Semiwiki on LinkedIn Semiwiki on Facebook Semiwiki on Twitter Register / Log in 
Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/synopsys-acquires-magma.1152/page-2
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021770
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Synopsys Acquires Magma!?!?!?

krylov_subspace said:
Innovation? In EDA?
Click to expand...

There are many innovative people in EDA, but my experience in the last couple of decades is that the big guys are not really interested in changing how they do things. E.g. we should have moved up from RTL to higher levels of abstraction for synthesis, but Synopsys makes it's money there so you can't get anything through the language committees, likewise Cadence blocks progress in analog. The tech-bust saw the demise of CAD methodology groups outside the EDA companies, so most users just buy into whatever their vendors tell them to do - the design tools suck because they are old and clunky, but never mind we can sell you all these neat verification tools that catch bugs on the back-end.

Personally I'm looking to be innovative in other domains where the big guys can't get in the way - http://parallel.cc
 
simguru said:
There are many innovative people in EDA, but my experience in the last couple of decades is that the big guys are not really interested in changing how they do things. E.g. we should have moved up from RTL to higher levels of abstraction for synthesis, but Synopsys makes it's money there so you can't get anything through the language committees, likewise Cadence blocks progress in analog. The tech-bust saw the demise of CAD methodology groups outside the EDA companies, so most users just buy into whatever their vendors tell them to do - the design tools suck because they are old and clunky, but never mind we can sell you all these neat verification tools that catch bugs on the back-end.

Personally I'm looking to be innovative in other domains where the big guys can't get in the way - http://parallel.cc
Click to expand...

Yes, there are many innovators in EDA but ,alas, there is not much innovation.

That in and of itself is not catastrophic, it just means that we should treat Synopsys buying Magma as we would treat consolidation in the toy industry (which is , BTW, around 75 Billion $ annually, not a measly 3-5 Billion like EDA is).

It's done to reduce expenses,and increase revenue and shareholder value.

So what will happen? Easy:
  1. Synopsys will kill the magma products that were eating into their margins
  2. Synopsys will lay off a bunch of people.
  3. The reduced competition will enable them eventually to increase the price of their tools.
 
We had heard Rajeev tell media, there will be only 2 EDA players... and as far as Magma he had hinted that ".. they would be bought out by one of them".. We knew Cadence was going after Mentor..that leaves one other..
Posted by HariPrasad
 
Wei Shi <q> two great timer (goldtime and tekton) have been pushed down under the water by SNPS, and then how they keep up two PnR tool, is doubtful? so eventually, what the user community ended up one timer, one pnr tool, with no options, risk mitigation left on the table. not a good sign. </q>
 
Sandeep Maheshwary <q> @Gopi, I agree, front end architectural system design capabilities is the space for a market leader. It is also a cheap acquisition for them. SNPS has strong cash flow that will accelerate within 1-2yrs of this acquisition: R&D cost savings, Consolidate development talent & IP, Significant savings on SGA, and Higher License fees on renewals. Good for engineering, great for business. </q>
 
However, this is good for Cadence as well, because while Synopsys pays for finishing off Magma, Cadence also gains from the lack of competition.
Posted by Kamran
 
I asked pretty much everybody I met with in Taiwan last week what they thought about the acquisition, which was one of the hot topics of the trip. The most common theme of EDA discussions is why our industry is a mere crumb of the total semiconductor pie. During a lunch conversation last week I proposed that ALL of the EDA software licenses be deactivated for one month so the semiconductor industry better appreciates EDA. With the recent acquisitions and lack of capitol investment in new EDA companies, that scenario is now much more plausible! ;-)
 
Daniel Nenni said:
I proposed that ALL of the EDA software licenses be deactivated for one month so the semiconductor industry better appreciates EDA.
Click to expand...

Why stop there? Let's also turn off all the hard drives so the semi industry appreciates Western Digital, and deactivate all the toilets so that we appreciate American Standard.
 
Daniel Nenni said:
Prices will go up. We will make more money. All is good! Right?
Click to expand...

I think that would be fine if value had gone up, but in this case the users will just be paying more so it will probably drive some more customers out of business - so it'll be back to the status quo.

Disabling the licenses would be good for open-source so I'm all for that!

The only way to expand EDA (IMO) is to improve the results so that the whole process of designing chips is cheaper.
 
Magma was offering a chance for startup as Magma tools are competive in price. With Magma possibly gone,and probably killing tools like talus and tekton, it would be detrimental to the chip design industry.
Posted by Alan
 
I hope the government blocks this deal. It is very anti-competitive. There is no doubt Synopsys will eventually kill the Magma tools except for maybe their timing tools which are much faster than PT.
 
Yes ... they did. Aart De Geus is brilliant. While everyone else is focused on the "product overlaps", what I see is sheer genius. Think about it from the other *angle*. Who cares if there are overlaps? Synopsys acquired Avant! and has done a superior job of entering the Place&Route market in a level playing field. Idon't think Synopsys will have a difficult time bolting the goodparts of Magma tools into their own.

No ... what Aart has done is really amazing. He's kept Cadence and Mentor from buying Magma. If Cadence buys them, they get areally, really good Physical Verification tool in Quartz. Cadence'sPVS tool is in a far 4th place in this market. They've been trying for almost 10 years to get off the ground with a good PV tool. From what I've read and heard, the PV tool ranking with regard to speed, accuracy, reliability and support is - tied for first, Quartz(LAVA)/IC Validator(SNPS), then Calibre(MENT) and then PVS(CDNS). Sure, Calibre has the market share because they've been around the longest, but Synopsys is moving up fast with their new tool. Quartz would have filled a huge hole for Cadence. If Mentor buys Magma, then Mentor can toss 75% of the Calibre guts, insert Quartz and call it "CalibreII".

They would also get the Magma P&R tool, Talus Vortex, and a lot of front end tools. They would probably move up into the #2 spot. Not now ... with this buy Synopsys really has the best of everything, or very, very close. They're taking market share with their Virtuoso-like tool (Custom Designer), huge in the IP market, will augment their Place&Route tool offering and probably bolt some under-the-hood goodies into IC Validator as well. I'm buyin' more SNPS ...

I see them at $2B revenue in the next 2-3 years. My hat's off to Aart and Co.

S20Y
 
Last edited:
Improving the results and saving design cost might have a long term benefit to grow the customer base but that will be a very very long and painful time to nurture new customers as the semi industry itself isn't in the hot spot of VC money. Prospect of new big customers emerging isn't that good. Improving results significantly itself is a very hard problem to solve and needs heavy R&D investment for the cash strapped EDA companies that are already losing top talents to software companies. Investors are not interested in EDA startups any more and the big ones are not doing all that well. They can't afford any risks.

For the small and consolidating customer base that EDA has, (the top 20 or so semi houses own the market, and they pretty much have a fixed EDA budget and try to cut each year), the only way to grow the industry that will see results in the next 5 years is to consolidate and increase the price.

simguru said:
I think that would be fine if value had gone up, but in this case the users will just be paying more so it will probably drive some more customers out of business - so it'll be back to the status quo.

Disabling the licenses would be good for open-source so I'm all for that!

The only way to expand EDA (IMO) is to improve the results so that the whole process of designing chips is cheaper.
Click to expand...
 
Last edited:
That article couldn't be more wrong about Mentor. Mentor will be the biggest winner from this merger. Caliber will now eat more market share and Cadence will have no option but to acquire it. Cadence will be the biggest loser from SNPS's acquisition of LAVA. Not Mentor!
 
"That article couldn't be more wrong about Mentor. Mentor will be the biggest winner from this merger. Caliber will now eat more market share and Cadence will have no option but to acquire it. Cadence will be the biggest loser from SNPS's acquisition of LAVA. Not Mentor!"

Not true. The Magma Quartz DRC tool that competes with Calibre has no market share. The only big name company that uses it is Nvidia and they also use Calibre for sign-off, Quartz is not a sign-off tool at any of the foundries at any node.

The Nvidia business will go to Synopsys (my opinion) not Cadence or Mentor.
 
You must log in or register to reply here.
Back
Top