Problems connecting an old HP LaserJet 2100 to an XP Pro machine

[Ben Myers]

Obsolescence of the parallel port is a mindset of the computer industry, not me.
But I did not make that clear, did I? Every year, there are more and more
"legacy-free" computers which lack parallel, serial, and PS/2 ports, and have no
floppy drives.

Still, driving any sort of parallel device with a PCI-Express is a waste of
bandwidth because the parallel port hardware is a very small bottlenect compared
to PCI-E. This is true no matter what the expense of the parallel port printer.
Far better to simply get a PCI (not -Express) parallel card and get on with it.

Taking a cue from the legacy-free part of the computer biz, expensive bar code
and label printers are now more and more USB or sometimes are BOTH USB and
parallel with connectors for both... Ben Myers

 

[Ben Myers]

I'm have to beat this one to death.

"The maximum data transfer rate achievable with this (parallel port)
architecture is around 150 kilobytes per second and is extremely software
intensive."

A 1x PCI-Express slot has a transfer rate of 250 MEGAbytes per second, or about
1600 times faster than a parallel port.

Tell me how ANY parallel port device, no matter how expensive or exotic, can
take advantage of a PCI-Express connection to the motherboard/system?

.... Ben Myers

 

[nobody >]

How much data is transfered reading a bar code or to print a label? I
have a plotter that has a parallel interface and sending an average
24"x36" CAD drawing to it takes maybe 15-30 seconds. I don't see that a
barcode or label gets anywhere close to that amount of data. Even
considering your comment, I still don't see where an older parallel port
device could take advantage of the increased speed of this card.

It's not the bandwidth of the card at issue, it's getting a working
parallel port for what you have for a printer when there isn't one on
the new 'puter. If it takes a PCI-X card to do it, so be it. How many
people are going to use every slot on the mobo?

By your reasoning, we shouldn't be using USB keyboards either. They
don't use more than 5% of the bandwidth on even a USB 1.1 port. I'll
admit that I try to use the old PS2 ports (if available) for the mouse
and keyboard to leave more USB ports available. I've got a double
handful of USB to PS2 adapters left over from work. A surprising amount
of USB KBs and mice that don't say 'PS2 Compatible' do work with the
adapter. If it doesn't, no damage; just use up one of your USB ports instead

 

[Michael Johnson]

It's not the bandwidth of the card at issue, it's getting a working
parallel port for what you have for a printer when there isn't one on
the new 'puter. If it takes a PCI-X card to do it, so be it. How many
people are going to use every slot on the mobo?

I don't see where buying an $80 PCI-X parallel card makes any economical
sense when PCI versions are sold for less than $10. Other than the rare
instance that all the normal PCI slots are full what practical use is
the high aspect of this card? It is like putting 200 mph speed rated
tires on a Yugo.

By your reasoning, we shouldn't be using USB keyboards either. They
don't use more than 5% of the bandwidth on even a USB 1.1 port. I'll
admit that I try to use the old PS2 ports (if available) for the mouse
and keyboard to leave more USB ports available. I've got a double
handful of USB to PS2 adapters left over from work. A surprising amount
of USB KBs and mice that don't say 'PS2 Compatible' do work with the
adapter. If it doesn't, no damage; just use up one of your USB ports
instead

This isn't my reasoning it is yours and it is flawed because there are
devices that use the full bandwidth of USB 2.0 or 1.1 ports and devices
that use just a fraction of it. Parallel ports are slow, period. There
are no older parallel devices I know of that can use the claimed triple
speed of this card. If there are such devices they must be very rare.

 

[nobody >]

I don't see where buying an $80 PCI-X parallel card makes any economical
sense when PCI versions are sold for less than $10. Other than the rare
instance that all the normal PCI slots are full what practical use is
the high aspect of this card? It is like putting 200 mph speed rated
tires on a Yugo.


This isn't my reasoning it is yours and it is flawed because there are
devices that use the full bandwidth of USB 2.0 or 1.1 ports and devices
that use just a fraction of it. Parallel ports are slow, period. There
are no older parallel devices I know of that can use the claimed triple
speed of this card. If there are such devices they must be very rare.

My point is *still* that if you *need* a parallel port and the only
available slot is a PCI-X version, you use it. Bandwidth be damned!

By that reason again, bicyles should be banned because they put down
such a small footprint on the asphalt compared to an 18wheeler.

As for devices that can exceed the 150kb/s 'standard' parallel rate,
think ECC/ECP capable tape drives, Zip and LS120 drives and 2X CDR (on
up) burners (all external parallel). Years ago I did the math and found
speeds up to 300kb/s when the data was streaming solidly. (and yes, that
was running a Colorado 250 backup with compression turned off to get
real results.)

 

[Ben Myers]

YOU can spend the money on an extra expensive PCI-E parallel port card. I'll
spend the money on an inexpensive PCI parallel port card. In the foreseeable
future, I'll probably find more free PCI slots in computers than PCI-E slots.

This is an interesting defense of the parallel port with all the bidirectional
kludge devices that were sold in the mid-90s. Every one of them was a horrific
kludge that managed to work OK if it was the only device attached to the
parallel port. But try to daisy-chain TWO devices to a parallel port, as some
manufacturers claimed was possible, and, more often than not, one or the other
device would stop working.

I have no problem with parallel ports and parallel port printers, for which Wang
originally engineered the port. But, let's be honest here. The disasters with
all these other devices is what gave stimulus to the industry committee that
worked out the design for USB.

I don't understand why one would go hog wild in defense of the parallel port,
unless one was either one of its inventors or if one is a troll.

And I agree with you 100%! Let's ban bicycles and all drive 18-wheelers. Great
idea! ... Ben Myers

 

[Michael Johnson]

My point is *still* that if you *need* a parallel port and the only
available slot is a PCI-X version, you use it. Bandwidth be damned!

By that reason again, bicyles should be banned because they put down
such a small footprint on the asphalt compared to an 18wheeler.

I am not saying this at all. This is your analogy. I have a PCI
parallel card in one of my computers. None of my devices could use the
added bandwidth of that PCIe card. If I didn't have a PCI slot I would
buy a USB to parallel converter as they are far less expensive than the
$80 for this device.

As for devices that can exceed the 150kb/s 'standard' parallel rate,
think ECC/ECP capable tape drives, Zip and LS120 drives and 2X CDR (on
up) burners (all external parallel). Years ago I did the math and found
speeds up to 300kb/s when the data was streaming solidly. (and yes, that
was running a Colorado 250 backup with compression turned off to get
real results.)

The cheap cards are capable of ECC/ECP transfer. Like I said, I know of
no parallel devices that can utilize 3X the speed of a standard ECC/ECP
parallel port device.

 

[nobody >]

YOU can spend the money on an extra expensive PCI-E parallel port card. I'll
spend the money on an inexpensive PCI parallel port card. In the foreseeable
future, I'll probably find more free PCI slots in computers than PCI-E slots.

Reread my posts.
I've been under the assumption that the OP only had a PCI-X(or E) slot
available. If a plain old vanilla PCI slot is available, go for the PCI
parallel card.

This is an interesting defense of the parallel port with all the bidirectional
kludge devices that were sold in the mid-90s. Every one of them was a horrific
kludge that managed to work OK if it was the only device attached to the
parallel port. But try to daisy-chain TWO devices to a parallel port, as some
manufacturers claimed was possible, and, more often than not, one or the other
device would stop working.

Been dere, dun dat!

I have no problem with parallel ports and parallel port printers, for which Wang
originally engineered the port. But, let's be honest here. The disasters with
all these other devices is what gave stimulus to the industry committee that
worked out the design for USB.

The M$ interpretation of USB printers hasn't lived up to the promises
though. I've had 98SE and XP both "lose" USB printers far more often
that parallel printers.

Case in point:
I had an old workhorse HP Deskjet 970 attached to this computer (and the
two previous ones). It has both USB and parallel interfaces. 98SE would
constantly lose the 970 on USB and would need a USB unplug/replug to
work again on the first two computers. I upgraded puter #2 to XP, same
thing. Present puter#3 was XP from the start. Still kept losing the 970.
I went back to parallel on all three computers for this printer and it
worked fine every time.

There's an HP 1315 PSC USB on #3 now, wife wanted a copymachine and
scanner. XP has lost it twice in the last 6 months. Both times it took a
USB unplug/replug to resume operation.

I don't understand why one would go hog wild in defense of the parallel port,
unless one was either one of its inventors or if one is a troll.

I'm not defending the parallel port, just defending the use of a
perfectly good parallel printer.

The 970 was free, was sitting in a box in a dusty closet at a local
garage ('Here, take this, I'm tired of moving it' deal). It has huge ink
tanks that refill easily, so operating costs are low.

 

[nobody >]

I am not saying this at all. This is your analogy. I have a PCI
parallel card in one of my computers. None of my devices could use the
added bandwidth of that PCIe card. If I didn't have a PCI slot I would
buy a USB to parallel converter as they are far less expensive than the
$80 for this device.

From personal experience, your mileage may vary greatly as far as the
USB > parallel converters. There's been a few threads here on that. I've
been burnt twice on those. I'm the "family IT Guy". Using a vanilla PCI
parallel or even a pricey $80 PCI-e card works if it saves me a round
trip of 200+ miles to redo a job that requires opening the case.

 

[Ron Hinds]

A colleague has just bought a new computer and is struggling to connect his
LaserJet 2100pcl6 (yes, I know it's old!). Because it only connects via
an LPT cable and the new machine doesn't have an LPT port, he's got hold of
an LPT-USB adaptor and plugged it into his USB port.

Windows XP Pro SP2 recognises it and prints a test page, but it won't
print anything else (such as a Word document). Under the properties he's
selected the USB port, and the print processor shows "WinPrint" and RAW
data type.

Can anyone suggest what else he might be able to do to get this bloody
thing working, or is he going to have to buy a new printer?

Many thanks.

TrentSC

I've experienced a similar problem with a LJ4200. I solved it by using the
PCL5e drivers instead of the PCL6 (but don't ask me why that worked ;-)