aking@BFLY-VAX.BBN.COM.UUCP (09/25/87)
1. DEBUGGERS: Is there any way of reinitializing the program in SID without exiting and reentering (and destroying debugger state?) Like the "run" command in dbx. Speaking of the devil, does amiga have source level debugging? When is atari's new debugger coming out? 2. EMACS: Who can tell me when ^y (yank) in micro-emacs will bring back just the last thing deleted (per the emacs convensions) or a whole bunch of other deletions which seem to be trying to make up for the lack of a kill ring buffer, and is quite annoying. Sometimes, for long streches of time, it just (corectly in my opinion) puts just the last thing, and sometimes, the whole kitchen sink gets plopped down in the middle of what you are doing, without even setting the mark on the other side! (Oh, I guess that was a flame!) Perhaps I'm suffering from old version-itus? is it fixed in the latest? (oh please add word wrap!) 3. COMPILERS and LINKERS: Atari's new linker was a godsend. It takes links with vdi & gem down from 2 min to 18 seconds (both with Supra and much ramdisk). Bravo. And filenames can have backslashes in them too -- quality work! Now, if the assembler in the latest version of Alcyon (4.14) would only do what earlier versions could, namely put temporary files on ramdisk. When I upgraded to 4.14 (I need its improvements in structure support) the time to assemble trippled, presumably due to the loss of ramdisk for temporaries. If I sound like a speed freek, its because my world clock runs on compile-load-go time periods. I have definately been jaded by the SUN-3 that sits on my desk at work. Does Anybody know if the Light Speed C people have any ST intensions? Will it run under magic sac? Allen King aking bfly-vax.bbn.com ucbvax
dyer@atari.UUCP (Landon Dyer) (09/25/87)
> 3. COMPILERS and LINKERS: Atari's new linker was a godsend. It takes > links with vdi & gem down from 2 min to 18 seconds (both with Supra > and much ramdisk). Bravo. And filenames can have backslashes in them > too -- quality work! Now, if the assembler in the latest version of > Alcyon (4.14) would only do what earlier versions could, namely put > temporary files on ramdisk. When I upgraded to 4.14 (I need its > improvements in structure support) the time to assemble trippled, > presumably due to the loss of ramdisk for temporaries. If you are a registered developer, you can get a beta copy of MADMAC, a high-performance assembler written here at Atari (by people who were pissed-off at AS68). Briefly: o 20,000 to 40,000 lines a minute (on a hard disk or ramdisk); o Macros, include files, repeat blocks, etc... o Will act as a back-end to the Alcyon C compiler; o Can generate Mark Williams object code modules (so MWC people don't have to use the icky MW assembler); o Will generate ".PRG" files directly; It's on Compuserve, in the developer's SIG. If you're not on Compuserve, you can ask John Feagans to send you a copy via snail mail. -- -Landon Dyer, Atari Corporation {sun,amdcad,imagen,hoptoad}!atari!dyer The views expressed here do not necessarily reflect those of Atari or the AI software that has taken over my brain. YOW! I am waiting for my warranty-expired interrupt!
nwd@j.cc.purdue.edu (Daniel Lawrence) (09/28/87)
>2. EMACS: Who can tell me when ^y (yank) in micro-emacs will bring back >just the last thing deleted (per the emacs convensions) or a whole >bunch of other deletions which seem to be trying to make up for the >lack of a kill ring buffer, and is quite annoying. Sometimes, for >long streches of time, it just (corectly in my opinion) puts just the >last thing, and sometimes, the whole kitchen sink gets plopped down >in the middle of what you are doing, without even setting the mark on >the other side! (Oh, I guess that was a flame!) Perhaps I'm suffering >from old version-itus? is it fixed in the latest? (oh please add word wrap!) > > > Allen King aking bfly-vax.bbn.com ucbvax MicroEMACS 3.x keeps a flag indicating what kind of command the last command was. Whenever you do a kill, if the immediatly previous command was NOT a kill, then the kill buffer is first cleared, and then the text is saved. Consecutive kills accumulate in the buffer, and kills after a cursor move or other command clear the buffer and start over. This is very predictable and I have had no problem reports on it. MicroEMACS, from version 3.3 and up ( 3.9 is current) has had WRAP mode. Just type ^XMwrap<RET> and the word WRAP will appear on the mode line. If you are using a very early MicroEMACS (Dave Conroy's original probably) I would suggest updating to something more current. Daniel Lawrence (317) 742-5153 UUCP: {ihnp4!pur-ee!}j.cc.purdue.edu!nwd ARPA: nwd@j.cc.purdue.edu FIDO: 201/2 The Programmer's Room (317) 742-5533