From owner-oi-users Fri Mar 10 12:58:48 1995 Return-Path: oi-users-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA24058 for oi-users-outgoing; Fri, 10 Mar 1995 12:58:48 -0800 Received: from lehman.com (Lehman.COM [192.147.66.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id MAA24052 for ; Fri, 10 Mar 1995 12:58:28 -0800 Received: from relay.lehman.com by lehman.com (8.6.4/LB 0.1) id PAA19028; Fri, 10 Mar 1995 15:57:12 -0500 Received: from cfdev1.lehman.com by relay.lehman.com (4.1/LB-0.6) id AA25800; Fri, 10 Mar 95 15:55:46 EST Received: from cfdev1026.lehman.com by cfdev1.lehman.com (4.1/Lehman Bros. V1.6) id AA17301; Fri, 10 Mar 95 15:55:44 EST Message-Id: <9503102055.AA17301@cfdev1.lehman.com> Date: Fri, 10 Mar 95 15:55:43 EST X-Mailer: Mail User's Shell (6.5 4/17/89) From: fgreco@lehman.com (Frank Greco) To: oi-users@freefall.cdrom.com Subject: Re: Rules to speed up the creation of OI objects Sender: oi-users-owner@FreeBSD.org Precedence: bulk : Can this help the speed of uib coming up? Uib makes my SS10 feel : like a SS1+. Warner says: > uib coming up is due to a number of things. One of them is the fact > that OI was asking about fonts in a way that was slow. uib comes up > much faster now than it used to with this simple change. Steve says: > No, that won't help with uib. > I applied all of those speedups years ago. > > We believe that part of the uib slowness was due to a font cacheing problem > which the guys at Pure discovered after 4.5.3 was released. This has been > applied to the 4.6 release. Thanks guys. I hope to get 4.6 soon to get the new-n-improved uib. Since we're on the uib topic... I noticed that if uib is having problems with the colormap it streams warning messages to stderr first, then it seems that you guys redirect stderr to the footer of uib and the "cannot get color blah...blah..." messages get blasted to the uib footer. Of course uib resizes to fit the text and I get the "Oprah Winfrey Effect"... uib expands and contracts in an seemingly endless loop. Is there anything I can do to StopTheMadness(tm)? Frank - yeah... I know about the "get a 24-bit frame buffer" solution already... ;)