there's actually no reason that you couldn't have a 20th note, you just couldn't express it in terms of quarter notes because you get an infinite series. it's convergent. but it doesn't help the notation.

you couldn't count it, but you could feel it.

i flipped my fraction over and thought i had an answer (i do this, flip the fraction over, to myself all the time), because i'm actually used to dealing with these kinds of weird conversions as a relic of writing drum sequences in a scorewriter. there's lots of ways to convert weird signatures. the series just has to terminate. it doesn't here, but that's just bad luck.

if you split the bar into five equal temporal spaces, each one will be four fifths of a quarter note. .8/4 = 0.2. so, a 20th note would be 20% of a quarter note - which is 40% of an eighth note and 80% of a sixteenth note. indeed, .8*(1/16) = 1/20. you can't notate this using western music theory (because the series happens not to terminate....), but that doesn't mean it's undefined or unplayable

80%(1/16th) =
160% (1/32nd) =
1/32nd + 60%(32nd) =
1/32nd + 1.2(64th) =
1/32nd + 1/64th + .2(64th) =
1/32nd + 1/64th + .4(128th) =
1/32nd + 1/64th + .8(256th) =
1/32nd + 1/64th + epsilon =
3/64th + epsilon

you simply couldn't hear 80% of a 256th note. that's indistiguishable from human error.

so, a 20th note would be indistinguishable from a triplet of 64th notes. that means that 4/20 would be four triplets of 64th notes.

....which is a blast beat.

http://thehardtimes.net/2016/08/29/stoner-tech-metal-band-trying-really-hard-write-song-420-time/?fb_comment_id=1161463340592034_1163796750358693