From ccl@www.ccl.net Fri Feb 28 11:26:10 1997 Received: from bedrock.ccl.net for ccl@www.ccl.net by www.ccl.net (8.8.3/950822.1) id KAA26868; Fri, 28 Feb 1997 10:50:55 -0500 (EST) Received: from relay5.UU.NET for frisch@lorentzian.com by bedrock.ccl.net (8.8.3/950822.1) id KAA18449; Fri, 28 Feb 1997 10:50:54 -0500 (EST) Received: from uucp3.UU.NET by relay5.UU.NET with SMTP (peer crosschecked as: uucp3.UU.NET [192.48.96.34]) id QQcewd01854; Fri, 28 Feb 1997 10:50:55 -0500 (EST) Received: from m10.UUCP by uucp3.UU.NET with UUCP/RMAIL ; Fri, 28 Feb 1997 10:50:56 -0500 Received: from mjf by m10.lorentzian.com; Fri, 28 Feb 1997 10:47:35 -0500 Received: by mjf (AIX 4.1/UCB 5.64/4.03) id AA19330; Fri, 28 Feb 1997 10:47:34 -0500 From: frisch@lorentzian.com (Mike Frisch) Message-Id: <9702281547.AA19330@mjf> Subject: Re: CCL:G:G94: calcall BUG ! To: chemistry@ccl.net Date: Fri, 28 Feb 1997 10:47:34 -0500 (EST) In-Reply-To: from "" at Feb 28, 97 02:18:19 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Steven Creve writes: > > > Hi, > > Has anyone ever come across this apparent bug in G94? : > (I use IBM-RS6000-G94RevC.3) > > The frequencies in the output differ between "opt=calcall" and "freq". > A small calculation on H2O makes this clear: > using the following input: > The two frequency outputs are the same in G94RevE.1 on the RS/6000; this was not known to a problem in earlier versions, either. Problems of this sort should be sent (preferably with both input and output) to help@gaussian.com. Mike Frisch From ccl@www.ccl.net Fri Feb 28 14:26:10 1997 Received: from bedrock.ccl.net for ccl@www.ccl.net by www.ccl.net (8.8.3/950822.1) id OAA28016; Fri, 28 Feb 1997 14:01:49 -0500 (EST) Received: from ETSU.ETSU-TN.EDU for R29CLOSE@ETSU.ETSU-TN.EDU by bedrock.ccl.net (8.8.3/950822.1) id OAA24055; Fri, 28 Feb 1997 14:01:47 -0500 (EST) Message-Id: <199702281901.OAA24055@bedrock.ccl.net> Received: from ETSU.ETSU-TN.EDU by ETSU.ETSU-TN.EDU (IBM VM SMTP V2R2) with BSMTP id 3790; Fri, 28 Feb 97 14:01:31 EST Received: from ETSU.ETSU-Tn.Edu (NJE origin R29CLOSE@ETSU) by ETSU.ETSU-TN.EDU (LMail V1.2c/1.8c) with BSMTP id 1136; Fri, 28 Feb 1997 14:01:31 -0500 Date: Fri, 28 Feb 97 13:26:55 EST From: R29CLOSE@ETSU.ETSU-TN.EDU Organization: East Tennessee State University Subject: G94 on LINUX To: chemistry@ccl.net Dear Members of CCL: Last week I posed a question about problems compiling G94.D4 on RedHat Linux 4.0. I have had a number of interesting replies which I plan to summarize. But I still have a problem. Gaussian, Inc. wrote me to suggest a hardware problem. It turns out this this is quite likely the actual problem. But how to fix? It seems that I have a faulty memory chip. My computer, a Gateway P6 has 64 Mbytes of new memory. The machine says all is ok on boot- up. But I have random problems with the C-compiler quitting. So maybe there is a memory problem. Gaussian suggested removing or switching memory. Now it appear I can compile the program with only 2x16 Mbytes of memory. So the question is, how does one test memory? OK, the computer thinks the memory is ok. But all the boot-up test is doing is writing 1, 0 into memory and then seeing if it is still there in a few microseconds. But it would seem that to compile and run a big FORTRAN program, one would be interested in having this same test performed 10 seconds later. How then does one test the long-term refresh function of big memory blocks? Regards, Dave Close.