Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!mailrus!cornell!uw-beaver!rice!sun-spots-request From: hedrick@geneva.rutgers.edu (Charles Hedrick) Newsgroups: comp.sys.sun Subject: Re: Summary of responses for the 4.0 performance on 3/50 Message-ID: Date: 14 Dec 88 08:57:55 GMT References: <8811211803.AA22290@syteka.syteka> Sender: usenet@rice.edu Organization: Rice University, Houston, Texas Lines: 14 Approved: Sun-Spots@rice.edu Original-Date: 3 Dec 88 10:10:06 GMT X-Sun-Spots-Digest: Volume 7, Issue 52, message 2 of 14 >One user said that the problem was not in 4.0 as such but in 'suntools', >and he was using 3.X suntools. I haven't tried that, I am not sure if the >binaries are compatible. Almost all 3.X binaries will run under 4.0. The only case I recall where that wasn't true was Common Lisp, because it depended upon details of where the stack is put, and that changed. I've actually been surprised at how much compatibility there really is. The 3.2 Suntools binaries work just fine under 4.0, and perform better than the 4.0 Suntools. I understand that if you have lots of windows, there are some performance problems with Suntools under 4.0, even if you move back to the 3.2 version. Our operators typically keep a window open to just about every machine on the network. I'm told that this worked OK under 3.2, and is terrible under 4.0. More normal Suntools uses seem OK as long as you use the 3.2 Suntools.