[Kst] You cannot load this much data

[Barth Netterfield]

On Mon, May 7, 2012 at 4:46 PM, Ben Lewis <egretengineering at gmail.com> wrote:
> After I posted this bug I restarted Kst and the problem disappeared so I
> can't easily recreate this problem to test the new release. I'm currently
> using 2.0.4. I gave 2.0.5 rc3 a try recently but I could not switch over
> because it did not interpret my existing 2.0.4 files correctly.

Whoah, Whoah, Whoah.... That's not OK and needs to be fixed!

Can you send me some example 2.0.4 files that 2.0.5 has trouble
loading and send them to me?  I will get on that asap.

> I don't think this problem is due to an actual shortage of memory. I have
> used much bigger data files in the past.
>
> If I retrace my steps, my Kst session started to run very slowly (taking a
> long time to respond to mouse inputs) when I plotted one time dependent
> variable against another to produce a correlation scatter plot. I think, if
> I remember correctly, I restarted Kst and this fixed the slow response
> problem. Then, not long after this, I got the memory error. I restarted Kst
> again and everything is back to normal.

Hmmm.... not cool, and hard to fix.

Peter has fixed the "Windows Always Has 1024 Megs free" bug though.
But I really don't understand how you could have gotten the error message
you got without asking for more than 1024 megs at a time....  Most interesting.


-- 
C. Barth Netterfield
University of Toronto
416-845-0946