ferdimh@gmx.de wrote:
Are you saying that you try to build a software program to record all read pulses from the IBM-PC parallel port? x86 machine is too fast to capture all read pulses from the Apple II disk and put all read pulses into binary image. Flowing read pulses through read signal is limited to 2 MHz.
I don't try to build it, I just suggested it. And yes, the x86 machine IS much too fast. But if I let it do some nonsense calculations, one can slow it down to any speed wanted. So it is possible(IMO) to do a much better timing than the original apple ] [ with its fixed clock. PS: the data to the disk drive is clocked at 250kHz, not 2 MHz.
I think it is usually a 2MHz clock for an MFM DD controller so it can do precompensation. I don't know if the GCR data uses precompensation or not. As bits (magnetic transitions) get closer together their position on read shifts. To correct for this some are shifted either slightly early or slightly late. The PC/AT floppy controller has a 24MHz clock so it can do 250kb/s, 300kb/s, or 500kb/s bit rates. That is unlike hard disk controllers which tend to only need a clock twice the data rate. Also, the UART for a serial port needs a clock at 16 times the data rate. -- glen