The 'Fucntion' page of UM Doctor Pro II has some interesting information:
http://www.digital-sanyo.com/BURN-Proof ... f-dvd.html
Take this quote for example:
PI Error：PI Error means Inner-code-Parity（PI）error. Normally, if the number of error is over 280/8ECC, data error correction is difficult. However, since UM Doctor Pro.Ⅱmeasures error status with a constant speed without changing the readout condition, please refer the written quality of disc as a guideline
That's all very understandable and nothing new. However the fact that they already use 250 PI / 8 ECC as the upper limit for errors implies that they've done testing with their own drives using UM Doctor Pro II to find a practical useful maximum PI error count (for 8 consecutive EC blocks):
PI：250 and over = There is a problem for data error detective amount. Although data error doesn’t occur when UncorrectablePI is not detected, data error may occur when UncorrectablePI is detected.
UM Doctor Pro II measures error status of the disc with a constant speed continuously. Therefore, if UM Doctor Pro II detects UncorrectablePI, data may be read out normally in some cases. For DVD player, if UM Doctor Pro II detects UncorrectablePI, the players may read the data normally and playback with no problem in some cases.
Although they are giving disclaimers, the upper limit of sum of 250 PI / 8 ECC seems to be their measurment guideline.
Now, this is IMHO more useful than what we can deduce currently with KProbe and LiteOn drives, where we are nowhere near understanding which levels start to be 'bad' under normal conditions for many DVD reading situations.
Maybe we should start making a test were DVD-/+R discs of varying PI sum counts are tested in DVD-drives to see if the data can still be read error free from the discs.
Even with the statistical nature of the read and variance from model to model, we could probably find some practical upper limit scale.
PS Looks like I have to buy an Optorite too :)