I was thinking of a code tweak.  There's a loop in ulpt.c (around line 
231) that
tries to find the bidirectional mode.  Simply commenting out this loop
should avoid that mode.

A different device might work too.  My usb-printer adapter works fine.

    -- Lennart

David Brownlee wrote:

>	Is there a separate device that can be used, or is it a code
>	tweak? As a data point I don't _think_ resetting the printer
>	has any effect. I've got a different usb->printer (by belkin
>	again) and will test that.
>
>	I've opened a PR on this (17948) and invite people to contribute
>	by sending email to 'gnats-admin@netbsd.org' with a subject
>	starting 'kern/17948:'.
>
>On Thu, 15 Aug 2002, Lennart Augustsson wrote:
>
>  
>
>>An idea for chasing down this bug would be to disable the bi-directional
>>mode.
>>Maybe some printers can't handle it.
>>
>>    -- Lennart
>>
>>David Brownlee wrote:
>>
>>    
>>
>>>	Unless there are any better suggestions, does anyone have a chance
>>>	to look at trying to track down when this problem started?
>>>	The best way would probably be to install a 1.5.x system then
>>>	use anoncvs to update a source tree to various dates in between
>>>	and compile a test kernel. Might want to start by generating a
>>>	-current toolchain and compiling a 1.5.x kernel with that to
>>>	eliminate toolchain issues (and confirm you have a toolchain that
>>>	should be able to compile any intermediate kernel :)
>>>
>>>	Unfortunately our ulpt using box is also formula1.com's main
>>>	internal fileserver that makes it a little hard to play with :(
>>>
>>>On Wed, 14 Aug 2002, Scott Ellis wrote:
>>>
>>>
>>>
>>>      
>>>
>>>>I have similar problems with my Brother HL1650 as well.  It probes as:
>>>>
>>>>ulpt0 at uhub0 port 1 configuration 1 interface 0
>>>>ulpt0: Brother HL-1650_1670N series, rev 1.00/1.00, addr 3, iclass 7/1
>>>>ulpt0: using bi-directional mode
>>>>
>>>>About every-other time I attempt to print, BSD simply swallows things sent
>>>>to lpd (which goes to ulpt or ulpa...same result with either one).  Usually
>>>>re-plugging the USB cable makes it start working, sometime lpc stop/start
>>>>will make it start working.  Sometimes, I get the "disabling port 1" message
>>>>too.
>>>>
>>>>
>>>>        
>>>>
>>>
>>>      
>>>
>>
>>
>>    
>>
>
>  
>