Received: from proxy2.bigplanet.com (proxy2.bigplanet.com [216.169.193.160]) by hub.org (8.10.1/8.10.1) with ESMTP id e58E2qT74612 for ; Thu, 8 Jun 2000 10:02:52 -0400 (EDT) Received: from new6 (1Cust56.tnt1.bwi1.da.uu.net) by proxy2.bigplanet.com (Sun Internet Mail Server sims.3.5.1999.07.30.00.05.p8) with SMTP id <0FVU0046R9ON40@proxy2.bigplanet.com> for pgsql-interfaces@postgresql.org; Thu, 8 Jun 2000 08:02:49 -0600 (MDT) Date: Thu, 08 Jun 2000 10:05:23 -0400 From: Mihai Gheorghiu Subject: Re: Win ODBC drivers for Postgres To: pgsql-interfaces@postgresql.org Cc: pgsql-general@postgresql.bigplanet.com Message-id: <002f01bfd152$b3247e60$2716113f@new6> MIME-version: 1.0 X-Mailer: Microsoft Outlook Express 4.72.3110.1 Content-type: text/plain; charset="iso-8859-1" Content-transfer-encoding: 7bit X-MSMail-Priority: Normal X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 X-Priority: 3 X-Archive-Number: 200006/51 Ladies and gentlemen, Please accept my sincere apologies. I was just as frustrated as you were when reading that person's Memo. He tried to take advantage of what the French say, "Les absents ont toujours tort." - Absents are always wrong. Given my very limited experience with PostgreSQL, I needed arguments to support my option. Now the jury is out. I hope they will vote for Open Source, at least for financial reasons. Actually the document tried to justify a network architecture that required transactional replication (2way :-( MS claims it is built in their SQL7. If this will be a requirement, what is the timeframe for having that implemented in PostgreSQL? It is my understanding this can be done now too, only by programming. Best regards, Mihai >Hello, > >At 10.36 07/06/00 -0400, you wrote: >>Quote: >>[ODBC public domain drivers] are immature, inefficient and are network >>bandwidth hogs. The public domain drivers don't support connection caching >>which reuses an existing database connection for same queries to reduce the >>query result time. > >well, don't know about the internals of the PostODBC driver, anyway I have >performed many tests on it. The latest version of the driver coupled with >the latest version of PostgreSQL with my tests was about 2 times faster >than MS SQL Server 6.5 with its own ODBC Driver, especially when more than >one concurrent connection was used. And in my experience (real experience, >not benchmarks) most commercial databases are slower than MS SQL Server and >have ODBC drivers that are almost useless because of bugs, unimplemented >features and many other quirks. I''m just working with Centura SQL Base >(version 6.1.something) with its own ODBC. It's just a pain. It doesn't >report indices and primary keys so Visual Basic with DAO refuse to perform >any write operation with it. I have to build complex SQL statements to do >any change to the data contained in the database. Every now and then it >refuses to connect to the database with no apparent reason, and so on. > I simply completely disagree with the above statement. There's >nothing slower than something that simply doesn't work at all, and many >commercial databases simply doesn't work, who cares if they are faster when >they work? > >Bye! >-- > Denis Sbragion > InfoTecna > Tel, Fax: +39 039 2324054 > URL: http://www.infotecna.it >