From rodebshan at comcast.net Sun Jun 4 18:38:25 2006 From: rodebshan at comcast.net (Rodger) Date: Sun, 4 Jun 2006 14:38:25 -0400 Subject: [P112] P112 Partially working!? Message-ID: <4E8169F6-F3F9-11DA-BD5C-0050E4D00558@comcast.net> With the week off, I've finally had time to assemble my P112. It is working at least partially, and I was hoping someone would have some ideas about the problems. First, the bios reports a processor speed of "??-Unknown-??". The crystals are in the correct position, and seem to be on frequency. Second, the "Reset" switch does not work properly, when pressed, the P112 seems totally dead to the world, and the only way to recover is to power down for an extended period (over 15 seconds). Third, after setting the NVRAM settings, even accepting all the defaults, the P112 hangs at the "(Configuration Complete)" message, and once again requires an extended power down to recover. When rebooting, I get a plus/minus sign instead if an equal sign, and the P112 ignores all input (appears to hang). I have tried changing the baud rate of my terminal emulator, with no effect. Now for the catch to the whole thing. If I leave the NVRAM uninitilized, and give it a boot floppy, it will boot and function correctly, and it seems to run like a champ, as far as my limited testing has gone. Any clues!? Rodger! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 1143 bytes Desc: not available URL: From terrygski at cfl.rr.com Sun Jun 4 21:15:42 2006 From: terrygski at cfl.rr.com (Terry Gulczynski) Date: Sun, 4 Jun 2006 17:15:42 -0400 Subject: [P112] P112 Partially working!? In-Reply-To: <4E8169F6-F3F9-11DA-BD5C-0050E4D00558@comcast.net> Message-ID: <000601c6881c$09ebada0$6401a8c0@terry1> Comments interspersed. -----Original Message----- From: p112-bounces at lists.feedle.net [mailto:p112-bounces at lists.feedle.net] On Behalf Of Rodger Sent: Sunday, June 04, 2006 2:38 PM To: p112 at lists.feedle.net Subject: [P112] P112 Partially working!? With the week off, I've finally had time to assemble my P112. It is working at least partially, and I was hoping someone would have some ideas about the problems. First, the bios reports a processor speed of "??-Unknown-??". The crystals are in the correct position, and seem to be on frequency. This is fairly odd. The init code in the ROM uses the Z80182's CTC to count while outputting a character to an internal serial port. This timer count is then used to determine the system clock speed. While it's not perfect, it does work pretty well, and I've not seen one that was off before. The entire process takes place inside the Z80182 - the only components used that are external to the Z80182 are the CPU clock and memory. Although the Z80182 could be bad, I'd first concentrate on the clock and RAM. Can you swap out the RAM chips? If not, you might try simply swapping the RAMs between sockets - a bad RAM can give very odd symptoms, depending on which cells are bad. In this case, I'm assuming the (potentially) bad RAM is in context with the ROM. Swapping RAM chips between sockets will put the other RAM chip in context with the ROM. See the discussion below re: the NvRam being un-initialized. Because of those comments, I believe your clock is working correctly, as is the Z80182. Second, the "Reset" switch does not work properly, when pressed, the P112 seems totally dead to the world, and the only way to recover is to power down for an extended period (over 15 seconds). If there IS a bad RAM in the system - or a 'marginal' one - and the P112 cannot determine if RAM is present, it will execute a HALT instruction. If this happens, a logic probe will show that all address and data lines are in a High-Z state - floating, as it were, neither HIGH nor LOW. Third, after setting the NVRAM settings, even accepting all the defaults, the P112 hangs at the "(Configuration Complete)" message, and once again requires an extended power down to recover. When rebooting, I get a plus/minus sign instead if an equal sign, and the P112 ignores all input (appears to hang). I have tried changing the baud rate of my terminal emulator, with no effect. I'm gonna guess that the 'hang' is due to a bad baud rate rather than anything else. Baud rates are set as a function of the CPU clock that was determined during initial power-up. This is done via a 'look up' table that is based on the CPU clock. Since the system cannot determine the CPU clock rate ("??-Unknown-??" mentioned above), it cannot set a baud rate because it can't figure out which table to use. Now for the catch to the whole thing. If I leave the NVRAM uninitilized, and give it a boot floppy, it will boot and function correctly, and it seems to run like a champ, as far as my limited testing has gone. With NvRam un-initialized AND AN UNKNOWN CPU Clock speed, the P112 uses 'default' values for everything, including 9600 baud divisors that are BASED ON A 16MHz CLOCK. Since your system DOES work correctly with NvRAM un-initialized, I submit that the clock IS working correctly, and at the correct frequency of 16MHz. Any clues!? RAM. Best of luck. Terry -------------- next part -------------- An HTML attachment was scrubbed... URL: From terrygski at cfl.rr.com Mon Jun 5 02:11:12 2006 From: terrygski at cfl.rr.com (Terry Gulczynski) Date: Sun, 4 Jun 2006 22:11:12 -0400 Subject: [P112] P112 Partially working!? In-Reply-To: <000601c6881c$09ebada0$6401a8c0@terry1> Message-ID: <001401c68845$518772b0$6401a8c0@terry1> Sorry Gents - I don't know how I screwed my previous reply so badly - something to do with HTML. Here it is again with the correct formatting. Comments interspersed ? -----Original Message----- From: p112-bounces at lists.feedle.net [mailto:p112-bounces at lists.feedle.net] On Behalf Of Rodger Sent: Sunday, June 04, 2006 2:38 PM To: p112 at lists.feedle.net Subject: [P112] P112 Partially working!? ? > With the week off, I've finally had time to assemble my P112. It is working at least > partially, and I was hoping someone would have some ideas about the problems. > First, the bios reports a processor speed of "??-Unknown-??". The crystals are in the > correct position, and seem to be on frequency. This is fairly odd.? The init code in the ROM uses the Z80182?s CTC to count while outputting a character to an internal serial port.? This timer count is then used to determine the system clock speed.? While it?s not perfect, it does work pretty well, and I?ve not seen one that was off before.? The entire process takes place inside the Z80182 ? the only components used that are external to the Z80182 are the CPU clock and memory. Although the Z80182 could be bad, I?d first concentrate on the clock and RAM.? Can you swap out the RAM chips?? If not, you might try simply swapping the RAMs between sockets ? a bad RAM can give very odd symptoms, depending on which cells are bad.? In this case, I?m assuming the (potentially) bad RAM is in context with the ROM.? Swapping RAM chips between sockets will put the other RAM chip in context with the ROM. See the discussion below re: the NvRam being un-initialized.? Because of those comments, I believe your clock is working correctly, as is the Z80182. > Second, the "Reset" switch does not work properly, when pressed, the P112 seems totally > dead to the world, and the only way to recover is to power down for an extended period > (over 15 seconds). If there IS a bad RAM in the system ? or a ?marginal? one ? and the P112 cannot determine if RAM is present, it will execute a HALT instruction.? If this happens, a logic probe will show that all address and data lines are in a High-Z state ? floating, as it were, neither HIGH nor LOW. > Third, after setting the NVRAM settings, even accepting all the defaults, the P112 > hangs at the "(Configuration Complete)" message, and once again requires an extended > power down to recover. When rebooting, I get a plus/minus sign instead if an equal > sign, and the P112 ignores all input (appears to hang). I have tried changing the baud > rate of my terminal emulator, with no effect. I?m gonna guess that the ?hang? is due to a bad baud rate rather than anything else.? Baud rates are set as a function of the CPU clock that was determined during initial power-up.? This is done via a ?look up? table that is based on the CPU clock.? Since the system cannot determine the CPU clock rate (???-Unknown-??? mentioned above), it cannot set a baud rate because it can?t figure out which table to use. > Now for the catch to the whole thing. If I leave the NVRAM uninitilized, and give it a > boot floppy, it will boot and function correctly, and it seems to run like a champ, as > far as my limited testing has gone. With NvRam un-initialized AND AN UNKNOWN CPU Clock speed, the P112 uses ?default? values for everything, including 9600 baud divisors that are BASED ON A 16MHz CLOCK. Since your system DOES work correctly with NvRAM un-initialized, I submit that the clock IS working correctly, and at the correct frequency of 16MHz. > Any clues!? RAM.? Best of luck. Terry ? From dgriffi at cs.csubak.edu Thu Jun 8 06:08:15 2006 From: dgriffi at cs.csubak.edu (David Griffith) Date: Wed, 7 Jun 2006 23:08:15 -0700 (PDT) Subject: [P112] 8in floppies Message-ID: Has anyone here tried and succeeded at using an 8in floppy drive on the P112? -- David Griffith dgriffi at cs.csubak.edu A: Because it fouls the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail?