Re: comedi_to_phys

It's not about the resolution. It's just about the NAN after the 
conversion. But you are right. The user can use the raw data. I'll 
change maxdata to 4096.

/Bernd

www:    http://www.berndporr.me.uk/
         http://www.linux-usb-daq.co.uk/
Mobile: +44 (0)7840 340069
Work:   +44 (0)141 330 5237
         University of Glasgow
         Department of Electronics & Electrical Engineering
         Room 519, Rankine Building, Oakfield Avenue,
         Glasgow, G12 8LT


Frank Mori Hess wrote:
> On Monday 28 February 2005 09:24 am, Bernd Porr wrote:
> 
>>Dear all,
>>
>>does it make harm if we change the "data==0" into "data>=0" in
>>comedi_to_phys?
>>
>>The reason is that I use bit #12 to signal an out-of-range with a 12 bit
>>A/D converter. Bits 0 to 11 carry the data. 0..4095 is the "normal"
>>range and values from 4096 to 8191 signal out of range.
>>
> 
> 
> The driver shouldn't return any values larger than maxdata.  The user isn't 
> required to use comedi_to_phys() after all, or may have changed the the 
> oor behavior.  Plus any scheme you come up with won't be useful in general 
> since 16 bit cards don't have any spare bits to play with.  By having a 
> seperate out-of-range bit, you are only adding two more useful values, 
> from 4094 to 4096 on a 12 bit card.  That's only 0.02% improvement in 
> resolution.
> 

Received on 2005-03-19Z20:00:39