New Hardware Design Software

An anonymous reader writes "AScribe is reporting that mechanical engineers from Purdue University have developed software they claim will increase the efficiency of creating a wide range of industrial parts. From the article: 'The new approach integrates the design and analysis processes, which are now carried out separately. Currently, the geometry of a part is first created using computer-aided design, or CAD, software. This geometry is then converted into a mesh of simple shapes, such as triangles or rectangles, which, when analyzed using a computer, indicates the part's strength and other characteristics. The painstaking procedure, called finite-element analysis, is extensively used in industry.'"

Comments

Boiler Up!

[Anonymous Coward]

Boiler Up!

Re:Boiler Up!

[Anonymous Coward]

In soviet russia, software design new hardware?

Making a 2 stage process into a 1 stage process

[BadAnalogyGuy]

The gist of it is that when one polygon in a mesh is changed, that the entire mesh does not need to be redesigned. The updated polygon only affects the polygons it is connected to, so on the fly analysis of the design can be performed without the need for a separate analyst.

It's kind of appropriate that a Chinese guy and an Indian guy are the ones coming up with ways to eliminate jobs.

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

Tis funny how widespread computer literacy has eliminated so many jobs. Was a time when your boss would have a secretary to type up his reports/memos/whatever, now he does it himself. That same person would typically be responsible for getting the boss his coffee.. now he has to walk to the kitchen himself. I find it funny that so many tech companies offer their employees free snacks and drinks as a motivational aid, but none of them offer someone to get those snacks and drinks.

Re:Making a 2 stage process into a 1 stage process

[TubeSteak]

In case anyone is wondering why the parent post is bringing the Chinese and Indians into this:

Information about the software tool is detailed in a research paper recently published online and will appear in the May issue of the journal Advances in Engineering Software. The paper was written by doctoral student Xuefeng Zhang and Subbarayan. The software tool is based on theoretical work by another doctoral student, Devendra Natekar. Natekar graduated in 2002 and now works for Intel Corp., and Zhang graduated in 2004 and now works at General Electric's Global Research Center.

I think what's most impressive is that Zhang did it in 35,000 lines of java.

Reinventing the way parts are designed doesn't seem to be "big and complex code" even if it is 35k lines. I imagine that porting the program to another language, then optimizing the code would result in a much smaller program.

Re:Making a 2 stage process into a 1 stage process

[riprjak]

You can only do away with the analyst if the CAD operator is competent to correctly interpret the results.

The problem with simulation is just that, it is *not* real world performance. Gaining accurate and useful data from FEA or CFD requires competent, skilled and highly trained operators (yes, all 3). Otherwise all you get are pretty coloured pictures.

In line analysis tools built into CAD for near real time analysis of models is nothing new. Its been around for the better part of a decade.

The difference here seems to be swapping your traditional mesh for an interlinked series of geometric primatives; essentially trading one form of approximation for another.

More power to them if this is indeed faster and equally accurate; but I doubt it will be, most of the difficulty these days is not mesh generation (damn near automatic) or Finite Element mathematics (sparse matrix solvers on modern CPUs make a meal of such calculations)... the time comes from the hideously complex multi-physics equations themselves; and these must remain common for equivalent accuracy.

Either way, the press release was typical oversold hype, but the tool SEEMS like it could be useful in low end design descisions.

err!
jak.

Re:Making a 2 stage process into a 1 stage process

[samschof]

This may help, however, there are other promising approaches. We no longer have to rely on low order approximations to curves for finite element analysis. In some applications, like thin shells, low order representations of surfaces tend to give unreliable results. Prof. Tom Hughes and his research group at U. of Texas have been working on finite element methods using NURBS. The obvious advantage is the easier and more accurate representation of surfaces, as you are more closely tied to what CAD programs are actually doing. I don't know if they are doing anything with regards to CAD to mesh software, but a NURBS based method would seem to have an advantage is this context.

Sam

Disclaimer: I come from a spectral/high-order methods background, so I have a bit of a bias.

Re:Making a 2 stage process into a 1 stage process

[drgonzo59]

My company's CAD/CAE/CAM application used to be 1+ million lines of code (C,C++, and FORTRAN) back in 2000, so 35K lines of Java isn't that much at all...

Re:Making a 2 stage process into a 1 stage process

[Anarchitect_in_oz]

I think that's because they realise that having all employees occasionally moving and interacting with other staff, was good for the health, well-being and productivity of all staff.

Allot of "new age" managers rarely get offices, some don't even have a desk, having laptops and a range of meeting spaces, so they can move to the work instead of it coming to them.

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

Bullshit. It's simple cost cutting. Shit, when I want a day off I have to fill in a form and carry it to HR. Was a time when I'd just ask my manager and *he* would fill in the form. Can't have that anymore, he's too busy typing things.

Re:Making a 2 stage process into a 1 stage process

[duffahtolla]

My company's CAD/CAE/CAM application used to be 1+ million lines of code (C,C++, and FORTRAN) back in 2000, so 35K lines of Java isn't that much at all

I think that that was why he was impressed. It is very small incomparison to other software in the same arena.

Re:Making a 2 stage process into a 1 stage process

[be-fan]

The code size isn't really representative of much. You can do simple finite element analysis in a few hundred lines of Matlab. Indeed, once you've formulated your element equations, solving them for a given mesh is quite straightforward. The real meat in most finite element programs is in the mesh generator and the solver. Building a mesh generator that can create good meshes for a wide variety of situations is a non-trivial task. The same is true for creating solvers that can handle (quickly) the extremely large systems of equations resutling from complex problems. A lot of code gets spent on additional features, like allowing the code to selectively solve parts of the problem, etc.

So depending on the sophistication of the program, 35,000 lines could be a very small codebase, or an unnecessarily large one.

Re:Making a 2 stage process into a 1 stage process

[EndingPop]

It's hard to tell from this article, but it is unlikely that the mesh was actually generated by this program. They probably create an input file with a commercial code, and this program just runs the topology optimization.

Re:Making a 2 stage process into a 1 stage process

[Ohreally_factor]

I didn't quite understand that. Could you give me a car analogy?

Re:Making a 2 stage process into a 1 stage process

[jonwil]

The obvious question is why dont they just make the form electronic so you dont have to fill it in and take it to HR, you sit at your desk and fill it in and then it gets sent to HR automatically.

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

Because that would require IT to actually do something.

Re:Making a 2 stage process into a 1 stage process

[AuMatar]

That means your company is doing things the wrong way. At my last 2 jobs, when I wanted a day off I went to 1 webpage, clicked the day, and hit submit. About 30 seconds if the intranet is slow. Another 60 seconds to type an email to m group saying that I'll be off.

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

Well, duh. It's not like I'm suggesting that companies are actually run intelligently.

Re:Making a 2 stage process into a 1 stage process

[Anarchitect_in_oz]

Of coarse it's simple cost cutting.
It's making it buzzword compliant that's the hard part.
Mostly by trying to make people sick less, and wasting hours scheduling meetings for things that can be resolved in 5-minute casual encounters.

Re:Making a 2 stage process into a 1 stage process

[Anonymous Coward]

So the MESHING doesn't have to be done for the whole structure, but solving the stiffness-matrix, which is the time-consuming part of the analaysis, has to be done for every small change...

Brilliant, saves a couple of minutes on a few-hours-analysis.

Re:Making a 2 stage process into a 1 stage process

[iceburn]

That's actually a pretty good point. Technology doesn't actually reduce the overall amount of work, it just allows it to be dumped on somebody else. As an overworked sysadmin, I know all about getting dumped on. I'm writing this as I sit here babysitting a server at midnight. A few short years ago the tasks these servers preform would have been done by somebody else... with a pencil, paper, and a filing cabinet. Filing cabinets don't usually crash in the middle of the night, either. Nope, I'm not bitter. :)

Re:Making a 2 stage process into a 1 stage process

[Bush Pig]

I had this at a previous place of employment - based on Oracle Financials. It was a complete headfuck, because it didn't quite work.

Re:Making a 2 stage process into a 1 stage process

[AmericanInKiev]

I'm fairly certain that the part designer would have to understand the goal of the part - that is to hold this or that in position against the stress, thermal expansion, and vibration induced under the application scenario. A more interesting next step would be to mutate the part and hunt about for a more highly optimal shape (generally by removing unnecessary material where it is unwarranted). In fact there is an early American poem about such optimization which describes a horse carriage so perfectly balanced that no part wears out first, and the rather poetic end that the carriage runs until the moment of its demise, whereupon it is reduced in an instant to a state of unrecognizable ruble.
AIK

Re:Making a 2 stage process into a 1 stage process

[Ig0r]

It's the One-Hoss Shay!

Re:Making a 2 stage process into a 1 stage process

[poot_rootbeer]

That means your company is doing things the wrong way.

All companies do everything the wrong way. It's not worth having a superiority complex over.

At my last 2 jobs, when I wanted a day off I went to 1 webpage, clicked the day, and hit submit. About 30 seconds if the intranet is slow. Another 60 seconds to type an email to m group saying that I'll be off.

You must not have been a very essential part of the team if your absence didn't require prior approval from a higher-up.

Re:Making a 2 stage process into a 1 stage process

[AuMatar]

You must not have been a very essential part of the team if your absence didn't require prior approval from a higher-up.

I'm a programmer- just a general employee. As long as I'm not taking off around the time of a release, it damn well shouldn't matter. I get x days off per year, and will take them when I please. You don't like it- put it on my year end review, and I'll happily tell you how much I don't give a fuck. Hell, even if I was tech lead, so long as it wasn't a day I had an important meeting or deadline it doesn't require approval. Thats why we have personal days.

Now if I'm taking a week off rather than a day, you have a point. Its still a matter of telling my boss a few weeks ahead of time, not requiring approval. I tell them when I have plans with enough time to push back schedules or reassign manpower. They don't get an option of approving or not- I won't be there. Depending on how complex my plans are (other people taking time off at the same time, or a date I have to be somewhere), I might honor a request to change the date. But thats exactly what it is- a request I'll decide to honor or not. I own my time, not my employer.

Re:Making a 2 stage process into a 1 stage process

[iamlucky13]

The designer is supposed to understand that. Constraining, loading, selecting mesh types, and interpreting results is somewhat specialized, though, which is why the FEA work is often done by seperate people than the original design work in the first place. Simplifying the meshing process saves effort for those who do both modeling and FEA, but not in offices where one engineer does the model, and another does the FEA. Also, drafters are generally concerned soley with geometry, rather than analysis, so it's common to have a drafter design a part, and an engineer verify it.

I do think this interesting for those who do both, however. From my own experience with ANSYS, it's rather a pain in the butt to make modifications to parts, especially since the modelling tools in ANSYS do not lend themselves well to undoing steps or changing existing features. It's very common to build a model in other CAD software, then import the model into an analysis program like ANSYS. One thing that concerns me though, is that you might end up with a screwy mesh near the interface of modified and unmodified sections of a part that could improperly effect the analysis. I would hope they would've thought of this somewhere in their 35,000 lines of code, however.

Re:Making a 2 stage process into a 1 stage process

[drinkypoo]

Hey! Stop encouraging that shit! :D

Re:Making a 2 stage process into a 1 stage process

[drinkypoo]

Right, then they would be Information Services. I far prefer working in Information Technology departments, where you just play with shiny toys all day.

Re:Making a 2 stage process into a 1 stage process

[somersault]

Huh? We still have secretaries to do things like booking flights etc, and they take coffee up to meetings etc. I wouldnt exactly say everyone's boss is computer literate either, in fact the opposite could be true - the higher you go, the less people are concerned with how machines work (unless you're working in an IT company), and just get their underlings to fix their messes.

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

Heh, yeah right. In my company, *I* have to call American Express Travel and arrange a flight. Very few companies have secretaries that do those traditional roles anymore.

Re:Making a 2 stage process into a 1 stage process

[somersault]

wow, seems pretty strange to me.. well we have people going all over the world on seabed dredging jobs and stuff, so it does make sense to have a couple of people coordinating everything

Re:Making a 2 stage process into a 1 stage process

[QuantumG]

It makes sense yes, but if you can justify the cost cutting by claiming that you're "outsourcing" the position then you'll win brownie points with your bean counter boss and that's how things get done in corporate environments.

In other news...

[Anonymous Coward]

Linux is STILL for fags.

Damn

[Eightyford]

Damn, I thought there was some New Software Design Hardware. Ah well.

This is nothing new at all

[Anonymous Coward]

I am a Mechanical Engineering studnet in my last semester and I have to say that this is nothing new. There has been a program for years called Pro-Engineer with the add on Pro-Mechanica that does this more or less just as described and ive personally used to design lightweight parts for a racecar.

Re:This is nothing new at all

[Anonymous Coward]

"studnet"

Hmmm, Very Interesting...

Re:This is nothing new at all

[MichailS]

I think you should read the article.

Re:This is nothing new at all

[sadtrev]

What's new here is that they are not remeshing the part after a design change. There isn't enough meat in the aricle to be able to tell whether their approach can be applied to all engineering problems where discretized analysis software techniques are now used.

I think that a better approach to reducing the turnaround time for designs is to break down the distinction (that tends to exist in large engineering companies) between engineers, designers and draughtsmen. To a large extent the CAE software companies are doing this. There is still some way to go, however, before the existing generation of FE and CFD software can redo an analysis after a design change without further intervention from the analyst.

Re:This is nothing new at all

[makinola]

See also SolidWorks and CosmosWorks for design+FEA analysis packages. As argueably the best mainstream software used in the mechanical design industry, SW comes with tons of add-ons, COSMOSWorks being one of them, it's also a lot cheaper than most of the packages described here. This article describes a process in which the mesh is not re-created on part redesign, but since a new problem must be created once any force or input to the part is changed, you probably would not save any time in the long run. I see this as just a small evolutionary step in FEA analysis and it's integration into the design process.

Re:Pffft

[Anonymous Coward]

Yes, but on Digg you don't get the suicidal tendencies that one gets from even considering clicking on a link to potentially view comments. LOL BEOWULF CLUSTER XML FRAMEWORK!!!11one*={'o', 'n', 'e'};

VOTE +5 ...because you know what I say is true. YOU KNOW IT.

interesting..

[Pavel Stratil]

I am really amased that noone came up with anything like this before. The approach suggested is commonly used in computational physics (i.e. modelling EM fields, etc.). Still, the finite-element analysis is here to stay. While the method can suggest in what direction to go in terms of design can in no way replace the 'holistic' view FE analysis offers.

Re:interesting..

[blueboy31]

At my university (Michigan Tech), they just restructured the general engineering classes. All engineering students use to learn IDEAS, the CAD software designed by Nasa. Now they are teaching us Unigraphics' NX 3, which does modeling and stress tests for parts all in one program. It's pretty slick, but I believe the automotive industry has been using similar software for a while now. I don't think it's that new of an idea...

Re:interesting..

[HotNeedleOfInquiry]

Yeah, well it's too bad you don't get the Discovery Channel.

They had a show documenting Boeing's mechanical design process. Same thing exactly. And I'll bet Boeing is using something other than 35k lines of Java to do it.

hardware design

[doubleshot]

The title makes reference to hardware design, in most of us in the technological world this seems like it might be a new PSpice or the likes. However this is a mechanical engineering software program, not a hardware design program. Agh, I got all excited for a second.

I'm not sure I understand

[corngrower]

What's this fine-eyed elephant analysis they're talking about?

Re:I'm not sure I understand

[Miss Emily Litella]

I was about to ask the same thing. Thank you, Mr. Hornblower.

Re:I'm not sure I understand

[Mr Europe]

FEA :
Finite Element Analysis
Materials internal stresses can be calculated by modelling the part as small elements, each of which have a uniform stress and a simple form.

In areas where the greatest stresses are anticipated the elements are smaller and thus "the net" is more fine-grained. The problem is that the solving time of the matrix calculations increases with the amount of elements and calculation times can be hours or longer. It takes a pro to say where the smallest elements are needed.

This of course depends on 1) how complicated the part is, 2) how accurate results are needed, 3) how fast the results are needed 4) what kind of hardware (and sw) does the calculation.

It's NURBS for stress analysis. Here's the paper.

[Animats]

Here's the technical paper: Constructive solid analysis: a hierarchical, geometry-based meshless analysis procedure for integrated design and analysis. This extends finite element analysis from rectangular cells to elements defined by NURBS surfaces. Difficult mathematically, but if you set up the problem that way, fewer cells and less number-crunching is required. Very nice.

This may have applications for soft-body physics in games.

So you're telling me...

[Temujin_12]

So you're telling me that there isn't "New Hardware" that "Designs Software"? Whew! I thought I was out of a job there for a sec.

nothing here move along...

[Anonymous Coward]

Slashdot, please stop reporting lame hyped-up press releases.

Re:It's NURBS for stress analysis. Here's the pape

[afidel]

How is this new? CATIA v4 already supports NURBS curves in later revisions and CATAI v5 uses it as the core element.

Nothing new

[sdo1]

I don't see anything new here. A bunch of enterprise CAD programs including NX (formerly Unigraphics) and Pro/E have similar design/FEA capabilities. There are typically many options for optimizing certain aspects of the design be it strength in a certain area, weight, etc. I haven't used them, but I'm sure higher-end versions of AutoCAD and the like have similar capabilities.

Maybe I missed something, but I didn't read anything in that article that isn't already done in industry every day.

-S

Re:Nothing new

[EndingPop]

This particular paper isn't new, and neither is design optimization. NX, Pro/E, ABAQUS, ANSYS, etc. are all just analysis tools. True design optimization, that is, the ability to find non-intuitive solutions to problems with literally hundreds of variables is a burgeoning market. This article is surprisingly sparse with details, but topology optimization is powerful in that it can bypass the mesh generation because it uses the same mesh with every analysis. It does several analyses (sometimes hundreds) and each time it changes the "density" of each element until you reach a predetermined weight reduction while maximizing stiffness. This is actually only a tool for preliminary design, since it only deals with strain-energy; it never even looks at stress. In fact, the solution is independent of the actual magnitude of the loads applied, only the location of the BCs.

In other news...

[Anonymous Coward]

Linux is STILL for fags. part2.

Re:Nothing new...maybe

[Overzeetop]

The press release sounds like it was written by somoeone who has never worked in an integrated production environment. Pro/E has had the end-to-end capability for close to a decade (though at $50k a seat). I didn't like (or didn't trust) the meshers in the 90s, and solid elements also didn't always behave properly, especially at boundries, so I did most of the work by hand. Parts of the writeup seem to suggest that the FEM created would contain boundries that would remain static, and the re-meshing would only occur in defined areas, reducing the amount of time to create and (ideally) invert the matrix on the next iteration (sorry it been a long time since I did FEM, excuse the glossyness of that statement). If thats the case, you're really only talking savings in computer time, which for common manufactured parts is probably down in the noise, costwise - who cares if you have to solve a 1e4DOF model if you can do it in a few seconds? The abstract of the actual paper, linked in another post, seems to imply that they're not using FEM to solve the optimization, which would be revolutionary.

Re:It's NURBS for stress analysis. Here's the pape

[be-fan]

I don't think CATIA actually does finite element analysis on NURBS. It may use it as the core element for handling geometry, but I think they're decomposed to more traditional elements for the actual FEM process.

how will this help?

[stephenMF]

"we are trying to unify both the CAD design and analysis so that they are carried out concurrently." I'm not sure how this is going to help me. When I design a mechanical part, necessary dimensions and geometries are the very first things that I consider. After that, *if* FEanalysis is necessary, it's no big deal to just re-do the mesh and start over if I change part of the design. It only takes a few seconds on a p4 3.2ghz machine. My point is that I *rarely* need to use FEA software... if you have a fundamental understanding of the strength of materials and some common sense, then most design decisions are easily made. But then, I'm only a machine/tool designer/programmer/controls engineer/gopher... not a product designer, so I am not the expert here. Show me some fatigue simulation software (it may exist) and I'll be impressed. In fact, I may even go ahead to make the purchase order.

I think I'm already doing this

[JoshWurzel]

I'm a mechanical engineer who uses Pro/Engineer at work. I design geometry and then use built-in software (Pro/Mechanica) to analyze the parts every day. How is this new?

At first it sounded like it was simultaneously computing something about parts as you design, but this just isn't possible or even useful as far as I can tell. The whole point of having a separate step is that you can define different loading scenarious. "What is the strength of this part?" is a meaningless question. What you want to know is "how much force will it take to displace the end of this beam by 2 inches?". There just isn't anything to compute on the fly.

So again, what is it that this is doing? The integration of FEA modules in CAD is already pretty seamless. It saves the analysis features as part features.

Reading the article, I see some interesting quotes:

"The way it is now, the same CAD software used to make the shape of the part can't be used to analyze the mesh" - flat out wrong. Pro/E, ANSYS, and NX all have integrated FEA modules. Its a separate module, true, but you just push a button in the GUI to change your mode from modelling to analysis.

"After the designer designs the object, it is thrown over to the analyst, and the analyst says, 'OK, I think, based on my analysis, that your design has to be modified this way,' and then throws it back to the designer, who makes the modification" - sorry, but this is not the way things should be designed. The company I work for is small, but in general there are only a few designers who have *extensive* knowledge of manufacturing capabilites and extensive experience. This enables them to make good judgements when designing parts. When they do make a mistake, the engineers have sufficient CAD skills to make the change themselves. There is very little "back and forth". Now, if the analysis indicates the design is completely non-viable (i.e. "Not strong enough now and never will be!") then the designer goes back to the drawing board. But that's an entire re-design and not nearly as similar to phone-tag as the article makes it sound.

And don't even get me started on the fact that its written in Java...it may be a fine language for web apps, but I've used FEMLAB and if someone else writes a CAD package in Java I may well be forced to hit them in the face with a trout.

Re:I think I'm already doing this

[EndingPop]

The difference is that you can use software packages to optimize your design based on any design variable and response you can come up with. That's the whole point of design optimization. Check out http://www.redcedartech.com/ who make HEEDS. You set up your design variables, thickness here, length here, what material is this made of, etc. and then say that the max. stress must be less than this, the first natural frequency must be higher than this, the deflection of this point must be smaller than this, etc. It performs an intelligent search of the design space, even when you have a huge number of variables.

Design optimization right now is where FEA was 15 years ago, on the brink of becoming not just huge in the industry, but ubiquitous.

Re:I think I'm already doing this

[Anonymous Coward]

When you use pro mechanica, you are using finite elements underneath. You are just not seeing the mesh. This is very different from the meshless analysis they are talking about. Pro mechanica uses P-elements (this package used to be called RASNA, then PTC bought them) so watch out for sharp internal corners. I think it is ok for a first look analysis but it could have significant errors if you are not careful.

from someone who knows a bit about this stuff

[Anonymous Coward]

so the reason this hasn't been thought of before is because there is an issue of 'mesh quality' as it stands it is still a huge issue in and of itself... The whole idea of not re-meshing could easily lead to poorly formed meshes, so thier technology would have to have some way of evealuating just how much of the mesh needs to be recalculated, with it sometimes being the whole thing, and how many steps you can do without global recalculation....
to give you an idea if you just change 1 polygon without changing the others, then instead of having all nice uniform polygons you may have nice uniform ones and 1 that is ten times the size of the others.
that is the stuff their paper will have to prove...that they keep mesh quality without increasing analysis time, something that can only truly be done with empirical tests.

Re:It's NURBS for stress analysis. Here's the pape

[Spy Hunter]

Wow, way to cut through the press release BS and find the real story. Thanks. Is it just me or are we getting more and more of these breathless press releases and gushing blog wackos lately? They always seem to be doing their best to bury the real (often interesting and worthy) story under a mountain of hype and irrelevant tangents.

Yes...Its been around for ages...

[Tmack]

The fundementals behind it have been in use for a long time, and trace back to the basics of engineering (software engineering as well): define the inputs and outputs, and design an element to meet those needs. In this case, the inputs/outputs are forces, and the elements are fragments of a structure. In basic particle dynamics (physics), the same principals are applied to individual static bodies. Finite element analysis simply takes the same basics, throws in a bit of calculus with integrals to the point where you can calculate the exact forces/moments at any point in a solid body. For paper work, the number and size/complexity of equations involved requires summing over a limited series of rather large finite elements to get an estimated result rather than solving all the integrals to get a perfect result (think low-polygon surfaces, finding the forces at each boundary).

Software has been around for a long time that makes use of the computer's ability to solve things quickly and recursively and allows much faster solutions with much smaller elements. I used the IDEAS package back in '98 at GaTech (IIAME), and I know we had had it for a few years prior to that, and Pro-E does similar stuff and had been out for a while as well. MatLab even has some stuff to do simple solutions, tho its not a 3d modeling engine like IDEAS and Pro-E are. The only advantage hinted at with this one is that it apparently does not need to re-calculate the entire structure based on a change to a few polygons. Not sure how advantageous that is, unless its a minor change that doesnt really affect the overall structure, where the model is large enough that re-calculating it would take forever...

Tm

Obligatory reference

[LouisZepher]

"...The painstaking procedure, called finite-element analysis..."

I wonder when we'll have infinite-element analysis... Of course, whoever manages this, will most likely be lynched for being a smartass.

Re:Obligatory reference

[Anonymous Coward]

There are infinite elements in the FEM, mostly for modelling material effects of large objects (pressure of the ocean on a vibrating object etc.)

Classical material mechanics can also be regarded as "infinite", where integration is performed in closed-form. Usually beam elements with simple supports can be analyzed by one element of infinite accuracy, errors usually come fome the simplification of the boundary conditions

Re:Obligatory reference

[LouisZepher]

I'll admit it was a lame joke, but it was still a joke. That was interesting though...

Java Free calls on a blackberry

[Anonymous Coward]

http://www.minowireless.com/

Re:Pffft

['aspies' are retards]

Finite element modeling is the de facto analysis tool for numerous industries. When you use finite elements, you convert the complex differential equations that describe the physics of the part's behavior into simpler algebraic equations that the computer can solve. It's a powerful method because it enables you to take any complex problem and solve it.

To describe the geometry, you take this complex object and break it into primitive objects like cubes, spheres or cones. With this approach, if they only modify some portion of the part, they only modify the primitives directly associated with that portion they are changing and not all of the primitives. If they only change the shape of a specific hole in the part, for example, the rest of the primitive objects are the same shape, so why should they need to reconstruct the whole geometry and remesh the whole geometry?

I see it as a "hierarchical, constructive, meshless procedure" because it enables engineers to analyze the changing design of a part without recreating the complex mesh of elements. The way it is now, the same CAD software used to make the shape of the part can't be used to analyze the mesh. But now, the same CAD software or some similar CAD-friendly software will be able to do the analysis, and in a much more efficient manner because there is no remeshing.

Design first, then analyze

[Anonymous Coward]

Currently, the geometry of a part is first created using computer-aided design, or CAD, software. This geometry is then converted into a mesh of simple shapes, such as triangles or rectangles, which, when analyzed using a computer

So hardware guys design first and then analyze, just like us software guys? Cool.

Hey this **is** the first one

[EmbeddedJanitor]

this is just a dup from the 1980s.

You heard it last on /.

People's lives aren't meant for work.

[aquadivina]

Unless they want to do their work.. Ultimately, machines will do all non-creative work, just look at Moore's Law.. Efficiency is good.. It's *society's* duty to sort the social implications of it all out. Companies duty is to do what they do the best possible way they can.. which means doing it with efficiency. What we should be doing is training people to *think* not how to do specific jobs. Those jobs will disappear, its a given. If people have the skills and the inquisitive minds to understand and embrace change, we will do well. If not, we'll fail. Training people just to do todays 'jobs' means that their skills will be obsolete not long after they gain them.. OTOH, if we teach them the fundamentals they will need to understand tomorrows skills.. we'll do okay.. But we have to accept that the whole job thing is probably a 20th century thing.. the 21st century will probably be more about project based employment and gradual, accellerating replacement of people by machines..

Re:People's lives aren't meant for work.

[Anarchitect_in_oz]

Using current and even foreseeable technologies "machines" can only really replace precision jobs. Anything that requires even a mild degree of flexibility or reaction to prevailing conditions is going to need a person.

While agree that training people to use their innate flexibility is good for all concerned, and employment is certainly moving towards project based (although we may still more of the team moving,than the individuals, already the M.O. of most professions). The Efficiency of machines really has to be questioned, i think we passed the point where what we know as machines offer any real gain a long time ago. Indeed every machine that replaces a worker these days should also require a population reduction of least 2 to cover the energy footprint.

In order for a technology to replace all but the interesting jobs we are talking an event as significant as the the engine on the Industrial Revolution, or computers on Information. Given how long those two technologies existed as ideas before reaching tipping point of Revolution. Is there anything on the horizon, other than Television, capable of producing efficient mindless drones(machines) capable of filling the need for all the mindless jobs. Well at least in our lifetimes.

Re:It's NURBS for stress analysis. Here's the pape

[dbIII]

This extends finite element analysis from rectangular cells to elements defined by NURBS surfacesTypically you can have different shapes for different areas in a mesh - either in things like ABAQUS or in your own code. Obviously it's time for me to read the article to see what is new.

Re:It's NURBS for stress analysis. Here's the pape

[Anonymous Coward]

So you're telling me I have to pay $30 to read an abstract?

Replicators

[Slashdotgirl]

"Replicators", coming to a store near you, Very Soon.

I think I used to do this

[Inda]

"This geometry is then converted into a mesh of simple shapes, such as triangles or rectangles"

When I left the industry 8 years ago we'd already moved on to solid modelling. Meshes are so 20th Century.

I had a vast array of tools that seemed to do far more this new software is boasting. Nothing to see here as the Slashdot crowd would say.

I would love to get back into CAD work. Would anyone like to offer me a shot at a job in the South West of England?

Re:IDEAS

[meburke]

I couldn't find anything on this package. Could you provide a link?

Thanks,

Mike

Making a 10 stage process into a 1 stage process

[Becquerel]

Either way, the press release was typical oversold hype, but the tool SEEMS like it could be useful in low end design descisions

Nothing to see here

I absolutely agree, current academic research in the field is looking at the solution of 1e6 element problems in seconds in order that parameters can be changed in real time by someone with the skill of both an analyser and designer. 1e6 element would allow the design of the whole product, say a car, to be worked on in real time, by the aerodynamics, electical, comfort, safety, etc departments collaberatively. Which would appear to be a step on from this press release. For example see this paper written by a colleague of mine.

There must be some novelty in the work in order for it to be worthy of a PhD thesis...but this is not ground breaking so far as i can tell, and certainly not for the reason in the front page caption

Re:It's NURBS for stress analysis. Here's the pape

[gardyloo]

This may have applications for soft-body physics in games.

      Translation: jiggles

The Sofa/Stairwell Test

[Anonymous Coward]

Does the software manage to get the sofa past the turn in the stairwell is the determining question.