On Thu, Feb 24, 2000 at 10:43:31PM +0100, Ingolf Koch wrote:
> On Sun, Feb 20, 2000 at 09:22:01PM +0100, Ingolf Koch wrote:

> > compared with PAR: (AmigaOS), /dev/par0 seems to be _very_
> > slow. At least the printer connected to the parallel port
> > prints the _same_ file much faster with COPY file PAR:
> > than with cp file /dev/par0 (on -current NetBSD/amiga).
> > 
> > Is there something I should configure before accessing
> > /dev/par0?
> 
> To be more precise: 20 cps with /dev/par0. Is it really
> that slow? It turns out that the while() loop in parsend()
> (which waits until old data has been read by the printer)
> is almost never executed.

Sorry, I got distracted. 

There seems to be an ioctl, legacy from the hp300 printer driver, that can
tune the parameters. I'll ask whether there is a control program for this,
or whether we should write a new one.

Actually, I'll just cross-post. "HP300ers out there: is there a control
program for the 

#define PARIOCSPARAM    _IOW('P', 0x1, struct parparam)
#define PARIOCGPARAM    _IOR('P', 0x2, struct parparam)

ioctl(), which seems to have its origin in the hp300 port?"

Regards,
	Ignatios Souvatis