"St." <dot### [at] dotcom> wrote:

<*snip*>
>      Glad to at least try and help - I'm no programmer, so please bare with
> me, and I'll help if I can. If you want me to do anything specific, then
> just fire away with what you want.  :)

Thanks Steve :)

>      Ok, first of all, here's my setup if it's of any use to you:
>
>        XP Home Edition (5.1, Build 2600) Service Pack 2
>        AMD Duron(tm) Processor ,  MMX,  3DNow, ~600MHz
>        512MB RAM
>        NVIDIA GeForce FX 5700LE

It's always helpful to know how the program is being used and on what
system.

>    I've had a good play with Tesslesphere this morning, and I've noticed a
> couple of things. When I first started playing with it earlier this week, I
> think I was a little too impatient and not giving TS a chance to do its
> thing after clicking around on various buttons, which is why all of a
> sudden, it would just disappear from the screen, (literally, it would just
> disappear without any warning).

Hmmm... seems like a crash. I'll upload a degug version for windows so you
can trace out what's going on. I'll also code a progress box for big data
sets to give more user feedback.

> So I learnt to watch the FPS field and let  it go up to around the 40 to
> 50 mark - that helped a lot.

>     The other thing that I found was that TS will save a simple file, but
> won't save a complicated file, (even after doing the above). I don't know
> what a 'complicated' file is when using TS, but I've uploaded a screenshot
> so you can at least see what will cause TS to crash when trying to save a
> file: http://website.lineone.net/~25ct/test.htm

Thanks for the screen shot. I managed to export this to an INC file on Linux
using TesselSphere 1.3.3. I can send it to you, but please read on...

>     Again, this crashed TS as soon as it got to the "Making Cell Edges
> unique [pass2]" stage. It also took about thirty minutes to get to this
> stage, so would this be normal? I'm thinking it might be my setup, and maybe
> it's not powerful enough?

Your set up is well powerful enough. But even so, that's a pretty big file:
you are trying to export 14400 cells; the _ts_V array alone contains 41400
vertices. One way round this is to just export a single geodesic particle.
You can then write POV macros that use the symmetry of the system to rotate
the particle to other locations on the sphere. I will update the POV-Macros
so that you can do this. Of course, 30 minutes to export a file is not
normal. The reason for the delay is outlined below...

The problem you are experiencing with big data sets is due to TesselSphere's
INC file export code and the internal data representation. TesselSphere does
not yet use indexed cells, so it has to generate all the indices to the
vertex array when saving the INC file. This ensures that only unique
elements are stored in the _ts_V array. This delay does not occur with VRML
export as no indexing takes place.

Right now, INC file export delays with large data sets is a limitation in
the program and one I will endevour to overcome. I would really like to
thank you for your time in reporting this. User feedback is important; I
know TesselSphere's quirks, so when using it, I don't attempt the things
that others might.

Best wishes

N.Shea

Post a reply to this message