I have tested *ist D, *ist DS and Canon 300D and those tests are published in my geocities homepage.
I have also tested the ambient light metering systems as well as the P-TTL exposure accuracy of the K100D:-
RiceHigh's Pentax K100D Full Review
Furthermore, I have owned and used 3 units of *ist D body and 3 copies of the DS plus a K100D and a borrowed K10D - all of them are behaving the same, and exactly as what the thread open poster describe, if one really to compare the Pentax to other actually more superior systems in the metering department, namely, Canon and Nikon.
My opinion is simply that it is a fact for Pentax has an inferior metering and exposure system, no matter for ambient light (fixed pattern or not) and the flash system. Of course, you can have a different opinion on that. But just ask yourself, how did you compare different systems and have you actually ever been able to compare them in an objective way?
Originally posted by c.r.brown Fair Warning Notice: None of those tests show a control group in order to show that the tests are indeed valid, these tests seem to be done with a single istD camera with no mention of wether it had been certified or checked by Pentax beforehand, wether it was a NIB model or other. Neither does it mention what firmware version was used or if it was retested with subsequent firmware revisions. Nor do these specious tests correlate to the K100D or K10D cameras and their associated firmware versions.
And lastly that these tests were 'done' on JPEGs, the entire test conclusion if entirely valid could then be an issue with the internal conversion software from RAW to JPG and not the metering of the camera. if you shoot in RAW you nullify any argument ricehigh may pretend to make.