[email protected] wrote:

> Um Verschiebungen der Hand (beziehungsweise Mitbewegen benachbarter
> Finger) zu berücksichtigen, kompensiere ich in meinem Optimierer den
> Mehraufwand für Tasten abseits der Grundposition teilweise, falls die
> Taste nach einer Taste mit derselben Hand in derselben Zeile
> angeschlagen wird.  Die Kompensatio

Das ist eine klasse Idee! 

Habe ich gerade als zusätzliche Kostenfunktion eingefügt: 
„Malus, wenn auf der selben Hand in anderer Zeile“ (es gibt keine Boni :) )

Aktuell sieht eine ausführliche Ausgabe des Layout-Testers damit übrigens so 
aus: 

$ ./check_neo.py -v 
Neo
[['^', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', '-', '`', ()],
 [(), 'x', 'v', 'l', 'c', 'w', 'k', 'h', 'g', 'f', 'q', 'ß', '´', ()],
 ['⇩', 'u', 'i', 'a', 'e', 'o', 's', 'n', 'r', 't', 'd', 'y', '⇘', '\n'],
 ['⇧', (), 'ü', 'ö', 'ä', 'p', 'z', 'b', 'm', ',', '.', 'j', '⇗'],
 [(), (), (), ' ', (), (), (), ()]]
# 1.633683464 billion total penalty compared to notime-noeffort
# 3.9755737368 mean key position cost in file 1gramme.txt
# 4.82198888113 % finger repeats in file 2gramme.txt
# 4.56473130298 million keystrokes disbalance of the fingers
# 0.307224610884 % finger repeats top to bottom or vice versa
# 4.70844781487 % of trigrams have no handswitching (uppercase ignored)
# 0.126744245 billion rows to cross while on the same hand

Qwertz for comparision
[['^', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', 'ß', '´', ()],
 [(), 'q', 'w', 'e', 'r', 't', 'z', 'u', 'i', 'o', 'p', 'ü', '+', ()],
 [(), 'a', 's', 'd', 'f', 'g', 'h', 'j', 'k', 'l', 'ö', 'ä', (), '\n'],
 ['⇧', (), 'y', 'x', 'c', 'v', 'b', 'n', 'm', ',', '.', '-', '⇗'],
 [(), (), (), ' ', (), (), (), ()]]
# 2.506007036 billion total penalty compared to notime-noeffort
# 5.95987411112 mean key position cost in file 1gramme.txt
# 6.72632774411 % finger repeats in file 2gramme.txt
# 7.75577431662 million keystrokes disbalance of the fingers
# 0.233732243474 % finger repeats top to bottom or vice versa
# 11.5644849023 % of trigrams have no handswitching (uppercase ignored)
# 0.228822782 billion rows to cross while on the same hand

And the Nordtast Layout
[['^', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', 'ß', '´', ()],
 [(), 'ä', 'u', 'o', 'b', 'p', 'k', 'g', 'l', 'm', 'f', 'x', '+', ()],
 [(), 'a', 'i', 'e', 't', 'c', 'h', 'd', 'n', 'r', 's', 'ß', (), '\n'],
 ['⇧', (), '.', ',', 'ü', 'ö', 'q', 'y', 'z', 'w', 'v', 'j', '⇗'],
 [(), (), (), ' ', (), (), (), ()]]
# 1.610586855 billion total penalty compared to notime-noeffort
# 4.28745556484 mean key position cost in file 1gramme.txt
# 1.83742456166 % finger repeats in file 2gramme.txt
# 5.20508768865 million keystrokes disbalance of the fingers
# 0.260071071233 % finger repeats top to bottom or vice versa
# 3.80773505265 % of trigrams have no handswitching (uppercase ignored)
# 0.11622743 billion rows to cross while on the same hand

And Dvorak
[['^', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', 'ß', '´', ()],
 [(), '’', ',', '.', 'p', 'y', 'f', 'g', 'c', 'r', 'l', '/', '=', ()],
 [(), 'a', 'o', 'e', 'u', 'i', 'd', 'h', 't', 'n', 's', '-', (), '\n'],
 ['⇧', (), ';', 'q', 'j', 'k', 'x', 'b', 'm', 'w', 'v', 'z', '⇗'],
 [(), (), (), ' ', (), (), (), ()]]
# 1.74961443 billion total penalty compared to notime-noeffort
# 4.61956626355 mean key position cost in file 1gramme.txt
# 2.54541079218 % finger repeats in file 2gramme.txt
# 7.96104929652 million keystrokes disbalance of the fingers
# 0.432284336966 % finger repeats top to bottom or vice versa
# 3.64150111878 % of trigrams have no handswitching (uppercase ignored)
# 0.115537747 billion rows to cross while on the same hand

And Colemak
[['^', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', 'ß', '´', ()],
 [(), 'q', 'w', 'f', 'p', 'g', 'j', 'l', 'u', 'y', ';', '[', ']', '\\'],
 [(), 'a', 'r', 's', 't', 'd', 'h', 'n', 'e', 'i', 'o', '`', (), '\n'],
 ['⇧', (), 'z', 'x', 'c', 'v', 'b', 'k', 'm', ',', '.', '/', '⇗'],
 [(), (), (), ' ', (), (), (), ()]]
# 1.660105182 billion total penalty compared to notime-noeffort
# 4.08994395442 mean key position cost in file 1gramme.txt
# 2.7257344839 % finger repeats in file 2gramme.txt
# 6.30282470377 million keystrokes disbalance of the fingers
# 0.14253925436 % finger repeats top to bottom or vice versa
# 9.25218658326 % of trigrams have no handswitching (uppercase ignored)
# 0.146366287 billion rows to cross while on the same hand

Liebe Grüße, 
Arne

Antwort per Email an