MICR printing via pcl2afp transform.
ITEM: RTA000159651
Q:
Topic thread:
Printer Systems (PRINT - NA/ATS)
PSF/AIX
Customer sends a PCL check job to PSF/AIX. It goes through the
pcl2afp transform and is sent to a MVS/PSF host (via upload-n-print).
From MVS/PSF V2.2 the job is routed to an IP60 MICR printer (from
Rosetta). The customer has been having marginal MICR signal readings.
Rosetta has worked hard along with the IBM CE to rule out hardware
causes for the marginal signal readings. To confirm this, the customer
has an AFP check job that has very acceptable MICR signal readings.
Our focus is turning to the pcl2afp transform. The customer sent their
HP PCL MICR softfont to a developer in Boulder who added it to the
transform. Specifically, he takes the PCL softfont for the MICR and
the signature fonts from the customer and uses them to build a UserInit
file. He adds some escape sequences to make the fonts get loaded in the
interpreter cache and remain there while the interpreter is running.
Boulder tested the application before they sent it back to the customer
to load onto their system. However, the testing was for format and
text appearance. They did not test MICR strength and signal readings.
Our concern is the output of the pcl2afp transform is an IM1 image
including the fonts. We feel that the appearance of the MICR font as
image "softens" the edges of the MICR font and reduces the MICR signal.
Is this a valid concern? If so, is there anyway around this?
Specifically, is there a way to add the MICR font "after the transform"
versus "before the transform"? Is there a better quality MICR font that
could be used?
Is there anything else that can be done to / with the pcl2afp transform
to produce a "crisper" image or MICR line?
If this is the best that can be done using PCL, would we have better
results if the customer had PostScript output and we used the ps2afp
transform? I don't think the customer's application (PeopleSoft) can
output afp.
I really appreciate your suggestions and look forward to hearing from
you soon. Thank you.
A:
I'm still checking with one of our specialists, but here is the
response I have drafted to date:
**********************************************************************
The first thing to verify is that the resolution specified for the
pcl2afp transform (in the pcl2afp.cfg and pcl2afpd.cfg files) matches
the configuration set at the Infoprint 60. You will get sharper
output if both are set to 300-pel, partly because no bits are lost
if they're the same and partly because PCL was written by HP for their
printer line, which began with 300-pel printers. We do say in our
announcement letter for PSF/AIX that PCL output from our transform
that is printed on 240-pel printers may or may not be acceptable for
that very reason.
You asked about PostScript as an alternative. PostScript has always
been resolution-independent (with the exception of imbedded
resolution-specific images), and its use of Type 1 Outline fonts
can give crisper output, especially on a 240-pel printer.
You say that the output of the PCL transform is IM1 image. Why? The
default for the pcl2afp.cfg and pcl2afpd.cfg files is to produce
compressed IO1_G4 image, and the IP60 can certainly support compressed
image. From a crispness standpoint, there should be NO difference in
the image produced by IM1 and the image produced by IO1_G4.
I can't answer your question regarding if another PCL MICR font would
provide crisper output. Mark has left IBM, but I can ask whomever has
taken over the pcl2afp transform. Do you know which PCL softfont you
sent Mark for him to convert?
I know of no way to add the MICR font after the transform, given the
configuration you describe. It is possible to add an AFP overlay to
a transformed job, but I don't know how you could create an overlay
on the host with the dynamic data.
If PeopleSoft can output ASCII line data, it may be possible to use
ACIF and a pagedef/formdef, along with an AFP MICR font from Rosetta
(tweaked specifically for readability) to create an AFP text file and
forget about the transforms that produce image.
**********************************************************************
I'll update this item when I receive additional information from my
coworker.
Q:
Thank you for the guidance. We may come out of the transform as IOCA. I
was led to believe otherwise. I look forward to whatever else you can
find out for us.
A:
I checked with Jamsie, and here are her and my combined comments:
I have still not received any responses to my questions to AFCCU
development - on whether a font can provide different sharpness
than an image. I think the answer is that PQE can be set differently
for image and for fonts -- but I don't know this. This may be a
MINOR component of the issue, but I had wanted an answer on this for
completeness.
Using ASCII and AFP text is probably the best -- if they can do it.
My other thought is maybe we could use Rosetta's version of the MICR
font in the transform, rather than the PCL MICR font. Rosetta should
be able to say if they require special shapes to the characters -
different than the shapes that are in the PCL micr font.
If Rosetta has their fonts in PCL soft font format, then it should be
possible for one of the developers to wrap those fonts in the needed
wrappers and incorporate into the transform (UserInit file). But
they will still be converted to image through the pcl2afp transform.
I think you should have the customer work with Rosetta on this.
Rosetta should review the font that is being used in the transform,
and say whether it meets their standards. If they think there is no
problem with the font, then they should review the print output and
see if they think it is a problem with printer settings.
Please keep me posted and let us know if there's anything else we can
do.
Q:
Here are the results of a conference call with the customer, Rosetta,
and IBM participating.
1) The font that they sent to Boulder is a MICR font from PeopleSoft.
Rosetta will mail the customer their version of the MICR font. It
is yet to be determined how to use that font in the current process.
2) We need to find out if Peoplesoft can output ASCII line data.
3) IBM will verify that data stream resolution is compatible with
printer resolution.
Another conference call is scheduled for this week.
A:
1) In my past experience (limited), the Rosetta MICR font has been
a raster version which could be used with line data, but not PCL.
If a pagedef can be developed to call the MICR raster font for the
correct data record, then this should work.
I asked one of the developers if he thought the quality would vary
from one PCL soft font to another, and he said he didn't know for
sure, but didn't think so.
2) It is my understanding that PeopleSoft can output ASCII, PCL and
PostScript.
To be continued.
S e a r c h - k e y w o r d s:
psf/6000 psf/aix psf aix infoprint ipmgr aix pcl pcl2afp micr quality
resolution crispness discard rate reject read magnetic check font
soft hp userinit peoplesoft ascii postscript ps2afp line data output
WWQA: ITEM: RTA000159651 ITEM: RTA000159651
Dated: 10/1999 Category: XPSF6000
This HTML file was generated 2000/11/30~13:34:12
Comments or suggestions?
Contact us