I have done further investigation, and it appears that there
must be something wrong with the setup. The printer is indeed
not a postscript capable printer, so some preprocessing/rendering
is being done on the computer, and a graphical image is
being sent. The slownes is not due to CPU bottleneck, as
shown by top:
Cpu(s): 1.7% us, 5.0% sy, 0.0% ni, 93.3% id, 0.0% wa, 0.0% hi, 0.0% si
Copying text files directly to /dev/lp0 results in very fast
printing (the files must first be run through unix2dos to put
in the CRs). So the problem is not just a slow interface on
a byte-by-byte basis.
The usual Epson standard for transferring graphics images
(which I presume, but have not checked) this printer uses
is a bit-for-bit technique (almost). Since the printer is
300x300 bpi, and a full page has 1/4" margins top, bottom,
and sides, that's 8x10x300x300 = 7200000 bits. Dividing
by 8 bits transferred = 900000 bytes to transfer.
The Centronics spec shows that the minimum time to transfer
a byte is 2.75us, so the entire page must take at least
2.475 seconds to transfer. But I'm seeing 30 minutes.
So this achieves only 0.1375% of the maximum speed.
Something is wrong.
A few days ago, immediately after a reboot, I found that
a page printed in about 30 seconds. It was definitely
printing as a bit image. Now I'm back to 30 minutes
per page.
Can anyone point to help on configuring the printer device?
Perhaps interrupts are disabled, or sth the like.
Mike
--
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
