Skip to content

MTGAP

_-
Y
P
O
U
J
K
D
L
C
W
I
N
E
A
;,
M
H
T
S
R
"'
Q
Z
</
>.
?:
B
F
G
V
X
Finger 123478910
Usage 7.99% (P95)7.07% (P16)15.82% (P96)11.18% (P16)12.61% (P64)12.93% (P50)8.20% (P28)6.97% (P30)
Same Finger Full Scissor Half Scissor Lat. Stretch
Bigram 1.06% (P72) 0.04% (P14) 1.24% (P33) 0.50% (P31)
Skipgram 4.69% (P48) 0.14% (P21) 2.89% (P74) 1.04% (P45)
No Thumbs Left Space Right Space
Weak-ish Redirs. 0.45% (P21) 0.91% (P61) 0.55% (P21)
Weak Redirects 1.47% (P91) 0.71% (P93) 0.71% (P93)
Other Same Finger 18.96% (P47) 12.76% (P58) 12.76% (P58)
Rolls : Alts 1.57 (P52) 1.47 (P29) 2.13 (P88)
2-Roll In : Out 1.88 (P75) 1.93 (P94) 1.17 (P60)
3-Roll In : Out 3.52 (P75) 1.56 (P58) 3.53 (P80)
Author
Michael Dickens (MTGAP)
Year
2012
Finger Map
Traditional

MTGAP refers to a series of keyboard layouts generated by Michael Dickens, who also goes by MTGAP, using a keyboard layout optimizer that implements simulated annealing.

Michael Dickens’ blog contains numerous samples of layouts generated between 2008 and 2012, and some but not all of them carry the name MTGAP. Nowadays, MTGAP mainly refers to the 30-key version from this 2012 blog post—despite this particular variant not being called MTGAP (or any name at all, for that matter) by Michael Dickens himself.

MTGAP is notable for being substantially better than all keyboard layouts before it (with the arguable exception of Maltron, which has the advantage of E on thumb). The metrics used in its design share more in common with much later layout optimization projects than with the other projects of its time. This includes a stunning repudiation of the common belief at the time that putting the most common letters on the home row and minimizing finger travel distance are the primary goals of a good keyboard layout.