Home > Community > Blogs > Industry Insights > analog is back but not like you think
 
Login with a Cadence account.
Not a member yet?
Create a permanent login account to make interactions with Cadence more convenient.

Register | Membership benefits
Get email delivery of the Industry Insights blog (individual posts).
 

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

Analog Is Back – But Not Like You Think

Comments(0)Filed under: Industry Insights, Virtuoso, Analog, Mixed-Signal, advanced node

After taking a back seat to digital design for many years, analog is re-emerging as a key differentiator and competitive edge for IC vendors. But analog design today is very different from that of 20 years ago, and realizing that is critical for success.

In a recent EE Times column, Planet Analog editor Bill Schweber noted that IC vendors with analog portfolios are in a better position during the downturn than those with all-digital portfolios. There are a number of reasons. Analog is a “foundational” technology that is increasingly indispensible for electronic systems, and analog design is getting harder as digital systems speed up. Analog expertise, meanwhile, is scarce. “IC vendors know that analog markets are more defensible at the customer design level,” Schweber wrote.

So analog is back. But as I learned in a recent discussion with Steven Lewis, product marketing director for the Cadence Virtuoso product, that’s not the whole story. “People are going back to their analog roots,” Steven said. “But the difference between analog today and 18 years ago is that today there’s not as much time, you have to fit into a mixed-signal chip, and you have to reuse what you did before.”

In today’s economic climate, time-to-market is critical. Most analog circuits need to fit into systems that also include digital circuits, and the analog circuitry can’t hold up the entire project. Companies can’t afford to wait weeks while an analog engineer handcrafts a perfect RF circuit. And it’s just not cost-effective to design a circuit once, throw it away, and start from scratch on the next one, even if some of the original design or design environment could have been reused.

Analog designers have been very resistant to automation – and for good reason. As an EE Times editor, I wrote about various startups promising analog “synthesis” or some other form of automation. But analog “synthesis,” if that’s the correct use of the term, really only worked for a few simple circuits such as filters. Good analog design is still an art, and it still requires handcrafting. As Steven said, “in the end, nothing can beat an artistic handcrafted layout.”

So what to do? One possibility is rapid analog prototyping, a capability available with Virtuoso’s design constraint system. With this, you can do a quick placement using Skill p-cells and run extraction and design-rule checking. It’s not a finished layout, and you still have to do handcrafting, but the prototyping allows a first-order approximation that can help avoid big mistakes.

You can also, as Steven puts it, “tell the machine” about measurements and specifications rather than keeping them in your head or writing cryptic notes on schematics. With this information stored in a Virtuoso library, you don’t have to keep rebuilding test benches, and you can easily switch between behavioral, schematic, and layout views of a circuit. If you design one op amp and store the measurements, setup will be much easier for the next one.

Analog autorouting doesn’t work for full circuits, but there are still automated techniques that can help, such as channel routing. Some nets will always have to be handcrafted, but hopefully these can be reduced to a relatively small number.

Analog IP reuse is difficult because you can’t resynthesize the source description. Whoever designed the IP may no longer be with the company. For these reasons, there isn’t much analog IP reuse. Reuse is possible, however, if you document everything possible about the IP, including how it was built, verified, and tested. The documentation can include testbenches, measurements, routing constraints, and more. Again, “tell the machine” rather than scribbling notes on a schematic or writing a Word document that might get lost.

All of the above requires a different way of thinking about analog design. As Steven noted, some customers still insist on a handcraft-everything, build-it-all-from-scratch approach, and the tools do allow that. But economics are increasingly pushing design teams to find a new way. Somewhere between the unrealistic promises of analog “synthesis” and the pure handcrafted approach lies a methodology that preserves the “art” of analog design, while giving it a little more science.

Richard Goering

Comments(0)

Leave a Comment


Name
E-mail (will not be published)
Comment
 I have read and agree to the Terms of use and Community Guidelines.
Community Guidelines
The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.