Large data stacks - performance

[Sarah Reichelt]

On 5/1/07, J. Landman Gay <jacque at hyperactivesw.com> wrote:
> Has anyone worked with very large data stacks -- say, 500 megs or so? Is
> there an appreciable slowdown? I have a client who wants to convert his
> HyperCard stacks as-is, without a database. I'm concerned about virtual
> memory swapping and other performance issues.

I haven't worked with anything that large, but I did convert a
HyperCard stack that used one card per record. It had about 5000
cards. Searching using "mark card" was super-fast in HC and unusably
slow in Rev, so I had to convert it to a text file database.

Since Rev works by loading stacks completely into memory, I would
reckon that a 500 Meg stack would not be feasible.

Cheers,
Sarah