On Wed, 17 Jun 1998, Tony Duell wrote: I wonder if implemented if it would be necessary to let the read electronics know which sensor it was reading to compensate for the rotation positional difference. Or whether when reading the flippy side it would simply wait longer to read the sector header. - don

On Thu, 18 Jun 1998, Tony Duell wrote: **** snip **** I guess what was lurking in the back of my mind as I wrote the question was the situation of trying to read a 'normal' flippy disk (with 2 jacket index holes symmetrical about the centerline) in a two sensor drive. I think that there are some exceptions to that Tony. The following is a clip from some correspondence with Chuck Guzis at Sydex a while back: "Another topic that I'd like to see some commentary on is how people have handled those old diskettes that lack the Index Address Mark information, such as those used on the Cromemco C10. On a PC controller, the first sector on a track on such a diskette usually falls to be seen by the FDC because it falls in the "blind spot" (ostensibly PLL sync-up time) of the 765-family chip. "On 5.25" diskettes, you could sometimes tape over the index hole and with some finagling, get a good read. I've heard of some people modifying a drive by moving the index-sensing LED/phototransistor assembly slightly to gain a few microseocnds. "But 3.5" diskette drives are too difficult to modify. We've had good read and write results by passing the index signal through a 1-shot carefully adjusted to trigger slightly ahead of the actual index position. But this is a very touchy arrangement, though it does work. I know that I have experienced the problem of the first paragraph on more than the C10 disks. - don