The number of records in a table has very little to do with the size of the file. The number of bytes of data needed for a given record multiplied by the number of records would be a better indication.
But the largest factor in file size will be the presence of container fields in the table and whether physical copies of large files have been inserted into that container field or if data was inserted "by reference".
Large complex layouts with lots of graphics or a hi res graphic image as the back ground can be a significant factor increasing the size of your file.
And there can be blocks of space set aside for objects no longer in the file due to previous design changes made.
You might try using "save a copy as" with the "compacted copy" option to see if that reduces the size of any of these files by a significant amount.
All of the databases I am working with have container fields. They used to have physical copies of files, but earlier this summer I went through and changed them all to links ("by reference") which decreased the file sizes tremendously (they were hovering around 300 MB).
So, that shouldn't be the reason for why they have vastly different file sizes.
Also, these are very simple layouts with no frills.
I have tried the compacted copy option on all of the files, too, and they will not get any smaller.
Is there a way to check if the file has blocks of space set aside that it no longer needs and, if so, is there a way for it to "delete" those blocks?
Save a copy/compacted is supposed to do that.
Another user recently reported (with the current version of Filemaker, not v7), saving a clone of the file and importing all data back into the file to get a much smaller file size. This would also rebuild all field indexes--which can also be s significant factor in file size.
You might also try recovering the files and compare the size of the recovered copies to the originals.
I went in and removed all field indexes thinking that might be the issue, but it did not seem to have any effect. Most of the fields of these databases, save for one, contain very little text and the same values are repeated often.
The recovering did not change the file size, but the cloning and importing did. The 50 MB file reduced down to 200 KB. Amazing!
Thanks for the help! I'm not sure why it changed the file size so much since the indexing had already been turned off, but I'll take the results without complaining.