Future of MSC ? 4

[johnhors]

Poor old MSC have been taking quite a battering in the last 12 months, firstly with UGS terminating Nastran for Windows, then inviting N4W users to ditch MSC and take up Femap bundled with NX/Nastran, and more recently UGS have extended their agressive marketing campaign to target MSC/Patran users.

Now MSC announce a workforce reduction in a bid to "streamline" their business operations.

Their future strategy appears to be largely reliant on the all new (?) Bangalore developed SimEnterprise (SimExpert?) multi-disciplinary software package (which the MSC web site over hypes to the point at which you begin to question the validity of it all). However this product has still to hit the masses yet and has had a very long gestation peroid. I learnt of this product probably two or more years ago and saw a demo movie of it back in October 2005.

But I suppose there are enough hard core Patran/Nastran users (especially in aerospace) to keep them ticking over at least for a while, until they discover better things ...

 

[JakeRAD]

it's all a question of price/performance.
small engineering companies have no need in getting the whole blown nastran package, with MARC and DYTRAN, or getting nastran MD.

On the other hand, they are offering MSC/FEA and MSC/AFEA for a price similar to the one of neinastran/femap, and they have much more experience.

 

[GregLocock]

Not just aerospace. Automotive use of high performance Nastran solvers is on a (horrid phrase) exponential growth curve as well. Their own pre and post processing packages have always been a bit odd, so yet another attempt at an integrated package isn't surprising. It would be interesting to know what proportion of their income derives from the solver rather than those packages, but I'd guess they aren't telling.

Bear in mind that they do /far/ more than just FEA, and seem to be attempting to become the Microsoft of automotive simulation software.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JakeRAD]

another aspect that I forgot, is that NASTRAN and MSC products in general offer the most comprehensive solution for composites, including laminate modeler, and it's integration with compro, and fibersim.

and you know about composites.. they are the present/future

 

[johnhors]

JakeRAD,

Sorry, but there are several other vendors that also offer very comprehensive composite solutions (Lusas, Samcef ....) with pre and post processing facilities to match. Unless you've tried them all, I'd step back from making a judgement!

 

[fkmeyers]

My guess is that UGS and ANSYS and Noran will continue to take business away from MSC. This would be MSC's 3rd major layoff but not the last one I bet. MSC gets most of their revenue from big aerospace and automotive companies so if UGS gets at those accounts, MSC is done for. Otherwise they should be around for a long time but in a diminished capacity. SimEnterprise (SimExpert?) is a huge failure. There was a time when MSC was strong and the market leader. They were respected in the industry. Things seem to go down hill shortly after they purchased PDA.

JakeRAD, too bad MSC is not hiring, you could be their new spokes person!

 

[ThomasH]

A question:

When MSC bought PDA, was it that purcase that gave them control of Patran? If so, before 1994 (?), didn't MSC have any pre/post in-house?

Regards

Thomas

 

[fkmeyers]

When MSC bought PDA they had just released Patran 3.0. Patran had the market share with most big aerospace companies. They bought the company and totally controlled it. They also bought a lot of headaches with Patran 3.0. MSC had previously attempted several times to either purchase or develop a pre/post but that failed every time. Yes, they had in-house pre/post but they were not popular.

 

[johnhors]

Frank

Wasn't Patran 3.0 hugely unpopular at the time with many users of 2.5 reluctant to upgrade?

What have you heard about SimEnterprise?

Some time ago MSC funded Fegs Ltd to write a HEX mesher for arbitary volumes, which spawned their medial object technology before MSC pulled the plug.

 

[fkmeyers]

Yes. Patran 3 was very unpopular because it was so different from 2.5. There were lots of enhancements but many features that power users used were removed. No command line for example. Also, it was very unstable and crashed often. That part was a big problem for many who used it day long. With Patran 2.5 we knew where the bugs were and what would crash it and we avoided those areas. With Patran 3.0 it was all new stuff, a completely new package requiring weeks of training to get proficient. It was slow too since it was written in PCL which was useless to most of us.

SimEnterprise does not seem to be what MSC had in mind, or least what their customers did. Not an expert on it but have not heard anything good about it. What you can expect from MSC now is development and support with an Indian accent.

 

[ThomasH]

I have heard that at some point MSC had the opportunity to buy Femap but declined. I guess that UGS "attack" on Patran and indirect also on MSC.Nastran must hurt. I mean, if oyu like Femap (or dislike Patran) UGS can offer a complete package in the same manner as MSC.

I think that MSC will be around but I'm not sure it's because of technical superiority.

Regards

Thomas

 

[outpost11]

MSC have really dropped the ball in terms of their core programs (Patran and NASTRAN). Minimal enhancements over the years have cost them dearly. PATRAN is not a native PC product and as such was doomed from any significantr enhancements unless a re-write was carried out (come in Bangladore). MSC/NASTRAN is still a dog. Memory handling is awful, huge disk writes with scratch files, no contact option other than sol 600 (Marc) but you need a campus license for that option. The cost structure is just another thing and what really got me was their arrogence in believing they could do anything to shaft customers. Their whole direction with SIM Enterprise is to accomodate the big auto companies, a direction that is relatively new (now over 50% of their analysis business). They have always being into aerospace only but this direction of seamless integration of multi disciplinary tasks is primarily focused at automotive conceptual to final design processes. This approach has left many other companies with little option but to look elsewhere.
UGS NX/NASTRAN obtained v2001 and have carried out quite a few enhancements since. FEMAP is a native windows product that allows considerable customization with excellent macro's and API programming possibilities. I have used MSC products for 20 years but we made the switch recently to UGS NX/NASTRAN and FEMAP. We are a NASTRAN house since we deal with NASA.
Personally, I think ABAQUS is a far superior finite element program (both linear and non-linear). ABAQUS CAE (pre and post) still has a way to go but it has improved significantly over the years.

 

[olofg]

I use MSC.Marc for contact analysis and I get the feeling that MSC.Software seems to focus on their new multidisciplinary codes instead of developing old codes. The selection of available courses for marc users is limited and I would like some suggestions of alternative finite element programs. Today my company have 2 marc licenses and 1 LS-dyna license. Can ABAQUS compete with the contact options in Marc? How is the pre/post processor in ABAQUS?

Regards

Olof

 

[fkmeyers]

ABAQUS is a very capable program. I recommend it. It can be difficult to use however. We also use NEi Nastran which has much better nonlinear than MSC.Nastran (without SOL 600) including true surface contact and specialized elements.

 

[outpost11]

ABAQUS or MARC ?
I have used both programs in the past for contact problems. In fact we bench marked MARC with ABAQUS and found MARC was not able to handle some of our complex non-linear problems. These included non-linear geometry, material, and contact. In all instances we got answers from ABAQUS and in comparative runs it performed significantly faster. Marc was easier to define the contact problem by defining "bodies" as opposed to the slave/master node approach in ABAQUS (I think the more recent version of ABAQUS now has body definition). The issue with ABAQUS may be the pre-processor CAE. It has some nice features but it is also missing some of the more useful features that PATRAN has. We used ABAQUS with PATRAN. When it comes to complex non-linear 3D problems (i.e. geometric plus material non-linearity along with contact) there are very few programs that can compete with ABAQUS.

 

[SkyD]

Personnally I think MSC are starting to get what they deserve!!! They have been very arrogant over the years thinking that just having the recognition of some industries, they could do nothing to improve their products and keep the same market share and the same prices. I've got a friend who has just bought a license for Nastran/Patran... They are charging him £75k a year!!! For a programme that has to my sense very limited capabilities... Especially on the pre-processing side... Ever tried to mesh anything complex without using Tet mesh? Unless it's got square edges everywhere you can't... I think I would agree with the comments that Femap is better and certainly Abaqus... But I think we need to keep an eye on all the other smaller products that are developing here and there...

I've started doing some work with Altair's Hyperworks including their Optistruct product. It's a very nice product with lots of capabilities for a fraction of the prices proposed by MSC. Certainly the interface of Hypermesh has a long way to go before it becomes userfriendly but if you haven't seen it before, you should check it out. It's very flexible when it comes to pre-processing.

We've started working with them to support the development of some new functionalities and I must admit I'm impressed by the effort they are making to meet the targets we give them...

I think on the long run, MSC will pay dearly the price of their arrogance and it can only be a good thing. It means cheaper and better softwares for us.

I've certainly noticed recently in the Aerospace industry, a change of mentalities. People are starting to agree that MSC is not the God Almighty of FEM and that other product could suite their needs better.

What do you guys think?

SkyD.

 

[dmacx]

We used to use MSC.Nastran in my group for many years with Patran. I remember a different MSC 15 years back than what it has become now. It is a shame, but the good news is that other companies are listening. MSC thinks they can solve their problems by just raising their prices justified by giving people features they do not need.

I agree with you SkyD about keeping your eye on other products. We have been using NE Nastran for several years now with very good success. They are taking a different direction for Nastran which seems to be to listen to what the customer wants, not just the big companies that still use MSC. I think MSC development is driven by their big accounts only. I heard they are outsourcing a lot of development now to India. That should be interesting. Maybe they will relocate all their staff over their as a way to save money. I would not be surprised if their support department is moved (replaced) over their soon.

 

[Jordonlaw]

"I would not be surprised if their support department is moved (replaced) over their soon."

I used to be able to phone a guy up in the UK to talk through problems, most of which were solved over the phone. Now I have to do it via e-mail and some MSC guys admit at least some of the support people are in India. They are also aware that this doesn't work very well and that customers are not happy; there is a hint that some of the MSC organisation hopes this will fail presumably for the benefit of their part of the empire.

I still think that the product I use (MSC MARC) is better than anything else on the market for what I need to simulate having benchmarked alternatives when I started this work. However, the cost is high and upgrades, particularly to multi-processing, are ridiculous. Their new licenese to token 'upgrade' path is little better if like me you use a single product.

I hope they sort things out.

 

[mikel526]

I agree that MSC is on the ropes. Ansys workbench with ANSYS clasic is far ahead of them on all fronts. I don't agree that PATRAN is better for composites at all. Its all MSC's fault, they have been a cash cow company relying on the fact that engineers are horrendously conservative. Anyone who is interested in performance and productivity and bothers to learn other softwares sees the enormous gap between MSC and their competitors. For the sake of engineering as a profession, I hope MSC either dies or gets its act together.

http://www.arclen.com

 

[Jordonlaw]

mikel526,

"Ansys workbench with ANSYS clasic is far ahead of them (MSC) on all fronts."

I'm going to defend MSC here. You can't possibly say that ANSYS is ahead of MSC on all front; its nonsence. I've benchmarked ANSYS against MSC products and ANSYS couldn't produce a technical result for the job I had to solve - pure and simple. Even their support people said it couldn't (and still can't) solve the problem I have. With your problems your statement may be true but then you should qualify it accordingly.