|
JCapper Message Board
JCapper 101
--
Changing data bases need help
|
|
By |
Changing data bases need help |
rosenowsr 8/3/2012 8:18:41 AM | I have 2 databases set up one is 2012-Q2 and the other is 2012-Q3. I finally have enough races to forward test, 3 months in database Q2 and 1 month in database Q3, HDW data.
When I set up a SQL in Q-2 and then run it on Q3 I get the same number readings, win% number of races, all the same. They should be different using different databases. I have changed DFM default and still the same thing happens.
What am I doing wrong? Or point me to the proper video, which I can't seem to find.
Trying to learn to forward test.
Thanks rosenowsr
P.S.After doing a little thought processing I realized I can do the same thing by adding a date limiting factor to theSQL UDM instead of changing folders. Is my thinking correct in this?
~Edited by: rosenowsr on: 8/3/2012 at: 8:18:41 AM~
| jeff 8/3/2012 1:15:05 PM | Running the Data Window in sql mode is different than running it in playlist file mode.
Playlist File Mode When you run queries in playlist file mode you are pointing the Data Window at a single folder - and by hitting the ALL or UDM buttons are commanding the Data Window to read data out of a text file sitting on the folder where you have the Data Window pointed.
When desired, you can get around the single folder per query limitation through the MENU - EXPORTS - QUICK INDEX FILE options in the Data Window.
FYI, procedure for doing that spelled out here:
http://www.jcapper.com/messageboard/TopicReader.asp?topic=710&forum=JCapper%20101
But for the most part, playlist file queries that you run will be using the program default - where you are telling the Data Window to query text files sitting on a single folder only.
SQL Mode When you run the Data Window in SQL Mode, the source data for your queries isn't being read from text files and the folder that you happen to have the Data Window pointed at doesn't matter.
In SQL Mode, when you hit the Execute button you are commanding the Data Window to query data sitting in the JCapper2.mdb file that you currently have the Data Window connected to.
Note that the default file that the Data Window (auto) connects to each time you launch it is:
c:\JCapper\Exe\JCapper2.mdb
Also note that you can have multiple JCapper2.mdb files - and can make the Data Window connect to a different JCapper2.mdb file by hitting MENU - CONNECT TO JCAPPER2 FILE - answering Y when prompted - and using the dialog box to select a (new) desired file.
Above I said the following:
In SQL Mode, when you hit the Execute button you are commanding the Data Window to query data sitting in the JCapper2.mdb file that you currently have the Data Window connected to.
Continuing with that train of thought, it is perfectly normal for the StarterHistory table in a JCapper2.mdb file to be populated with data spanning multiple playlist file data folders.
In my own case, my current StarterHistory table has data in it from each of my quarterly calendar folders for years 2011 and 2012.
I'm guessing, based on your post that your StarterHistory table has data in it from each of the playlist file folders you have compiled so far. I'm further guessing that nothing is wrong - that this came about as the result of normal program operation.
Do this:
Select "+ Show Individual Plays" and then run any of your sql expressions through the Data Window with the following line added to the end of it:
ORDER BY [DATE], TRACK, RACE
The above order by command will cause the Data Window results to sort chronologically first, by track code second, and lastly by race number.
Further, you should be able to see the individual starters - and from that determine the following:
a. ALL of my starter information is there - nothing is missing.
-or-
b. For some reason, I'm missing starter data for one of my calendar time periods. (If that's the case, let me know and we can go from there.)
If case a. above best describes what you see, from there it becomes a matter of answering the following question?
Q. What should I use as a development sample? (And what should I use as a validation sample?)
More to come...
-jp
.
|
|