Jim, That's an interesting and colorful narrative, and much of it makes good sense - in the giddy days of the 1980s home computer market, and for CBM in particular. But speaking as one who was fairly hot & heavy into the VIC-20 (and then almost immediately the C-64), this 'drive ID' issue has more-or-less dogged us since day one.
While it's true that +most+ end-users couldn't or didn't buy a second 1541, it was very common to connect two or more drives to a single machine for demonstrations, software shows, user group meetings or what might have been the 1980's equivalent of the LAN party.. where we shared & swapped software. And on top of that, as time wore on, there were a number of programs that supported multiple drives. So we can agree that Jack, being Jack, had his reasons. But ultimately, I still see this decision as short-sighted. There's a reason that the circuit board jumper pads were well-publicized, as were the many text files on adding device ID switches to the 1541. On Sat, Jan 23, 2016 at 7:33 PM, Jim Brain <br...@jbrain.com> wrote: > On 1/23/2016 7:15 PM, drlegendre . wrote: > >> " I am saying don't make a permanent hardware change to a 1541 that >> does not have the switches unless you really want it to be permanent >> because there is a software method of assigning drives that is good enough >> most of the time. BUT if you must make it permanent and you don't have >> the >> external switches, consider adding some form of external switch so you >> don't ever have to open the case again to put it back to the default." >> >> Well then, we're having a major agreement. ;-) >> >> The device ID switch is the ultimate fix for Commodore drives, and I'm >> really not sure why CBM didn't incorporate one into the design - at least >> from the 1540 on upwards. Can't have cost much to add a discretely located >> access hole (or a knock-out) in the case, along with a 2-place DIP switch >> for controlling device ID. >> > But, they did. The 1541-II, 71, and the 81 have switches. > > I would disagree on your point that Commodore should have made it part of > the design... > > Let's travel back in time. > > After the PET intro, Peddle designs a drive, a beast of a device, with 2 > CPUs and it costs a fortune. Peddle is convinced a smart drive is best, > and the delay allows other manufactures to create "dumb" drive options (saw > one at World of Commodore, forgot the name). > > Tramiel is mad, and the drive doesn;t sell terribly well at first. > > When the VIC-20 comes out, the drive is just marketing materials. Why > would someone want a business storage system for a home computer basically > built because Jack needs to unload a bunch of VIC-I ICs that he can't sell > to video game firms? > > But, people asked for it, and CBM delivered, taking Peddles IEEE design, > ditching the IEEE cabling (Jack: "Get OFF that bus", as the cables from > Belden/Beldin(sp?) cost a mint and supplies had been spotty at times). The > unit is retooled to use 1 drive, 1 CPU (and in the process the changes > introduce the fabled "save with replace" bug. Enter the 1540 > > But, the designers put little jumper pads on the drive to set dev number, > as they were very sure most people would never buy two of them. > > Then the 64. Paddle's ColorPET design loses out to a hacked VIC-20, which > became the 64 > > Now the drive seems more useful, and can be made more cheaply. As well, > other units in the marketplace have drives, so it is a competitive need. > All is well, though, as the 1540 is already in production. Dump some gray > tan/gray colorant in the injection mold machines, and you're done. That's > classic Jack. > > The 1540 has a problem though. The bit banging protocol (introduced late > in the 1540 design cycle after it was determined that the VIC shift > register has issues) fails on the 64. high/low times are 20uS/20uS, and > sometimes the CPU is stopped for 50uS on the 64. Enter the 1541. No > changes except the ROM, which slows down sends from the drive to the 64 to > 60/60uS clock. Jack is done. He doesn't care if people buy them per se, > just that the drive is there so he can tout it being there as he competes > with the Apples and Ataris. > > So, at the time, with the drive being as much as the computer, no one > thought people would buy more than 1, and thus no need to pay for extra > parts (switches) and changing the molds. > > With the C128 and the case redesigns, people were buying multiple drives, > and it was a selling point, so the switches made their appearance. > > Jim >
