On 16 Sep 97 at 1:24, Antti Lukats wrote: > At 07:00 AM 16/9/97 -0600, you wrote: > >The new Scenix chips are supposed to comply with some UL 98 standard that I've done a bit of a search for this, or similar standards, on the web. Anyone got a url for it? > >requires the microcontroller to verify the integrity of the code. Does anyone > >have the details of this standard? Several microcontrollers allow access > to the > >ROM code, so a checksum is possible. However, how do I know an ROM error does > >not exist in the code that I wrote to check the ROM? Will the new Scenix chip To make it truly safe, it should be in real rom, that just runs thru the eeprom and compares it with a crc in a known site. > >have a hardware checksum circuit? > >How important is this checksum? Question is - Would working in a faulty manner be better or worse than not working at all? Follow up question - how is the 'not working at all' implemented? examples- Space probe - better to work faulty - or - request a d/l to ram Pacemaker - better to fall back to a 'metronome' cct. ABS - disable - go to normal braking - warning light Traffic lights - disable MikeS (remove the you know what before replying)