so what you say is "unless the chairman of the board confess his sin, it is not evidence".

As far as I recall from the judicial system, if 5 people (list in previous threads) state the same fact (in this case "using a supervisor takes care of it all") it is considered evidence.

If you refer to 1) I have had dealings with Brent and, yes, the reaction is not immediate but the reasonable complaints he has gotten from me have all, eventually, been addressed satisfactorily. That some complaints such as "this feature of the debugger, that does work, is cumbersome to use" have yet to be deemed 'reasonable' is another story. However, anything related to an actual malfunction has been addressed.random

f you refer to 2) then the fact that NO ONE have in a thread that long reported flash erasures with correctly configured and coded chips should say something.

There's no evidence presented here. Just because the chip salesman or his FAE takes your calls or, maybe even, buys you lunch, doesn't mean what he says is true or accurate. .

above you will see "been addressed satisfactorily" and "no one" what more evidence do you need? There is no reference to "what he says", only to "been addressed satisfactorily".

Further, the fact that you have to add a supervisor to a chip that apparently has features that should alleviate the need for one says something too, doesn't it?
you do not need to "add a supervisor to THIS chip" in http://www.8052.com/forum/read.phtml?id=134533 I state: "On a one uC board ... with a f124 there is no supervisor. This design is about 3 years old and 1000+ boards ...t without any memory loss". THUS, "that you have to add a supervisor to a chip that apparently has features that should alleviate" is refuted by 1000+ "witnesses".

You do not want to use chips that do not conform to your old sockets, should that allow you to judge todays state of the art by what is true for yesterdays chips?

Erik