Ask Steve Borgatti: [log in to unmask]

In a message dated 3/15/02 2:11:01 PM, [log in to unmask] writes:

<< I have used both the DOS and the Windows version with networks with
1000 nodes. I never had any problems with any versions regarding the
size of the matrix. I sure don't have a supercomputer (266 Mhz, 98 M

Of course it all depends on the operation you plan to do with a large
matrix. Check the manual of UCINET, they tell the time demand of each
routine expressed in the power of N (number of nodes). (That's a smart
thing that I would appreciate in other software manuals too.) So you
can run a small matrix doing the same procedures that you would do with
the large one, record the time demand (UCINET tells you that), and
multiply it with the appropriate power of the numner of nodes.

E. g. for me to run CONCOR at 5 cut depth with 1000 nodes takes about
35-40 minutes on the 266 Mhz laptop. (So at this level it is coffe
brake software.)

Good luck

----- Original Message -----
From: "Andrew V. Shipilov" <[log in to unmask]>
Date: Friday, March 15, 2002 12:48 pm
Subject: Re: UCINET capacity

> I have used the UCINET Windows version and it can crunch up to 255x255
> matrix. The DOS version should be the same. If you need to analyse
> larger datasets, you might have to turn to MATLAB, however, you will
> need to use your own (or developed by someone else--like Professor
> BillRichards from SFU) codes. There the size of a network is not
> an issue,
> but the computer specifications are. I managed to crunch 2000x2000
> matrix in MATLAB. When I went over that size (3000x3000), my computer
> (one of the lastest Dells) gave me a memory error.
> Andrew V. Shipilov
> Ph.D. Candidate
> Strategic Management and Organization Theory
> Joseph L. Rotman School of Management
> University of Toronto
> 105 St. George Street
> Toronto, ON, Canada
> M5S 3E6
> Fax: 416-978-5433
> -----Original Message-----
> From: Jenny Godley [mailto:[log in to unmask]]
> Sent: Friday, March 15, 2002 10:27 AM
> To: [log in to unmask]
> Subject: UCINET capacity
> Dear Socnet List:
> What is the largest network one should analyze with UCINET (currently
> using IV, DOS version)?
> Any advice appreciated.
> Thanks.
> - Jenny
> Jenny Godley
> Dept. of Sociology
> Haverford College
> Haverford, PA 19041

----------------------- Headers --------------------------------
Return-Path: <[log in to unmask]>
Received: from ( [])
by (v83.45) with ESMTP id MAILINXG13-0315171101; Fri,
15 Mar 2002 17:11:01 -0500
Received: from ( [])
by (v83.45) with ESMTP id MAILRELAYINXG45-0315171056;
Fri, 15 Mar 2002 17:10:56 -0500
Received: from PEAR.EASE.LSOFT.COM ( by
(LSMTP for Digital Unix v1.1b) with SMTP id
<[log in to unmask]>; Fri, 15 Mar 2002 17:11:16 -0500
Received: from LISTS.UFL.EDU by LISTS.UFL.EDU (LISTSERV-TCP/IP release 1.8d)
          with spool id 82363 for [log in to unmask]; Fri, 15 Mar 2002
          17:11:12 -0500
Received: from ( []) by
 (8.12.1/8.12.1/2.2.3) with ESMTP id
          g2FMBBEx046440 for <[log in to unmask]>; Fri, 15 Mar 2002 17:11:12
Received: from (LOCALHOST []) by (iPlanet
          Server 5.1 (built May  7 2001)) with ESMTP id
          <[log in to unmask]> for [log in to unmask]; Fri, 15 Mar
          2002 23:12:05 +0100 (MET)
Received: from [] by (mshttpd); Fri, 15 Mar 2002
          17:12:05 -0500
MIME-version: 1.0
X-Mailer: iPlanet Webmail
Content-type: text/plain; charset=us-ascii
Content-language: en
Content-transfer-encoding: 7BIT
Content-disposition: inline
X-Accept-Language: en
Message-ID:  <[log in to unmask]>
Date:         Fri, 15 Mar 2002 17:12:05 -0500
Reply-To: "Balazs Vedres (bvedres)" <[log in to unmask]>
Sender: Social Network Researchers <[log in to unmask]>
From: "Balazs Vedres (bvedres)" <[log in to unmask]>
Subject:      Re: UCINET capacity
To: [log in to unmask]