On 10/12/2013 2:39 AM, Baltissen, GJPAA (Ruud) wrote: > Hallo Jim, > > > So far I didn't have time to have a good look at it. > > >> If I disconnect ATN input to 6522 and ATNACK from 6522, ... > Looking at the SCH the ATN goes to two inputs, PB7 and CA1. I trap them both. > > >> (thereby tying both pins of the 74'86 to the same value), > Just to be sure, you connect pin 1 and pin2 of UD3 together. If both pins have the same value all the time, the output is always (L). The output goes to pin 11 of the 06 OC inverter UB1 and therefore its output will always be tri-stated. Yes, meaning the hardware assist will not bring the DATA line low. > > >> I can create a device not present condition. > I'm not familiar with the communication protocol of IEC drives. IEEE drives pull NDAC (L) the moment ATN becomes (L). The very first moment this is done by hardware but it is taken over by software. This is the first check for the computer that there is a device present at all. Looking at the SCH I think the data line has this function. > The first byte to be sent by the computer, while ATN is still (L), should be the device number. If it is not the number of the device, it will release NDAC. No NDAC present at all after sending the device number means there isn't any device available with this number. So IMHO the data line should be kept (L) to tell that the device is present. > > The only thing that can make the C64 hang is that the data line is made (L) at a certain point AFTER your first try. The only one who can do that is the drive. If both inputs had been disconnected, I cannot see any reason at all for the 1541 to do such a thing. But you mentioning disconnecting only one input, my interpretation, I can imagine that the 1541, acting on CA1, activates the data line in software after the moment the C64 has decided that there is no drive present. At the next access, the data line is still (L) so the C64 kindly waits until it is released. No, I disconnect both inputs. Strangely, if I reset the drive after disconnecting them (and connecting the two inputs of the '86 together, all works as expected. > Jim -- Jim Brain email@example.com www.jbrain.com Message was sent through the cbm-hackers mailing listReceived on 2013-10-14 04:00:08
Archive generated by hypermail 2.2.0.