DAC 2011 – The Feel Good DACDAC is back, EE Times is back, the EDA forecast is up and we have made the transition to ESL design. It was a great DAC. It was almost as if the last decade hadn’t happened.
Yes we survived!
My Sunday Night Talk
I don’t usually review my own talks but it looks like I hit a nerve this DAC. For those that didn’t attend DAC I followed up on a talk Lucio Lanza made at IC CAD last year on the Responsibilities of EDA. I showed the ITRS Cost Chart with two lines added. The first line was at $25 million, below that point VCs will start funding Semiconductor Start-Ups. The second line was at $50 million above that line even IDMs start cutting back on new SoC designs (See Figure 1).
Figure 1: ITRS 2010 Cost Chart
The bottom line is:
• EDA is responsible for developing the design tools that enable the IC design process, at a design cost that allows the ecosystem to operate at a profit.
Keynotes and Other Talks
Following my Monday morning pavilion talk on What To See at DAC list (video of band), I went to see Jim Hogan and Paul McLellan give their investor talk. Great talk; it was both fun and informative at the same time. It is talks like these that make DAC such a bargain. You generally pay a lot of money to get the quality of advice that these guys dish out.
Unfortunately I came down with a bad cold and ended up sleeping in Tuesday morning so I missed the opening Keynote. From what I hear Lisa Su, from Freescale continued the theme of asking the EDA vendors to keep up with Moore’s Law.
Gadi Singer of Intel is close to the top of my list of people to listen to. It was another great talk. Gadi continues to challenge the EDA Industry to think bigger, work faster and adopt your customer’s problem as your own. Again he was driving the theme of taking responsibility for the cost of design.
IBM’s Dharmendra Modha’s talk was the kind of talk we need to have more of. These talks that look into the future of semiconductors and computing are indispensible for keeping our eyes on the future. If you don’t know where you are going there is a good chance you’re not going to get there. His talk was on the study of the brain and how this may change the way we look at computer architectures. It was a very inspiring talk.
Word on the Street
One of the things I liked most about DAC 2011 is that I got a chance to talk to a lot of design engineers. I do get the opportunity to talk to many of the design and CAD managers during the year but generally don’t talk to many of the “in the trenches” design engineers. Their input on the major EDA vendors was interesting, to say the least.
Mentor is trucking along as always. Not only are they staying number two in revenue, they are catching up with number one slowly but surely. They seem to have the top R&D team now and in general are considered the best major EDA vendor to work with. The only concern was Carl Icahn. Word from the trenches to Carl is don’t mess up a good thing. Keep Wally in place and keep the momentum moving. The engineers are relying on Mentor to solve many of the 22nm problems.
Magma is back! Looks like their ASIC Layout tools are doing well and they’ve added Analog Layout, SPICE simulation and a killer Timing Analyzer to their tool offerings. One piece of advice I got was that Rajeev needs a vacation. He’s been working way too hard the last three years.
Apache was another big topic. Their domination of the Power market and their recent “possible” IPO has really impressed the engineering community. The possible comment was that many engineers wonder why Andrew wants to go public. Especially with the recent mess at Mentor people are starting to think that going public is way more trouble than it’s worth. Especially for successful EDA vendors that have a good cash flow. Wall Street might need to take a look at the health of their market. These engineers are the ones that will man the next start-ups. With the recent group of start-ups avoiding the Venter Capitalist and now questions on the need for an IPO, there seems to be an anti- Wall Street trend that could spread beyond the High Tech world.
The Good News and the Bad News
The good news is that Cadence was the talk of the show. Their tools have improved significantly and they are taking market share away from Synopsys. This was the show me year for Cadence and they are pulling it off. Not only are engineers starting to like their tools, they really like working with Cadence as a vendor. Customer support was mentioned multiple times. Unfortunately this was usually in context with Synopsys poor customer support, which is our next topic.
The bad news was the overwhelmingly poor view of Synopsys as a vendor. Somehow Synopsys has turned from being the most respected vendor to the most disliked and in some cases feared. Most engineers no longer consider them the technology leader and you throw in their recent arrogant approach and it doesn’t look pretty. This is not the Synopsys of only a few years ago. Aart’s a great guy and I hope he can turn the company back around.
All in all it was a great DAC. The EDA industry is looking good. We should have a nice run for the next five years. Now if we can only do a better job of keeping up with Moore’s Law we can keep the engineering community happy. I’m looking forward to DAC 2012 in San Francisco.