LivinLoud, you got your Ophir sensor from clcen1 on ebay, correct?
A while back Jerry and I both did some testing on Ophir heads and found that there was some level of variability among the sensors from ebay. Most are within 3%, but we've found some as high as 8-10%. That's why ARG makes sure each Ophir sensor that comes bundled with a Rubicon is on-spec.
In Jerry's defense, my four-year-old LaserBee I read within 1% of a NIST-traceable LPM without the need for recalibration. Some LaserBee models might be slow, but for the most part are quite accurate.
I can't speak for the LPM Lazerer uses (I think it's a LaserBee?), but the slow response time of the TEC might be covering up a short-lived peak.
Like you said, if the laser in question is a DPSS laser, it's probably the testing conditions.
Trevor