Hello M. Frisch - may i call you 'M'.
Anyway, thank you for the further explanation in translating the Profs book. Yes, i can see clearly
the big problem of screenplots and pictures etc. and it is
not, the most interesting of work to do either - well it wouldn't be for me. I would be very pleased to cross the marathon finish line, but, to do the run without a cheering crowd... not easy at all.
Any Other Business.. well, i feel like a type- please forgive me.
With the release of V2a i'm amazed at the speed increase!
I haven't been using the Pardiso solver before the advent V2a but now as i gain experiance with Aurora, i experiment a little more and more.
I'm staggered to find that Pardiso solver is
x 14 faster than V2a SICCG with the same results! The theory manual says SICCG is 'fast' and Pardiso is 'very fast'. Please change that to ' MIND BLOWINGLY FAST' or 'SWOOSHHH' or 'SNAP' or 'PRONTO'.
I'm working on a normal core2 duo 32bit win7 4gb platform with complicated, feature rich, machine parts with holes, cross holes, small chamfers and fillets etc. - yes, all the small features left in(!) and not removed and still it computes in JUST.......26 secs.
I expect 30 minutes,
not 26 secs.
26 SECONDS! *********************THAT'S JUST AMAZING!***********************
However, as a word of caution here, i import .stl into aurora from the Netgen meshing process as proposed by superuser 'selopez'.
It works for me
everytime and the upfront effort pays back massively. For anyone
'having issues' with .stl or .stp importation, or with how Aurora makes a mesh, i would
recommend experimenting with the procedure as it has completely changed my experience with FEA for the better.
For anyone who doesn't have importation issues.. well, i wish i had your CAD software. Mine is 5 years old now!
I see 'FBook' likes are steadily increasing and rightly so.
Thank you M and thanks to all the Aurora team. I'm a very happy Easter Bunny..
Regards. Tim.