From owner-freebsd-hackers Tue Mar 19 16:18:12 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id QAA13820 for hackers-outgoing; Tue, 19 Mar 1996 16:18:12 -0800 (PST) Received: from mail.think.com (Mail1.Think.COM [131.239.33.245]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id QAA13815 for ; Tue, 19 Mar 1996 16:18:10 -0800 (PST) Received: from Early-Bird-1.Think.COM by mail.think.com; Tue, 19 Mar 96 19:18:07 -0500 Received: from compound (fergus-29.dialup.cfa.org) by Early-Bird.Think.COM; Tue, 19 Mar 96 19:18:05 EST Received: (from alk@localhost) by compound (8.6.12/8.6.112) id SAA11217; Tue, 19 Mar 1996 18:18:30 -0600 Date: Tue, 19 Mar 1996 18:18:30 -0600 Message-Id: <199603200018.SAA11217@compound> From: Tony Kimball To: hackers@freebsd.org Subject: Re: GAS question Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk From: Nate Williams Date: Tue, 19 Mar 1996 13:21:49 -0700 Subject: Re: GAS question > not so much the Emacs command set, per se, as the fact that it's a > huge memory pig. And VC++ isn't? I can run Emacs IDE in a smaller memory footprint than VC++ if I leave out X. 5MB for emacs w/X. I'm guessing 4MB for vc++. Emacs tends to keep a lot resident on my machine, but then I rarely swap so I don't complain. > I guess I could live with unguessable command syntax (how do you > exit microEmacs, anyway?) if I had printed documentation. Which > I have for VC++. So buy the doc from FSF. That's not hard. You get a half-decent Lisp environment too. Actually, it seems a worthwhile effort to provide a stone-dumb interface package for X Emacs, and a VC++-mode, just to lure in the suckers. Once they write an elisp rmail sorter, they're hooked for life. What would need to be added to Emacs? 1) Alt-C Alt-V Alt-X Alt-Z F-keys, in compile-mode bindings 2) More menu items 3) Improvements to tags 4) dissassembler From: Terry Lambert Date: Tue, 19 Mar 1996 13:30:48 -0700 (MST) Subject: Re: GAS question VC++ has a 2-4M footprint... or, actually, "Microsoft Developer Studio" has that footprint. My machine that it's installed on has 16M, but 12 of that is for Windows 95 and the broken VCACHE code for cache utilization backoff (that isn't fixed, even in their most recent update, publically available soon). The speed difference between a VC++ build-all w/16MB and one with 32MB is almost a factor of two for me. For gmake on a similarly sized project it is less than 10%. This is WinNT vs FBSD. > $VC++ 4.0 is $495 w/out documentation. Docs are another $150 + > shipping, and are now superceded by the pending VC++ 4.1 release. This is retail price. This is not what you pay for an MSDN Level 2 SDK/DDK/VC++ subscription. You don't get VC++ w/ the level 2 subscription (which was $495/an last I bought in). You get it w/ the level 3 subscription (which I *think* is $995/an). > You're arguements against using Emacs apply as well to VC++, so are > moot. Not so. I can click Icon's and menus without having to remember it. That is true enough. But then I don't remember the Emacs key commands myself. My fingers do. Emacs is my login shell on some machines, so I suppose I'm biased.