Perhaps others have the same problem - exporting cel/vrt from Star-CD v4 into a form that can be read by older versions. Nastran/Patran export works somewhat, but has problems with pyramids etc.

I have a workaround (a perl program for converting the cel/vrt files). Since I cannot seem to attach the file without butchering all the lines, I'll send a copy to cd-adapco support - so you can also bug them for a copy of it.

/mark

usage:

star-backport [OPTION] file1 .. fileN

options: -h usage

Attempt to backport .cel/.vrt files from star4 to star3 format. The new names are prefixed with '__star3__' and suffixed with ''.

Unfortunately, cd-adapco provides no means for this.

Polyhedron and trimmed cells are not supported. Only the following cell types are attempted:

* hexa tetra prism pyramid (fluids and solids) * triangle quad (shells and baffles) * line and point

The program is provided as is, without warranty, of completeness, accuracy, usability, fitness for a particular purpose.

version 1.2 (2006-04-11) copyright 2006, Mark Olesen <Mark.Olesen@ArvinMeritor.com>

This is not bright at all and a totally pointless waste of time to say the least. Backward compatibility has never existed.

If a simulation has to be run into star 3, it should be created in star 3.xx or earlier version.

and then use upward compatibility if necessary.

"Don't-even-think-about-it" or "Dont'-even-go-there", as everyone says

you can see the consequence: "The program is provided as is, without warranty, of completeness, accuracy, usability, fitness for a particular purpose."

What if you incidently save your v3 model when playing with v4 without having a backup? cdsave was always an option to go back to an earlier version, which doesn't work any more.

Assuming that 'Yogurt' is not simply trolling, despite the very distorted interpretation of the usual legal disclaimer, I'd like to respond.

The point of moderate backward compatibility is to recover mesh geometries from newer into older versions. Examples of such would be for benchmarking wall-functions and turbulence models between versions, particular customer requirements, or simply for other programs that do not yet support the new formats.

whimsical...

William Blake is right: it is by mistake