View Full Version : Teaching Spyder2


scooper750
02-09-07, 07:18 PM
Got a quick question. Colorfacts has got a training wizard. If I train my Spyder 2 with my Display LT, will the Spyder in essence be as good as the Eye-One Display LT or are there some other shortcomings with the Spyder. The Display LT isn't supported in Colorfacts so I was simply going to take my red/green/blue/white measurements in calman and see if I can import it into Colorfacts and train the Spyder 2 with it. What do you guys think?

derekjsmith
02-09-07, 07:34 PM
Yes and No :)

Yes the S2 will give the same results as the D2, for the same bulb/display type and light level for a single xyY coordinate.

For this to really work you need use a different colorspace other than XYZ because it's not linear and you need to have at least 44 data points. WRGB for 0,10,20,etc.., 100 Because the S2 and any other meter will have a different response curve.

This is why in CalMAN v2.1 we do at least 11 data points with White but that is still is not good enough.

jimwhite
02-10-07, 08:36 AM
This is why in CalMAN v2.1 we do at least 11 data points with White but that is still is not good enough.

why only 11 then ?????? :D

scooper750
02-10-07, 09:53 AM
11 is good enough to read greyscale, just not good enough to train a meter since you have to train the meter with colors also.

why only 11 then ?????? :D

derekjsmith
02-10-07, 12:23 PM
why only 11 then ?????? :D

Because that's what how it was original designed.

V3 is very different in how we calculate meter to meter and screen to PJ offsets. In fact they are not even called offsets any more we are calling them profiling.

derekjsmith
02-10-07, 12:28 PM
11 is good enough to read greyscale, just not good enough to train a meter since you have to train the meter with colors also.

You are right but we do extract the color data from the White so it's a pseudo 44 point. But this does account for non-linear meter response curves.

For V3 we have a much better and more accurate way of "profiling" this for meter to meter and screen to PJ.

Bear5k
02-10-07, 02:02 PM
why only 11 then ?????? :D
Because that was what others in the industry were doing (sort of), and we had some time-to-market issues to get through. The other aspect of this is that people do not necessarily have the patience to fully profile their meters to eliminate the non-linearities in response. We, obviously, think we have a better way now that we've had some time to build it out. ;)

jimwhite
02-11-07, 12:07 AM
Oh, Hi !!! you caught me counting my probes while I'm waiting for Ver 3 :D

derekjsmith
02-11-07, 12:29 AM
:cool: Mine is at 10, how many do we really need for calibrating :p

trueblue
02-11-07, 11:35 AM
How many.....just one thats accurate........ :)

derekjsmith
02-11-07, 01:00 PM
How many.....just one thats accurate........ :)

Very true

And can calibrate all display types now and unknown, does not need recalibration, cost less than $100 and will perform the calibration for you... wait that was the dream I had last night never mind, maybe I need to get away from the hardware for a few days :D