HC 3.0 generates gibberish with rules and full syntax on WPA

[setekh]

1. Wrong / gibberish output when full syntax is used and rules applied.
When run like so :

Code:

hashcat64.exe  -r ".\myrules.rule" ".\test.txt" --stdout

all output is fine...

password
11pass11
00pass00
12pass12
1234pass
4321pass
pass4321
pass1234
drowssap
PASSWORD
Password
passwordpassword
...

but when using full syntax :

Code:

hashcat64.exe -a 0 -m 2500 --session=all -o ".\found.txt" --outfile-format=2 -w 2  -r ".\myrules.rule" ".\Capture_File(WPA).hccap" ".\test.txt" --stdout

the result contains wrong generated words / gibberish then the proper generated words from rules :

G±P(æ1yy܃l-
 w

11G±P(11
00G±P(00
12G±P(12
1234G±P(
4321G±P(
G±P(4321
G±P(1234

w 
-lƒÜyy1æ(P±G
g±p(æ1YY܃L-
 W

G±p(æ1yy܃l-
 w

G±P(æ1yy܃l-
 w

...
...
password
11pass11
00pass00
12pass12
1234pass
4321pass
pass4321
pass1234
drowssap
PASSWORD
Password
passwordpassword

Rules file :

# Dict word unchanged
:
# Ninja stuff
x04^1^1$1$1
x04^0^0$0$0
x04^2^1$1$2
x04^4^3^2^1
x04^1^2^3^4
x04$4$3$2$1
x04$1$2$3$4
# Reverse
r
# Upper - unsupported by GPU .. trying t for toggle will do in all lower case and upcase first only
#U
t
c
# Dupe
d
# Change first
T0
# Dup first 4 and trunc to 8
y4'8
# Append common
$!
$.
$1
...

the test.txt contains one word only - password

Any ideas why this is happening?

2. Also i want to override the drop of initial word when using WPA/WPA2 when it is < 8 chars and drop it after the rules are applied, but <8 is not supported, is there a way to do that and how.
Thank you.