[cups.general] Fwd: Fwd: very strange dot-matrix printer problem: randomly wrong print result (ghostscript-esc p2)

[Ambrose Li]

I want to know who is resposible for fixing this borken spam filter on
this list...

---------- Forwarded message ----------
From: Ambrose Li <ambrose.li at gmail.com>
Date: 14-Sep-2006 09:06
Subject: Re: [cups.general] Fwd: very strange dot-matrix printer
problem: randomly wrong print result (ghostscript-esc p2)
To: "Mirror of cups.general Newsgroup" <cups at easysw.com>


On 14/09/06,  <zhangweiwu at realss.com> wrote:
> I am not sure... I saw Chinese text directly printed out when power-on
> the printer with 'font' button pressed (means to print a setting menu).
> Besides, the printer said it has a Chinese font stored in the printer,
> however I cannot manage to print them out. E.g. 'echo 中文 > /dev/lp0'
> only produce junk. Both UTF-8 and GB18030 has been tried.

Sorry for not checking first, and for the quick assumption. At that
time I forgot that there were things like Japanese printers and
Chinese terminals.

I found an online manual and it says the printer accepts GB18030. I
don't know why the GB would give garbage, unless there is some special
way to tell it to accept GB.

> Ah, yes, and now I purchased a second new wire, the result is.... the
> same! This is really frustrating because this is the final last
> show-stopper for us to move this shop to move everything to Linux.
>
> Is it possible that this has something to do with the special printer
> model? Because it is LQ-300K model, K probably means it has Chinese font
> in it, there might be tiny difference between the real LQ-300 model.

Yes, this is possible. The manual says the command set is "ESC/P-K",
which I assume must be somewhat different than normal "ESCP/2". But I
don't know what the differences are.

Perhaps we can do a little test to check whether it is the command
set. I can think of two ways to check this:

1. Go to Windows and do a Print to File of something. Take the
resulting raw print file to Linux and cat it to /dev/lp0. If it is
wrong then the problem is with Linux's parallel port driver (or the
hardware, but I guess we have just ruled it out).

2. Go to Linux and generate a raw file from Ghostscript using the
correct printer driver. Take this to Windows and copy it to lpt1 with
the /b option. If it is right then the problem is not with
Ghostscript's epson driver.

Perhaps the first is easier to do as a quick test.
--
cheers,
-ambrose