w8sdz%brl@sri-unix.UUCP (12/22/83)
From: Keith Petersen <w8sdz@brl> Phil Lapsley has released a new overlay for MDM7xx for the new US Robotics 300/1200 baud S-100 modem. The overlay works on MDM712 and later (including the latest MDM715). It's available on SIMTEL20 in the MICRO:<CPM.MODEM7> directory as M7US-1.ASM. Dave Hardy (Microsystems author and Sysop of TCBBS Dearborn MI) had one of these modems for evaluation several months ago. He ran into some problems trying to use it with the BYE program on his Sysop Clearinghouse RCPM but eventually found what was wrong. I suggested that Phil contact Dave for details and here's some information that Dave Hardy passed along which may belp others trying to use this modem with BYE. ----- Forwarded message # 1: Date: Wed, 21 Dec 83 11:14:48 pst From: The tty of Phil Lapsley <jlapsley%D.CC@Berkeley> To: w8sdz@brl Subject: USR overlay I talked with Dave Hardy this morning, and we hashed out a few things. His problem was on auto-answer, which mdm715 doesn't really have to cope with. On AA, the USR changes the baud rate to suit the incoming call, but it never tells anyone that it changed. Therefore, things get excellent when one tries to send commands to it when it is at a different baud rate than one expected. But he confirmed what I suspected: the board is not meant to run faster than 4 MHz (i.e., my 5) and he agreed I should have some time-padding do nothing stuff in there to slow things down (as an aside, another flaw on the board is there is no provision for wait states). In M7US-1.ASM the initialization routine are two XCHG statements after each OUT statement. They are commented-out (i.e., "; XCHG") so they will not be there for people who use slower systems. However, there is a note explain- ing what they are for and telling people with fast systems to remove the semi- colons and fix things up. I have tested this one fairly extensively, and without the delay it locks up. With the delay I have yet to be able to cause it to lock. ---Phil ----- End of forwarded messages