Look at the new cfd code APU:

http://www.flowgrid.com/

They even do flow around the woman

Have you heard of this code?

Thank you come again,

Apu

Its interesting that these new codes keep cropping up (CFD2000 is another example) does anyone think that the marketplace is now at a point that any of these are likely to survive? The big three is about to become the big two with ansys buying fluent and other codes such as CFDesign, flowmerics etc already have a foothold in the industry. Personally I think that 10 years ago, yes they might have a chance, but now CFD is becoming a serious industrial tool is there any likelyhood of any of these newer small guys getting anywhere?

Please not I'm not knocking the new codes, im sure they are fine I just cant see them getting that far in the comercial world now.

I think that new CFD codes such as this can gain marketplace and continue to alive with focusing on especial application, because they don't compete with well-konen general purpose CFD codes. This also permit them to give better feedback and support for consumers.

Also the populare codes such as FLUENT can't model special phenomena also their performance is very low for some application.

As an example the MAGMASOFT code is a commersial casting simulation tool (fluid flow and heat transfer) its price is very high ($100000-500000), it taken good market more than 500 user 4 years ago from my nowledge (note that is computational kernel is very simple, it use uniform cartesian grid and FDM method).

I model casting with magma and fluent (unsteady free surface flow) the fluent is about 20-times slower and also results of magma is more reliable.

Generally it can be said that developping comercial code for especial application not only needs expert CFD-team (probably big codes has this) but also needs expert people in that especial field. Satisfying the second criteria is not possible for general purpose CFD codes.

There's a LOT of truth in that statement!

It looks like CD-Adapco is going to end someday in five years time. Then we can see CFD monopoly globally.

>>As an example the MAGMASOFT code is a commersial casting simulation tool (fluid flow and heat transfer) its price is very high ($100000-500000), it taken good market more than 500 user 4 years ago from my nowledge (note that is computational kernel is very simple, it use uniform cartesian grid and FDM method)

This is very interesting information, and corroborates with past information that special purpose codes sell for more than $100k, and what's interesting/funny about this is that the engine is very basic FDM with either rectangular grids or some simplified grid-generation tools (templates that are modified within a narrow range of parameters). This is a very smart business model.

Now curious in your statement above: do you mean 500 end users or 500 licensees. That is, do you mean 500 separate entities paid >$100k each? This is very valuable/important info, if you could share with us

Thanks

Adrin Gharakhani

My purpose from 500 is 500 licence or 500 buyer. Also it is important to note that additional feedback and supports is usually as important and lucrative as primary marketing software is (note that my opinion is based on my previous experience related to working in developer team of SUTCAST software, www.sutcast.com).

what makes you say that? The company has been growing 15-20% the past few years and if anything is the best placed company to challenge the ansys monopoly as current cfx/fluent users get nervous about changes to their codes. And that is of course not considering if a company like dss buy them out to try and compete.

what I ment in my first post wasn't really aimed at specialist codes (such as those for casting) more at codes like CFD 2000 and so on which seem to be aimed at general industrial CFD which is already a market with a lot of well established widely used products. I doubt they can survive or compete with fluent, STAR-CD, CFX and so forth.

CD-adapco may or may not be bought by someone else but if they are, the buyer certainly will not be Ansys. The US govt was not sure they even wanted CFX and Fluent owned by 1 owner so for sure they would not allow Ansys to buy STAR and create a true monopoly.

I am completely agree with you.

I think the future of CFD lies in GPU/CG language and what NVIDIA is delopping.

(http://sourceforge.net/projects/libsh/)

within this development frame microsoft campaign to remove Linux and Unix (Get the fact campaign)

what is most likely to disappear is UNIX/Linux and the software who do not work with microsoft:

http://www.microsoft.com/windowsserv...s/default.mspx http://www.microsoft.com/windowsserv...grateunix.mspx

Roger

I cant see it myself, microsoft are a long way away from the type of performance advantage that unix gives and programming ease, its cluster based system is still at beta level where as unix clustering is now firmly established.

"Blatant, inappropriate advertising", that's a long phase isn't it?

blatant advertising is not on www.cfdreview.com from the main cfd-software provider anymore?

Rob

I understand the point you're trying to make, but the one example you cite, CFD 2000, doesn't fit. CFD 2000 has been around for quite some time already so I don't see it going away anytime soon just because ANSYS acquired CFX and Fluent. In fact, what you'll see are some of the existing CFD companies start to grow and fill the void left by CFX and Fluent.

What I infer from your statement is that ANSYS+CFX+Fluent is really big (i.e. lots of money) and therefore they'll provide a better product than anyone else. I don't believe that'll be the case. All CFD codes have unique features, from very technical things to price to level of service to type of customer they intend to serve. There are many opportunities for CFD solver companies and continued growth should be expected. Some recent numbers show CFD leading the growth of the CAE market, growing 18% a year through 2010. Not all of that will be from ANSYS/Fluent.

Thanks for information. I still prefer using Linux/UNIX.