CREO Simulate eigensolver?

[JBlackhole]

Does anyone know which eigensolver is used by Simulate?

I can't find anything in the doc on the matter/subject.

[KenFarley]

Such details are likely of little to no concern for users of the software. I don't know if it's helpful, but Creo Simulate was based upon Mechanica (it used to be called Pro/Mechanica). Perhaps you can find some sort of esoteric details in past discussions of that package, but I really doubt it.

The actual algorithms used are likely proprietary.

[JBlackhole]

Thanks. Will do some digging on the old mechanica stuff. I am very surprised that such details are likely of little to no concern for users. [quote]. Might be for a designer who just clicked on buttons but any professional (structural) analyst will pay attention to such things. How the method is implemented will indeed proprietary but it's quite possibly based on well-known method (lanczos, etc). 

[KenFarley]

Most of the past discussions I've seen with Mechanica were concerned with how the engine runs dealt with the rather unusual (for the time) approach to the meshing, its adaptive methods, shape of elements, etc. I'm not a dedicated 100% structural analyst, I use the program as a tool to get answers to things like "Is this structure sufficiently rigid at this temperature to hold a wafer with less than an allowed amount of deflection?"

There have been, in the past, users with very esoteric knowledge of how this tool works and its advantages/disadvantages with respect to other software available in the world, but I've not seen anything from those folks since the multiple changes of platform for this forum made the usability of it much less pleasant.

The only discussion on the PTC site about eigenvalues, eigenvectors and the like is in MathCAD. Which is to be expected since that piece of software is used for pure mathematics as well as more pedestrian endeavors like what I do.

[JBlackhole]

Thanks. I have hit nothing so far. It seems there's only 1 solver anyway. As far as I can see there is no option to select a "eigensolver". Not sure if the eigensolving is efficient if one does dynamic analysis with Simulate then .I guess it's not top of the list of issues. As you flagged there are other issues that needs attention; adaptive methods, etc. Mind you one can turn on the Sturm sequence check so there's something

It's seems that will have to go on the "don't know yet" list and move on. It's even possible that sort of knowledge might be superseded as one moves on to CREO Ansys Simulation.

[Chris3 · Accepted solution]

PTC support has documentation on this. You can request it. If they initially deny having it keep pushing. They do. Attached is an example.