July 2008 Archives

Penny Herscher, who was a Cadence Design Systems executive for a while after the EDA giant bought her company Simplex Solutions, reckons her old employer should have done the same as Synopsys some years back and just moved to a 100 per subscription licence model and have done with it.

But where's the fun in that? We couldn't have had the ten years of licence-mix changes that Cadence has treated us to.

See if you can guess when each of these changes took place. And check out the talk of "visibility" and "transparency":

"Additionally, this product sales model lets us build revenue predictability and visibility, while receiving appropriate value for our products. Based on this new approach, we're confident that we'll achieve the revenue performance and profitability that has been characteristic of Cadence in the past."

The Company's product sales model will include a new subscription license (usually two years), which makes available both current and new products. Because it includes undelivered technology, the subscription license requires revenue to be recognized ratably over the license period.

The Company estimates that over time 30 per cent of product bookings will follow this subscription model.

As a result, total revenues in the third quarter are expected to decrease from the current quarter.

Or this one:

Cadence expects its subscription bookings, which going forward include subscription licenses and maintenance combined, to grow from 48 percent of the mix of software bookings in Q2 to 65-70 per cent for Q3. For Q4, the company anticipates approximately 70-80 per cent of software bookings to be under subscriptions.

The company expects that the change in mix of software licenses should enhance revenue visibility and, beginning in the second half [of the following year], accelerate revenue and earnings growth.

The joy of flexibility:

"In the quarter, approximately 49% of our product business was represented by ratable licenses.

This percentage is lower than our typical rate in the mid 70s, because we elected to continue to work several contracts for better value. Our forecast for the revenue mix in Q4 is in the low 80s and in the 70s for the year. For the year, we expect to generate approximately 2/3 of revenue from backlog."

And why 100 per cent ratable is bad:

"I think as you compare over the long term it’s the amount of business that you’re able to win and I fully admit that our model is a much more transparent model than one that is 100 per cent ratable. [The mix was about 60 per cent upfront/40 per ratable at this point]."

...because he fixed it in the mix.

Don't believe me? From the Seeking Alpha transcript (I haven't had a chance to listen through the audio yet) of Cadence Design Systems' Q2 conference call:

"...we made the difficult but necessary decision to lower our outlook, and transition to an approximately 90% ratable license mix. We believe this transition will enable to us keep our focus on the value of our technology, and this decision is the right one for our business over the long-term and for building sustaining strong customer relationships in the future."

Which, as I recall, was the reason last time for adjusting the balance between how much revenue Cadence recognises upfront for licences in its quarterlies versus the amount spread out over the term of the licence. I need to dig out the old 10Qs and 10Ks and transcripts to find the previous changes in but, as I indicated in my first blog post about the attempted takeover of Mentor Graphics, this kind of financial shuffling is a familiar one to anybody who watches the electronic design automation. I didn't think they'd have the balls to try it again. I await the day when one of the big EDA companies goes for a 110% rateble mix and starts invoking complex numbers in its non-GAAP reporting.

This means that the discrepancies between Cadence's and Mentor's financials will be that much larger if the merger goes through, which makes any comparisons between the businesses now and in the future even tougher for anyone trying to analyse the figures.

And with that, over to Indeep.

Researchers at MIT have developed a way to draw lines on wafers that are just 25nm across and spaced 25nm apart much more cheaply than today's fancy immersion lithographic steppers. The bad news is that the $2m tool developed by Mark Schattenburg, Ralf Heilmann and two graduate students at MIT can only do parallel lines right now. But the technique could provide an alternative way of doing double patterning that could potentially push optical lithography well into the next decade.

The technique, which is based on the 'nanoruler' developed a few years back, uses interference patterns generated by pumping 100MHz sound waves into a crystal through which they shine a laser beam. Heilmann said: "We use a transducer to change the properties of the sound wave at a very fast rate. We have updates rates of 10kHz or so. Changing the sound wave changes slightly the phase of the light wave that goes through the crystal."

By recombining the light split from a laser under the control of vibrating crystals, it is possible to create very fine interference patterns. These produce extremely thin lines on the wafer. The researchers worked with a laser operating at a wavelength more than 50 per cent longer than the 193nm deep-UV light used by today's steppers. But the interference pattern leads to sub-30nm lines. But they are quite widely spaced: about 200nm apart.

To get the line density up, the team used four passes. This takes time but Heilmann said the process can be applied across a wafer rather than being limited to a 600mm2 reticle.

In its current form, the scanning-beam interference lithography can only make parallel lines. “We are working on changing the optical design to do a greater variety of patterns,” said Heilmann.

Although the technique might be used to create just regular structures, one possibility is that interference lithography might be combined with conventional lithography. For starters, companies such as Intel have already started to use more regular structures. In this approach, long lines could be patterned across the surface of the wafer and then cut into gates, for example, using an existing stepper. Very often, silicon transistor gates are way, way wider than they are long. And this pattern-and-cut trick is an approach that Intel already uses with conventional steppers on the 45nm Penryn devices.

One issue for the future is that of line-edge roughness. This roughness is becoming more and more of a headache because it increases the variability of transistors. Just cutting that variability in SRAM transistors, which tend to be more densely packed than others on a typical chip design, could pay off in better yield.

"Line-edge roughness is not a limiting factor but it is close to being one," said Heilmann.

Schattenburg said in the release: "There are several new technologies on the horizon that have the potential for alleviating these problems. These results demonstrate that there's still a lot of room left for scale shrinkage in optical lithography."