From my experience with Mac OS X,
1. GL version is faster than Tk only if you have a good enough
graphics card. Even then, as the number of spectra increases,
especially with many Windows displaying many spectra, it slows down.
I guess, it has to do with graphics card's memory mostly. On the
whole, I still find Tk version faster with a 64MB graphics card as I
have a lot of spectra and windows!
2. Slow quitting was solved by compiling all the components (Python,
Tk, etc.) with GCC 4.x and then using them with ccpnmr c-code
(compiled also with GCC 4.x.) Now it feels very snappy though it
wasn't bad before. There could be similar differences with the
components (Tk, python, etc.) you are using in SuSe and Red Hat distros.
On 6 Apr 2006, at 06:23 pm, Justin T Douglas wrote:
> Hi,
>
> I've recently changed linux distros from Suse to red hat and I
> noticed different Analysis behavior. First of all, the amount of
> time required to close Analysis increased several orders of
> magnitude. Secondly, if I got up to get a sandwich and the screen
> saver kicked on it took several minutes for Analysis to wake back
> up. Think Lazarus.
>
> This morning I switched from GL to TK and it seems to have helped.
> Any other suggestions.
>
> Justin
|