VE and BLM cell edges

[Bob Wooten]

Cool, sounds to me like adequate performance can be had w/a small amount of
processor.  this being evidenced by the small number of i/O readings & the
available processor time.  I would also think that noise is not a huge
issue.  if it were a simple solution to that would be to increase the
number of readings taken/sec & then average them to a rolling number of
some time constant & if need be an adaptive filter can be employed. 
anywho......

thanks Dave, more & more the picture is becoming clearer.  like when you
are on the net & have loaded a new web page.  the first few seconds are
just colors.  then some shapes.  then some images.  then they become
defined & now you can see edges to the pictures.  I am somewhere between
colors & shapes but the picture is getting clearer.  

BW 


> [Original Message]
> From: Dave Zug <dzug at delanet.com>
> To: <gmecm at diy-efi.org>
> Date: 1/20/01 12:50:07 PM
> Subject: Re: VE and BLM cell edges
>
> Canister enable has to be a very low priority item, I belaive the updates
> for the most important sensors is still not taking up near the time that
the
> processor clocking is capable of, thats a conclusion based on seeing 12.5
> msec being used for some important updates, the processor is going waay
> faster than one pass per 12.5 msec. I think the most taxed resource of the
> P4 is the onboard ram, the processor has plenty of speed to spare I think.
> the I/O stuff is even averaged or 'softened' so it seems truly accurate
> readings are not even a priority. noise may influince the reasoning for
> softening the readings too, I think I read posts about that.
> 
> ----- Original Message -----
> From: Bob Wooten <r71chevy at earthlink.net>
> To: <gmecm at diy-efi.org>
> Sent: Friday, January 19, 2001 9:38 PM
> Subject: Re: VE and BLM cell edges
> 
> 
> > Dave,
> >
> > does this not also relate to processor time & the whether or not the
data
> > point in question NEEDs to be an instantaneous value or if historical
data
> > of X age is acceptable for the calculation?
> >
> > BW
> >
> >
> > > [Original Message]
> > > From: Dave Zug <dzug at delanet.com>
> > > To: <gmecm at diy-efi.org>
> > > Date: 1/19/01 3:19:58 PM
> > > Subject: Re: VE and BLM cell edges
> > >
> > >
> > > That is called efficient coding! If some condition is already checked
> for
> > as
> > > part of another process (like the canister purge as discuvered here)
> then
> > > other processes only have to reffer to the result, not execute the
check
> > > again.
> > >
> > > This of coarse makes code hard to follow when not looking at its
> > > complimentary documentation. Keeps most of us from going the extra
steps
> > to
> > > find out more,  you are a rare breed Bob ;-)
> > >
> > >
> > >
> > > ----- Original Message -----
> > > From: <rrauscher at nni.com>
> > > To: <gmecm at diy-efi.org>
> > > Sent: Friday, January 19, 2001 10:07 AM
> > > Subject: Re: VE and BLM cell edges
> > >
> > > > However, some of the canister purge qualifications are very
> > > > similar to what would be idle: if tps% < 3% or mph < 1.8, then
> > > > no canister purge. Then cell 4 is selected. There are some
> > > > other quals such as coolant temp and map (set to 0).
> > > >
> > > > BobR.
> > > >
> > > >
> > >
> > >
> > >
> >
--------------------------------------------------------------------------
> --
> > > To unsubscribe from gmecm, send "unsubscribe gmecm" (without the
quotes)
> > > in the body of a message (not the subject) to
> majordomo at lists.diy-efi.org
> > >
> >
> >
> >
> > --- Bob Wooten
> > --- r71chevy at earthlink.net
> > --- www.r71camaro.homestead.com
> >
> >
> >
--------------------------------------------------------------------------
> --
> > To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
> > in the body of a message (not the subject) to
majordomo at lists.diy-efi.org
> >
> >
> 
>
----------------------------------------------------------------------------
> To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
> in the body of a message (not the subject) to majordomo at lists.diy-efi.org
> 



--- Bob Wooten
--- r71chevy at earthlink.net
--- www.r71camaro.homestead.com 


----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org