Re: VIGC study on spectrophotometers
Re: VIGC study on spectrophotometers
- Subject: Re: VIGC study on spectrophotometers
- From: "edmund ronald" <email@hidden>
- Date: Wed, 17 Sep 2008 20:53:52 +0200
Raymond,
Telling each user that inter-instrument agreement is his
responsibility is very sensible, I'm sure. Layiing on supplies of
isoproyl alcohol is easy too. But please explain how I'm supposed to
match backings across say iSis, DTP70 and i1 ? I won't insult you by
asking about other instruments from other manufacturers ...
Edmund
On Wed, Sep 17, 2008 at 6:36 PM, RaymondCheydleur <email@hidden> wrote:
> Ahh Fear Uncertainty and Doubt.
>
> Having 3 different reports like this surface in the last few months, I felt
> moved to comment, I've given it the requisite 2 hour cooling off period in
> my drafts folder and I still like it, so here it goes... Don't take this as
> the official stance of my employer but as my personal observances.
>
> I'd also like to thank Danny for his input (received during the 2 hour
> cooling off period) as it appears to be a reasonably well controlled
> experiment, easily understood and thus informative, which is refreshing. I
> think it also is still in agreement with the following, understanding that
> one data point from each instrument type does not tell us all we would like
> to know.
>
> Instruments are clearly not perfect and work is ongoing both from a
> manufacturing standpoint and in Standards work to improve them in many ways.
> As an example on the standards side, ISO TC130 is meeting next week and
> includes proposals that will help instrument data interchangeability (CxF)
> and instrument repeatability (Measuring conditions, illuminants/filtration,
> M0-M3).
>
> However
>
> This study like many others comes to some conclusions that are pure common
> sense: Newer instruments tend to be tighter to each other than older
> instruments. Using a visual tolerance (their suggestion is DE2000) the "big"
> differences being reported essentially disappear. (I say "big" because 4-5
> years ago much larger differences were considered unavoidable system noise,
> and this is why visual certification ruled the day.
>
> I'd suggest however that if tight tolerances of your measuring device are
> really a concern to you (no matter what your tolerancing preference) a few
> basics will almost always improve your results:
>
> - Calibrate - carefully
> Make sure your reference is clean, and for some instruments within
> expiration dates. Take care in the placement of the instrument to the
> calibration reference (if this is a manual process) The foundation of all
> your measurements is based on this one reading.
>
> - Recertify annually
> If you don't do this you are part of the problem not part of the solution.
>
> - Match the target to the device
> Don't make it smaller than the instrument recommends to fit your desires and
> then complain about repeatability, please.
>
> - Use a standardized backer that fits CGATS.5 or ISO 13655.
> Self backing will generally increase noise in the data - sometimes
> dramatically, sometimes only a small amount.
>
> - If your device has a measurement guide, use it.
> Generally guides increase repeatability and accuracy of data in a noticeable
> way.
>
> RayC
>
>
> Ray Cheydleur
> OEM Project Manager
> email@hidden
> X-Rite Incorporated
>
>
>
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> Colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
> This email sent to email@hidden
>
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden