[argyllcms] is my colorimeter broken?

Hi all,
quick question: Is there any way to detect whether my i1d2 is broken other than comparing its results against known good results?
I'm just wondering if it might be broken, because every display I have hitherto profiled was very weak on the red end. I'm sitting in front of this new 24" PVA panel and get easily beyond 1M cubic units of gamut but compared to sRGB those are very far on the blue side and lacking on the red side. 

For example:

http://www.jarre-de-the.net/computing/t6500-b120.icc
http://www.jarre-de-the.net/computing/t5000-b120.icc
http://www.jarre-de-the.net/computing/b120.icc

and visualized vs. sRGB:

http://www.jarre-de-the.net/computing/t6500-sRGB.wrl
http://www.jarre-de-the.net/computing/t5000-sRGB.wrl
http://www.jarre-de-the.net/computing/b120-sRGB.wrl
Using any of the whitepoint corrected profiles makes the display rather reddish and dark (hard to look at). Using the brightness@120 profile with a gamut mapping for sRGB, I can configure that profile in photoshop with perception intent, load an sRGB image, enable color proofing and get "close" to my printout, though the highest saturated red will still look odd.
So back to my original question: Is there an easy way to find out whether I can trust my color readings? Or (since I couldn't do this on the machine I intend to use), is it safe to assume that a LUT based profile created on one machine with an NVidia graphics adapter (on Linux) can be used on another with a different NVidia graphics adapter (on MacOSX)? 

TIA,

Karl.

Other related posts: