plustek: Thread: if epson esc/p2 driver loaded, some wrong colours may be in the scanned image

Hello,

I compiled another kernel and then it was possible to load the pt_drv module.
The standard SuSE EIDE Kernel was it, was made the problem in my last question. 

So, I have also an epson stylus pro xl printer using the ESC/P2 driver on my
parallel port connected. If the lpr service is loaded, I have sometimes the
problem, that in parts of scanned images are some colour sets wrong or missing.
The scanner (4830) stops on some parts of the scan (what is usual) and
continues the scan then. After a stop sometimes the next part is scanned with
the wrong colours and after it stopped again, the correct colours are used
again. 
If the lpr service is not loaded, it scans everytime right and makes no
problems. 

On Fre, 14 Jul 2000, you wrote:
> Hello,
> 
> I compiled another kernel and then it was possible to load the pt_drv module.
> The standard SuSE EIDE Kernel was it, was made the problem in my last question. 
> 
> So, I have also an epson stylus pro xl printer using the ESC/P2 driver on my
> parallel port connected. If the lpr service is loaded, I have sometimes the
> problem, that in parts of scanned images are some colour sets wrong or missing.
> The scanner (4830) stops on some parts of the scan (what is usual) and
> continues the scan then. After a stop sometimes the next part is scanned with
> the wrong colours and after it stopped again, the correct colours are used
> again. 
> If the lpr service is not loaded, it scans everytime right and makes no
> problems. 

Hi Paul,

I encountered similar problems using an Epson Stylus Color and the OP9636T.
Here it happens that you only hear an awful noise comming from the motor.
Since now it is quite not clear why this is so. I'm for sure that it has
something to do with the parport module and that the driver does not get all
data (on the OP4830) or can't send all data (on the OP9636T). This has to be
tested for future releases. I think this is not only a problem to Epson
printers, this is a common problem to all connected printers. 

I must admit that this datapath is not well tested yet. I'm looking for 
volunteers to check the code and provide some ideas or patches
(Kevin, what about you ?!)

Bye
Gerhard