On Fri, 25 Jan 2013, Jerome H. Fine wrote: AND, . . . once the disk is formatted, WD and NEC style controllers can usually read/write a formatted disK (unless it's an HP disC) without needing to see index. Therefore only the drive used for FORMATing needs the modification. 'course not have a functioning index can play havoc with the guesses that the FDC makes about what is wrong when there is a problem. INT13h may return an error code 128 (time-out/drive not ready) when an actual problem might be something else. -- Grumpy Ol' Fred cisin at xenosoft.com

On Fri, 25 Jan 2013, Jerome H. Fine wrote: What "SS/DS bit"? MOST systems do not have any such hardware. They SELECT a head, as a signal FROM the controller TO the drive, and then read/write, etc with no side feedback from the drive. <Oversimplified> It does not tell the hardware where the first sector is located. It tells the "hardware" and the software where the track starts. The software then ON MOST SYSTEMS assumes that the next sector is the first one. On MOST SYSTEMS: When a READ (or WRITE) command is processed, the head could be anywhere along the track. Reading begins, watching for the desired sector header. If it is encountered, then the READ (or WRITE) is done. It may or may not be found before the index pulse. After the index pulse, NEC controllers are momentarily "blind", so any sector headers that are too soon after index (such as some written by WD controllers) are inaccessible. After several index pulses without finding the desired sector header, the controller declares "Sector Not Found" (Error code 4 of INT13h). If NO sector headers were seen, due to blank, damaged, or wrong density media, then INT13h gives an error code #2 ("General Failure" aka "NONE OF THE ABOVE"). If, after an excessive amount of time without seeing an index pulse, then error #80h ("Time Out" of "Drive Not Ready") If there is a successful read before Index, or with NO index, but before Time Out, then NOBODY CARES. Therefore, in normal use, index IS needed during FORMAT, to specify where to start the track (not bothered with on Apple][, or Commodore). But index is not needed for READ nor WRITE unless other problems occur. Some Teac 55 series drives rely on index internally to the drive, and therefore choke on lack of index. Option Board software relies on index, and will choke without it. </Oversiplified> -- Grumpy Ol' Fred cisin at xenosoft.com

I just looked up the dates on the modifications made by DSD to the DY(X).SYS device drivers in RT-11. The files on the 8" floppy disk supplied with the documentation for the DSD 880/30 range has the symbol defined as: DBSID2 = 2 The DEC RX02 does not make this bit available in the CSR since there is no second side. However, the file DY.MAC from the V04.00 binary distribution in 1980 contains at least some of the code that would be needed to support an RX03 DSDD drive. It does not look as if that code was ever properly tested since I remember noticing at least one mistake when I enhanced the DY(X).SYS device driver code supplied with V05.00 binary distributions to support the RX03 8" floppy drive on the DSD 880/30 that I had modified with the DPDT switch I made mention of in an earlier post. Fred, does that clarify the "SS/DS bit" reference? Up until 1985, the DEC PDP-11 had hardware and software which far surpassed any PC hardware and software. Of course, it was also much more expensive, probably overpriced and the volume relatively low, to some extent the direct result of the high cost. Jerome Fine

On 1/26/13 8:36 PM, Jerome H. Fine wrote: It seems the Xerox 820 was pretty pricey too at the time. With the configuration I have, (DSDD floppy drive and 10 mb. hard drive), it was around $9200.00 with the stock software package and all the hard copy documentation. Pretty steep for a desktop machine even then. :-) Dave Land Land Computer Service