Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 (Tek) 9/28/84 based on 9/17/84; site motel6.UUCP Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!tektronix!reed!motel6!keith From: keith@motel6.UUCP (Keith Packard) Newsgroups: net.games.rogue Subject: problem with rogue7 strings Message-ID: <174@motel6.UUCP> Date: Sat, 29-Jun-85 04:15:25 EDT Article-I.D.: motel6.174 Posted: Sat Jun 29 04:15:25 1985 Date-Received: Thu, 4-Jul-85 00:20:40 EDT Distribution: net Organization: 5440 SE 41st, Portland, OR Lines: 52 I have a set of rogue7 sources (the one with multiple character classes) that do run fine on the vax. But, when I ported them to a Sequent Balance-8000 machine (32016), I got strange errors in the messages printed at the top of the screen. For example, I am fighting a bat: You miss the bat --More-- ------- |..b..| # |.@...+######## |.....| ------- When I hit space, the message changes to: Tou miss the bat --More-- Yes, this is correct. I check the sources and find that it *really* wanted to print: The bat misses you. but only the first character was printed correctly. When I hit ^R the correct message is displayed. This is exactly the same problem I had in porting rogue7 to my pdp11/73 at home, at the time I was sure it was a space limitation problem or a problem caused by the massively horrible things I did to get rid of the strings. (they sit in a file and are read into buffers at the bottom of the stack segment - it worked for the other rogue (5.0)) But, now I get the same problem on a big machine with no modifications to the source! In the course of debugging on the 11/73, I modified the wputs call that was responsible for printing this message to a series of wputc(*s++); refresh(); call sequences and, lo and behold, it worked! It was amazingly slow but correct. Does this strike a familiar note to those of you who have ported rogue7 to other machines? I just tested *exactly* the same sources on the Balance and on an 11/780, runs on the vax, breaks on the Balance. I must assume that some weird machine dependency exists. Oh, the same problem occured on another 32016 machine, a tektronix 6130 running Utek - curses is probably compiled directly from the 4.2 sources so I think I have ruled that out as most probably cause. keith packard ...!tektronix!reed!motel6!keith