[argyllcms] Re: Display calibration with Eye-One Display 2
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 16 May 2007 19:32:29 +1000
Tilo Nitzsche wrote:
Under Windows XP, dispread can profile the laptop display fine. However when
trying to calibrate a Dell 1905FP (19" TFT, Samsung PVA panel), it hangs for
maybe 30s when trying to read dark patches and then times out. (dispread -v
-yl -K -p 0.5,0.5,2.0 dell1905)
Was this also under XP, or was it Linux ?
It's hard to explain this problem. The closest I can get is that indeed
there is an issue with instrument commencing measurement before the
new patch is stable on the display. There are two phases to
each Display 2 LCD read operation, the first being a preliminary
reading to establish a scale factor, followed by the real reading
using that scale factor. The scale factor and the light level sets
the time it takes for the second reading to complete.
So if there was a preceding bright patch that was still being displayed
as the preliminary reading was being done, there would be a high
scale factor (integration target threshold ?) set, and then if during the
second phase the displayed level drops to black, the instrument
won't respond for some time (minutes ?), because the integration
will take a long time to reach the high threshold.
The preliminary reading takes about 120 msec from start to finish,
so if this is the explanation, it's hard to guess how much delay
there is from setting a patch color to it actually turning up on
the display.
> My impression was that i1 Match and basICColor Display don't display dark
> patches longer than others (using the same Eye-One Display 2). So perhaps
> there is some reading mode that uses a fixed measurement time.
They usually take between 0.5 and 7 seconds. I think there is a bug
in the Gretag driver that sometimes causes even longer measurement
times than this too.
I'll see about uploading a version of the code (V0.70 Beta 4) that
adds a 50msec delay before doing a display reading, so that you
can see if this makes any difference to the problem.
Graeme Gill.
Other related posts:
