Joel wrote: >>>>Generally, in my experience, intermittent errors can sometimes occur on >>>>a CS cable with the drives jumpered as master/slave. I have never had a >>>>problem with the drives set to CS though. >>> >>> >>>I wonder if this might be due to selecting a drive as master but hanging >>>it in the middle. I'm thinking that floating input could cause the >>>controller to miss selects forced against the pattern, and suspicious >>>that manufacturers may not be testing the controllers, cables, and/or >>>drives with slave and master strapped, but master in the middle. >> >>Well how do you explain in binary logic the float thing ? > > > You're gonna love this -- > > I haven't seen the spec, but one typical way to use float as a state is > to check the pin's state first, then try driving it to the opposite > state and checking the result, and then try driving it again to the > original state and checking again. If the pin is strapped, it will > refuse to go to one state. > > IIRC, IIRC ? what does this mean ? there are also gates which can detect tri-stated inputs under > specific engineering parameters, as well. I remember looking at the > diagrams and thinking, "Man, some engineer has been drinking too much > coffee." > > Anyway, no, it's not binary logic. Well i haven't seen the spec either but possibly i will agree with you , when you say that some engineer must have drunk too much coffee/tee whatever that was . BTW : My cabling on this machine ( the Athlon XP one ) is non-cable select, the other machine ( the Pentium 166 MX one ) has o Promise controller in order to be able to see the 160 GB hard disk which was shipped with a cable select cable . Do you know if there is a difference in performance between the Cable Select and the non-cable select one ? > -- > Joel Rees <rees@xxxxxxxxxxx> > digitcom, inc. 株式会社デジコム > Kobe, Japan +81-78-672-8800 > ** <http://www.ddcom.co.jp> ** >
