ICGCPI(*CPI)

["Campden Wonder" <pete@xxxxxxxxxxxxxxxxxxxxxx>]

Can anyone shed any light on how DBCS font size is affected by the spooled 
file attribute IGCFNT(*CPI)?

The IBM documentaiton/help text is ambiguous.

CRTPRTF seems to suggest that (as you might expect), IGCCPI(*CPI) means that 
the DBCS font size is dependent on the CPI setting:
-----------------------------------------------
DBCS characters per inch (IGCCPI) - Help

Specifies the printer character density of double-byte character set
(DBCS) characters, in characters per inch (CPI).

    Note:  This parameter does not specify the printer character density
    of alphanumeric characters.  Alphanumeric characters are printed
    with the value specified for the CPI parameter.

*CPI
    DBCS character density is based on the values specified for the
    Characters per inch (CPI) parameter.  The system prints one
    double-byte character for every two alphanumeric characters.

     o  For CPI(10), DBCS characters print at 5 characters per inch.
     o  For CPI(12), DBCS characters print at 6 characters per inch.
     o  For CPI(13.3), DBCS characters print at 6.7 characters per inch
         (same as IGCCPI(*CONDENSED)).
     o  For CPI(15), DBCS characters print at 7.5 characters per inch.
     o  For CPI(18), DBCS characters print at 9 characters per inch.
      o  For CPI(20), DBCS characters print at 10 characters per inch.
-----------------------------------------------

which makes sense.

Meanwhile, DSPSPLF help text states:

-----------------------------------------------
                Characters Per Inch - Help

The number of double-byte characters to be printed per
inch.

The options are:
 o  *CPI        5 double-byte characters per inch.
 o  *CONDENSED  20 double-byte characters per 3 inches.
-----------------------------------------------

which suggests the DBCS font size is always 5 CPI whatever the setting of 
CPI.

I have a spooled file with CPI(15) and IGCFNT(*CPI) which seems to print 
DBCS characters at 5 CPI, as per DSPSPLF.

Can anyone confirm that  the CRTPRTF help text is wrong and IGCFNT(*CPI) is 
always 5 CPI, irrespective of the value of CPI?

Thanks

Pete