Proper training is a must with FEA. I have been using Algo since 1994 and
Plassotech within Inventor since it was released. If you set up your model
correctly you will receive a correct answer. They could handle
loads/constraint selection better (currently requires Split-Surface) and
better mesh control. (these may show up in the next release)

As Richard stated, you still need to understand the results and any
deviations from conventional training.

Blair

--
IV2010-Pro-sp1
Vista Business 64bit Sp2
Core i7 @ 3.07Ghz
12Gb DDR3-1600, 4 x 60 Gb SSD - RAID0
Quadro FX3800 - 191.00
SpacePilot 3.7.18 / 6.7.13
AVG8.5
"Richard Hinterhoeller (AISS 2010 Sp 0)"
wrote in message news:6283228@discussion.autodesk.com...
Almost. Because Autodesk bought the FEA program from Plassotech and dealt
with former owners (like yours truly) in an honourable way I have 4 years of
exposure to the FEA portion of the program. In addition my background
includes having learned the equations used for FEA from the late 1970's.
I've used other programs in the interim.

FEA is an incredibly powerful tool. If you ask the right questions, you'll
get the right answer.

My suggestion is that one approaches this feature slowly. When the
solutions deviate slightly from conventional training, examine whether the
solution was wrong or whether the problem was miss-stated. If the solution
seriously deviates from conventional training, take a jaundiced view of the
way you modelled the problem. If you don't know the difference, you have no
business using a tool that can provide a catastrophically wrong answer to 16
digits of accuracy.

Richard

wrote in message news:6283183@discussion.autodesk.com...
Has anyone used the FEA Study & Simulation Software from AutoDESK.

If so, please let me know what you think with regards to ease of use &
understanding the output.

Thanks in advance.