TMG-L Archives

Archiver > TMG > 2004-10 > 1096775407


From: Lee Hoffman <>
Subject: Re: [TMG] Splitting a V5 Project
Date: Sat, 02 Oct 2004 23:52:32 -0400
References: <415E60A4.3040607@fairhall.id.au><6.0.0.22.2.20041002095521.03d6ec38@pop3.norton.antivirus><6.1.2.0.0.20041002190007.03358eb0@incoming.verizon.net>
In-Reply-To: <6.1.2.0.0.20041002190007.03358eb0@incoming.verizon.net>


Walt wrote:
>At 10:27 AM 2/10/2004, Lee Hoffman wrote:
>... certain filters will be faster than others, but even so they are still
>searching the entire project.
--clipped
>>Are you still searching the whole project if you deactivate data sets?

Yes. All data for a data set in a project are maintained in the same files
as all other data sets in that project. For example, all event data for
persons in data set #1 are included in the same file as event data for
persons in data sets #2, #3, #4, etc. Only the display on your monitor is
affected by the disabling of one or more data sets. That is, if you disable
a data set, then the data in that data set is not display in the Picklist,
Project Explorer, etc. until you Enable that data set in the Data Set
Manager. The idea is to allow you to reduce the "clutter" on your screen.

If you want "less clutter" in the files to speed up searches then about the
only way to do it is to split your projects up so that they contain fewer
data sets and persons. However, the speed of searches is fairly fast
anyway for most users even for filters with many clauses unless you have a
very slow machine with few resources. I have a single data set project
with over 40,000 persons and most filters are processed within a few
seconds and the rest rarely take more than a 20 to 40 seconds. I do create
some filters occasionally that take longer but those are the exception and
involve thousands of combinations, hundreds of people, and many generations.

I tested this by making a copy of my 40,000 plus project and then merging
the original into the copy. This gave me a two data set project with each
data set being very large. Then I ran two tests -- timing each one. For
each test, I filtered the Project Explorer for "Surname Equal
Hoffman". For the first test, both data sets were enabled in the Data Set
Manager. This timed out at about 8 seconds. For the second test, I
disabled the second data set and it also took 8 seconds. So you can see
that whether a data set is enabled or disabled does not make a difference
in timing of a filter -- or not enough to be able to determine a difference
with the clock I had available to me.

Now this was a test of a simple one line filter using two 40,000 person
data sets in a single project on a single computer. The computer is not
very state of the art (couple years old) and could use more memory (only
has 384MB) to improve performance. Other computers with more resources
could reduce the timing of each part of the test. But the end result would
probably be the same with maybe minor difference between the two.

Hope this helps -

Lee Hoffman/KY
TMG Tips: <http://www.tmgtips.com>;
My website: <http://www.tmgtips.com/lhoffman>;
A user of the best genealogy program, The Master Genealogist (TMG)

This thread: