[ Closed ] Error message

Actually, I can't reproduce the error now either, but I can reproduce other errors that I've seen many times before, when I did following:
The case was this: (4b^2-12ac)^(0.5)
What happened was, that I made the base lower than zero, and this resulted in what I've attached as the First error (which I know is supposed to be displayed).
For example I would change the value of "c", so that 4b^2 < 12ac. I should tell you though that the actual values you see in Custom Functions.bmp is very different from the values I had used when I wrote about the error I described in the first message i posted, and sadly I haven't got the original data. Nevertheless when I've now tried to reproduce the original error I've experienced another weird thing.
For example I change c=(-1/30) to c=30, and I, of course, receive the first error. Then I click OK in the error-box and Annuller (Cancel) in the Custom Functions-box, and obviously nothing has changed from this.
Then I go to the Custom Functions-box where all my values are back to normal since I clicked Cancel and I can conclude that whatever change I make here (in this case I just added the constant n=2), I'm stopped by the second error, and I have to close and open the program to be able to make changes in the Custom Functions-box again.
I also noticed that if I didn't close and just repeated clicking OK in the Custom Functions-box the error message sometimes changes in the last part of the phrase as in Second error(2). And actually if I repeat clicking OK also this Third error pupped up one time randomly. The digits in the last part of Second error and Second error(2) also changes randomly to for example:
00000065
0000001C
and maybe others (I know it may not seem like it but I also do stuff apart from crash-testing), and will sometimes shut down when I do it long enough, just leaving me with the usual Windows-error that asks if I want to report the error to Microsoft. Usually though I see the error displayed in Second error.bmp when I repeat clicking OK. I'm testing as I sit here and this fourth error also just pupped up randomly:

Fejl 75: Ugyldigt navn %s til brugerdefineret funktion eller konstant.
Navne på brugerdefinerede funktioner og konstanter skal begynde med et bogstav og må kun indeholde bogstaver og tal. Du kan ikke bruge navne, som allerede er tildelt indbyggede funktioner og konstanter.

Just so you are still following me all these messages are displayed, while c=30, and I keep click OK ignoring the first message that actually tells me what the problem is. Therefore even though it's normal errors they should be displayed at this time. Finally I got this last error that I've also seen before, which actually closes the program completely and by this I mean it closes every grf window open at the time. I'll send it in the next message along with the .grf I'm working with.

Thanks a lot for the improved error description. I will look into these and try to fix them. I am not home the rest of the week, so it might be a while before I get back to you.

By the way, have you tried the beta version of Graph 4.4? Some of the problems you experience might have been fixed already.

And yes, I am Danish.
But as you said, it is probably best to keep the language in English if possible so other have a chance to comment.

I have now managed to reproduce the described problem with Graph 4.3. Fortunately it seems to have been solved already in the beta version of Graph 4.4 as I cannot reproduce it with the beta version.