Well, there isn't. :-) The cursor is generated by a microcode task. It does appear very early in the boot process, however. There are some good scanned documents about the Alto on Bitsavers -- Ian ________________________________________ From: cctalk-bounces at classiccmp.org [cctalk-bounces at classiccmp.org] On Behalf Of Nick Allen [nick.allen at comcast.net] Sent: Tuesday, October 19, 2010 10:14 PM To: cctech at classiccmp.org Subject: Updates on repairing Xerox Alto There must be some way to get the hardware cursor to appear even without booting to ethernet or the disk drive... I do have a disk drive, controller and disk packs, but as you mentioned it requires almost all the boards to be working, which may be very difficult if I can't gradually get things working and tested (such as the display board). I have however confirmed with an o-scope that I am getting the correct video output from the video board, so I assume that board is working properly. Now about a dozen more boards to go!

Provided the PSU is good (and you have checked that, right :-)), I can't see that plugging all the boards in can do any damage. OK, when it doesn't work, you don't know immeidately where the problem is, but it shouldn't be all that hard to identify sections that are working. Somebody mentioned the trick of tracing the microcode with a logic analyser. This is something I've done many times (not on the Alto, since I don;t have one, but certainly on PERQs and HP9800s). Assuming you have a sorce listing of the microcode, this will tell you a lot about what it's doing or not doing. I understand that the Alto doesn't alas, have a test connector carrying the microcode address bus (the 2 machines I mentioned both do). So you have to pick the signals off the backplane If you don't ahve a logic analyser, then another trick that can help is to connect the microcode address lines to one set of inputs of an <n> bit comparator. The other set of inputs comes from a suitable set of DIP or thumbwheel switcehs. Feed the 'equals' output of the comparator to a logic rpboe or a monostabel and LED. The idea is that while you can't trace the microcode, you can at least see if certain routines are being executed (set the address of the start of the rotine on the switches, if the logic brobe is triggered, then the microde is gettign to that address). Does it ever run any of the video routines, things like that. -tony