Over the Frail Dream

A guide to charting/playing minedodge

Minedodge refers to Etterna charts that use mines in a way that demands effort from the player to avoid beyond knowing not to hit them.

Mine Mechanics (In Etterna 0.74.4)

To know how to make a good minedodge chart, understanding of mine mechanics is needed.

Mine does not overlap a note's timing window

If a mine is not within a note's hit window (+-180ms), it has a window of +-75 ms to be hit, and will also be hit if the column is held when the mine passes the receptor.

Mine overlaps a note's timing window

If a mine's window overlaps a note's hit window, it's more complicated. If the mine is before the note (early mine), the mine will trim the note's early window to however large the gap between the mine and the note is in milliseconds. If the mine is after the note (late mine), the mine will effectively trim the note's late window to half of the gap. This is because mines after notes follow the same note priority rules as normal notes, so if you hit closer to the mine than the note your input will hit the late mine instead of the note.

As with the non-overlapping case, you can also hit a mine by holding the key while the mine passes the receptor.

Consequences of Mine Mechanics

The non-overlapping case can mostly be ignored for minedodge, since mines not near notes are usually trivial to avoid once you have the muscle memory to not try and hit them.

For the overlapping case, the asymmetrical windows mean that late mines are, at minimum, twice as hard as early mines at the same distance away. Possibility much, much higher when factoring in release time.

Calculating mine tightness

From these rules, a simple calculation can be done for how tightly a mine constrains the timing window of a note.

For early notes this is simple. The new timing window is just the millisecond gap, so:

earlywindow = ms_gap

For late notes, it's a bit more complicated because of the possibility of hitting a mine by pressing too long.

latewindow = min(ms_gap/2,ms_gap-tap_duration)

These formulas can be used for many interesting purposes, most notably determining whether early or late window abuse is better.

Here's a practical example: Noname Rhythm Edit in HSMP4 has 48th early mines and 16th late mines at 150 BPM. The gaps are 33.33 milliseconds for the early window and 100 milliseconds for the late window.

Lets plug these into our equations:

earlywindow = 33.33

latewindow = min(100/2,100-tap_duration)

So the early window is 33 ms (about in the middle of the perfect window) and the late window is whichever's smaller: 50 ms or 100 minus the tap duration.

If all of your keytaps are as short as possible, then the value is:

latewindow = min(50,100-0) = 50

So you would have a late window of 50 milliseconds, much bigger than the 33.33 ms early window, so it's best to hit late. But what if your keytaps are all 80 milliseconds long?

latewindow = min(50,100-80) = 20

Now your late window is 20 milliseconds, even tighter than the early window, so you should hit a bit early. This shows how the "optimal" way to play a file can change drastically depending on how fast you can release the keys.

Early/Late Abuse

It was implicitly stated in the last section, but if the early and late windows created by the mines are imbalanced, you can take advantage of this. For example, if a chart has only late mines, you can hit earlier so your late hits aren't as late, and consequently you'll be less likely to hit the late mines.

Hitting early is particularly effective on a lot of naively-made minedodge charts that have early and late mines the same distance away; in these cases the early window is much bigger, so hitting early lets you play the file with less precision and fast releases needed.

Some very difficult modern minedodge charts will deliberately use extremely tight mines on one side of the note in order to force early/late abuse. Difficulty of early/late abuse is much more about reading/memorization than precision, as you need to recognize a mine pattern that requires hitting early/late and then adjust your inputs accordingly. Consequently, early/late abuse minedodge difficulty is impacted much more by visual clutter. Frequency and predictability of changes between early/late/no abuse also matters; it's a lot less mental strain if you know that every note in a section requires hitting early than having to constantly be visually parsing mine gaps to know whether or not you need to hit the note early/late.

It's also worth noting that since you need to hit early/late, this effectively means you're hitting offsync. Audio timers may find early/late abuse very unpleasant if not impossible. That being said, there's nothing stopping you from making the notes offsync to compensate, so that timing by audio will make you hit early/late and avoid the mines. You'll just be making visual timers miserable instead.

General Tips on Minedodge Design

Minedodge difficulty scales much more on rates than normal patterns due to the timing windows shrinking along with the increase in pattern speed. As a result, charts that mix minedodge with non-minedodge difficulty tend to have their difficulty curves drastically changed by up/downrating, usually for the worse.

For basic minedodge, treat the mines as constraints on the timing window of the note. Closer mines = smaller window = harder. Remember the formulas mentioned above for calculating effective early/late windows caused by mines.

Good minedodge players usually minimize their input lengths to ~50 ms on average. There can be a lot of variance in the input lengths. It may be a good idea to use an input logger to determine how fast your own releases are.

Compare your gaps to existing minedodge charts to get an idea of how hard exactly various mine gaps are. Look at scores to see when people decide early/late abuse is preferable.

Sandwich mines have their purpose, but sandwiching every note tends to look ugly. If you want to prevent early/late abuse, another method is to switch between early/late mines very frequently, and/or make your minedodge chart easy enough that it's not worth trying to abuse.

Mineart and minedodge usually go hand-in-hand. You usually don't need to worry about any details of the mines besides proximity to notes, though realize that high mine density near notes may result in harsher punishment when hit. There is no grace period between mine hits, so holding a key over a stack of 192nd mines will result in instant death.

Since most players use small dot mines, realize that extremely tight mines before a note will be invisible due to the note covering the mine. This also applies to holds if the noteskin uses an opaque note tail.

General Tips on Playing Minedodge

Performing well on minedodge is a combination of:

Accuracy is straightforward. Since mines act as constraints on the timing window, if you can't hit the notes accurately then you're not going to be dodging any mines.

Technique is important due to being able to hit late mines if you hold the key for too long. You need a playstyle that presses the keys for as little time as possible. Otherwise, it's possible to have extremely good accuracy but still hit late mines. The traditional ways of playing minedodge are either to hit very lightly only exerting enough force to trigger the actuation points, or hit hard and fast so even though you bottom out, you rebound off and release the key fast enough that the time spent pressing the key is short anyway.

Rapid trigger keyboards make good minedodge technique easier by automatically releasing the keys for you once you aren't pushing down, but it doesn't trivialize it as you still need to minimize the time spent pressing and holding down the key. If you're going the full-force approach to minedodge technique, switches with less of a "buffer zone" (gap between the actuation point and bottoming out) help with this. Rapid trigger keyboards let you adjust the actuation point for a smaller buffer zone but you can also get this effect on flat scissor switch keyboards, like the ones you see on most laptops.

Reading is important at every level of minedodge for different reasons. When starting off, you obviously need to learn that mines aren't something you're supposed to hit. It's not as simple as just ignoring them though, as you'll need to sometimes deliberately change your playstyle to adapt to their presence, whether that be releasing quicker, releasing holds early, or knowing whether to early/late manipulate.

Early/late manipulation is the most niche skill but necessary to learn for certain minedodge charts. Being able to recognize when it's needed and override your usual instinct to hit the note itself as accurately as possible is not easy. This ties into reading but is distinct enough to be worth mentioning separately.

See also