DBASE 8 plus and older DBASE

[BIG49010]

Just looking into the new Dbase 8 plus http://www.dbase.com/ , it looks like these guys bought all the rights to the old dbase, and are trying to make a good, well supported product. Anyone using this, or the older versions of dbase?

[DJofSD]

With so many modern relational offerings available, why go backwards?

[BIG49010]

With so many modern relational offerings available, why go backwards?

What would you go with?

[DJofSD]

What do you need to do? On strictly Windows or do you need support across OS boundaries?

[BIG49010]

Windows seven 64 bit, I am looking for a database program that can be programed to automate the query process of looking up a couple hundred angles of 250 horses in less than an hour. Importing the comma delimited files from bris, or some other source, and grinding through all the data.

[Dave Schwartz]

Windows seven 64 bit, I am looking for a database program that can be programed to automate the query process of looking up a couple hundred angles of 250 horses in less than an hour. Importing the comma delimited files from bris, or some other source, and grinding through all the data.

An hour? That should be several minutes. Seriously, we compute 4,000 factors per horse for every race on every card in MAYBE 4 minutes.

Note that this includes the import.

Re-doing the entire process for factors effected by a late scratch (like ranking factors, tote board, etc) for a single race takes maybe 3 seconds on a slow computer. On a fast one it is an eye blink.

My point is that once you get it programmed it will just FLY.

Oh... re-figuring 150 trainer stats is another 2 seconds.

[BIG49010]

I guess I am just a dinosaur and my destiny is extinction.

[Dave Schwartz]

Not sure why you would say that. I think this would be a great product for someone trying to maintain a database.

My point was that it will go much faster than you think.

[DeltaLover]

Of course Dave is right.

You should optimize your solution.

When having a time consuming application that you try to make it faster, the first think you should consider is to improve the algorithm. Chances are that in several places on your code you are using brute force that can be accelerated if you improve your algorithm.

As an example of this, you can view the storage of your factors. A trivial approach is to create a database table containing rows consisting of horse id and factor id. So if you have a horse matching 200 factors you end up with 200 rows in this table. This is a very weak solution that not only complicates your processing layer but also pollutes the schema of your data base making queries awkward and slow. The solution to this is instead of requiring a table for each factor, to store all matching factors in a single field in your horse record. For this you can use a field typed as long where each bit will represent a specific factor which will be accessed through bit-wise operators. Now each combination of factors becomes a mask that you can easily use for retrieval operations.

There are many similar tricks that you can use to improve the effectiveness of your algorithms and you should spend time discovering them.

The other important thing you need to reevaluate when it comes to the performance of your application has of course to do with the underlined technology your are using. You need to have a good understanding of whether you need to change your database from SQLServer to MongoDb for example, what are the pros and cons of each one?

Do you really need a fancy GUI or you can do much better just by using console based tools?

How balanced is the workload between data and processing tiers?

Do you know where the bottle necks are coming from ? Do you use a profiler to detect them?

These (among many others) are some of the questions you need to answer as you are optimizing your applications.

[Dave Schwartz]

Delta,

I wish I could afford for you to come and write code for me.

Heck, I wish I could afford ME!


Great post.
:ThmbUp:

Dave

[DeltaLover]

Delta,

I wish I could afford for you to come and write code for me.

Heck, I wish I could afford ME!


Great post.
:ThmbUp:

Dave

I am convinced you are an extremely capable and creative developer Dave. Especially after you told me that decades ago you ended up coding a Neural network without even knowing the underlaying theory.

Of course if you have an idea that you do not have time to implement maybe I can help...

[Dave Schwartz]

LOL - The key word was "AFFORD."

[Red Knave]

Windows seven 64 bit, I am looking for a database program that can be programed to automate the query process of looking up a couple hundred angles of 250 horses in less than an hour. Importing the comma delimited files from bris, or some other source, and grinding through all the data.
There are multiple free SQLs that would fill the bill. They are all higher up on the learning curve than a dBase clone but the effort repays you twenty fold.

PostGreSQL, mySQL, MSSQL Server Express, FirebirdSQL, SQLite are some products to look into.

After saying that I see that dBase+8 allows some SQL statements in the language and can connect to other databases so it may be a tool you can use to do what you want to do.

[DJofSD]

There are multiple free SQLs that would fill the bill. They are all higher up on the learning curve than a dBase clone but the effort repays you twenty fold.

PostGreSQL, mySQL, MSSQL Server Express, FirebirdSQL, SQLite are some products to look into.

After saying that I see that dBase+8 allows some SQL statements in the language and can connect to other databases so it may be a tool you can use to do what you want to do.
Another to consider is Open Office Base.