated.

Every tool creates a draft

What comes out of these tools is a draft, and the last step is you reviewing it and deciding what ships. None of them upload anywhere.

A flood of unreviewed charts would land on the people reviewing submissions, and that is not a cost I want to hand them.

These tools generate first drafts

I do not think AI is good enough to produce a drum chart that needs no edits to clear the bar this community has, and I am glad the bar is where it is.

Everything here is a first pass for a person who is going to review it, which is why the output opens in an editor instead of a download, and why each page tells you which parts to check first. Getting closer still helps, because the real comparison is not a perfect chart against a draft. It is a draft against a song nobody ever charts.

If a tool of mine ever makes it easier to publish something bad than to publish something good, that is a defect in what I built, and I want to know. Tell me on Discord.