The PCB design software market has shown fairly lackluster growth over the past two years. However, there is reason for hope. One area that Dataquest has identified as an area of possible high growth is that of the PCB physical analysis tools. This article examines the market and its potential for growth in the short term.
PCB design software markets have traditionally suffered from a lack of attention, being overshadowed most of the time by their more glamorous counterparts in the computer-aided engineering (CAE) market. The market has not been helped by the fact that performance over the last two years has been fairly lackluster except for a few bright spots. However, it seems that the hibernation is over and the market is ready to enter a new phase. (Of course, in the meantime there has been plenty of activity in the market both in terms of mergers and acquisitions and in technological advances.)
Mergers and acquistions
Both 1999 and 2000 saw significant merger activity in the PCB market. Table 1 lists several of the significant mergers that occurred.
It is interesting to note the nature of the activity that occurred. Note that both PADS and VeriBest-the two mid-range PCB software vendors-have been acquired, which leaves only Protel as an independent vendor of reasonable size. The market is top-heavy with the top three companies accounting for more than 60% of the revenue in this market. Interestingly, the market is increasingly being divided up in North America and Europe by Cadence and Mentor, while Zuken continues to dominate Japan.
While the top three vendors-Mentor Graphics, Cadence and Zuken-have traditionally catered to the power user community, all the acquisitions they have made recently have been aimed at increasing the penetration of the user base and acquiring technology to either enter new markets or round out product offerings.
For instance, Cadence's acquisition of OrCAD immediately gave it a strong position in the "shrink-wrapped" and lower mainstream markets. Similarly, Mentor Graphics' acquisition of VeriBest allowed it to enter the mainstream market. It also gave Mentor access to new routing technology, a critical issue since Mentor did not renew the CCT router license. In this instance, Mentor Graphics was able to kill two birds with one stone, acquiring the necessary technology and penetrating a new user base.
Protel's acquisition of Accel has allowed it to establish itself more firmly in North America, something that was really necessary for the company since Protel's corporate headquarters are in Australia. Protel has since gone on to acquire an embedded real-time operating systems and tools company, Tasking Inc., in a reprise of Mentor Graphics' acquisition of Microtec Research a few years ago. At first blush the acquisition does not seem synergistic but if you evaluate it a bit more closely, it becomes apparent that there are some excellent opportunities for leveraging new technologies that it has acquired-especially in the "late adopter" market. Here, the PCB/FPGA designer (in many instances) is also responsible for firmware and device driver development. Since this is the natural target audience for a vendor that sells into the shrink-wrapped and lower mainstream markets, the revenue possibilities are strong.
In terms of market share, the vendor lineup has remained fairly static over the past three years. Most of the new entrants in the market were primarily start-ups that were trying to leverage new Internet-based technologies to design boards. Some of these start-ups focused on design collaboration software that could be used over the Net, some on offering new licensing models, and some primarily on component selection and purchase. Examples of such companies are SpinCircuit.com and EDAconnect .com. Marketshare for the established vendors grew mostly through acquisition.
As mentioned earlier, the sales of PCB design software were flat for the past two years overall, though certain submarkets did show the promise of good growth. Figure 1 below shows the growth trend in the PCB market as compared with the overall market for electronic design automation software.
Mentor Graphics, Zuken and Cadence have been rotating among the top three positions in the market for the past three years. Mentor Graphics gained the leading position in 1999 and Gartner Dataquest's estimates suggest that the lineup will be unchanged for 2000.
Gartner Dataquest believes that physical analysis tools show good potential for growth. Even though the demand for layout software will continue to grow, the growth rate will be much faster for the "five sisters" of physical analysis: signal integrity, timing, EMI/EMC, thermal and power issues.
The reason for this is that mainstream designers are starting to run into analysis problems. In the past, high-speed board design was perceived to be mostly in the domain of the leading-edge user community. However, a look at some of the user surveys that Gartner Dataquest has conducted over the past few years shows a steady creep in the number of users who report designing high-speed boards. Over 35% of designers reported in 1999 that they were designing boards that had clock speed of 50 MHz and higher, and we expect this percentage to rise in the future. With leading-edge processors slated to run at gigahertz speeds, it is not inconceivable that mainstream designers will be working with clock frequencies of 300 MHz and higher.
As a result, the biggest problems that board designers face today are primarily analysis-related problems. Of these timing and signal integrity problems have been around for a while in the PCB world. As the two domains are tightly connected, many of the leading PCB software vendors offer some kind of analysis solution that is closely integrated with the design and layout software. The other analysis areas-power, thermal and EMC/EMI-are now rising in significance. Solving power delivery problems that deal with being able to deliver the right power to different parts of the board as required by the components at the right times is becoming critical.
The problem on the supply side is that most of the solutions available in these areas are point tools that tend to be standalone and are not integrated with the design and layout tools. In some instances there are no tools. As a result, many users either tend to have their own tools-which they use to fill in the gaps that the commercial analysis tools leave-or try manual tweaking and massaging and hope that the problems will somehow solve themselves.
In the thermal domain, the problems become even more complex because of the need to take into consideration mechanical constraints, including external constraints such as ambient temperature within a device, the use of heat-spreading mechanisms and lower air flows within devices because of changes in the overall system configuration. Again, in this space a number of point tools exist that are not integrated into a board design software package.
The net result of all this is that users find themselves in the unenviable position of having to deal with higher board speeds, exploding numbers of critical nets, microvias and buried components on the board, as well as having to comply with federal regulations that govern EM emissions and thermal issues while still trying to meet their project deadlines.
Many of these problems apply equally to package design. In many cases, a package tends to be an instance of a highly advanced printed circuit board. From a system perspective, the package acts as a connector between the chip and the board. Given all these trends, Gartner Dataquest believes that analysis tools are an area of good growth for the market. Figure 2 has a forecast for the growth of the physical analysis tools market.
During the course of this article, there have been several references to the power user, mainstream and shrink-wrapped communities. Table 2 is a model of the user base for the PCB design software market. The smallest third of the pyramid is the power user community-who are traditionally users of high-end tools. They are the smallest in terms of numbers of users but they spend the most on tools. The mainstream users are almost 42% of the user community. The upper mainstream users tend to use a mixture of high-end and "ready to use" tools, and by operating system, tend to use a mix of UNIX and Windows NT systems. The late adopter market is a large community, but it spends the least amount of money. Late adopters typically tend to buy software packages that can be characterized as "shrink-wrapped" or requiring very little support. They also do not tend to make use of analysis tools. However, some of this is changing. As mainstream designers start to encounter analysis problems, they are likely to be looking for an integrated layout and analysis package with the support model that they are accustomed to. With some trickle-down effects in place, shrink-wrapped vendors also will need to include analysis tools in their product offerings, something that some vendors are considering very strongly.
The PCB design software market is now at an inflection point. The market is poised for growth but a significant portion of that growth will come out of the analysis tools. The big void today is the lack of a universal database that can store not only layout information but also take in information from all the different point tools that are available for the different domains, including the die information and packaging information. Today's engineers tend to use Excel spreadsheets to conduct their analysis.
Some of the vendors are trying to address this problem, but given the number of standalone tools that exist today it will take a while for this to become a standard package. The bottom line is that revenue opportunities exist for vendors who recognize this need for a more integrated layout and analysis package.
Daya Nadamuni is a senior analyst for the technical software programs of Gartner's Dataquest Software group. She contributes to the research effort for mechanical CAD/CAM/CAE, AEC, electronic design automation, embedded software and Real Time Operating Systems worldwide programs. She can be reached at firstname.lastname@example.org
Mergers and acquisitions
in 1999 and 2000.
Source: Dataquest (May 2001)
The PCB user pyramid
50% Late Adopters
Source: Dataquest (May 2000)
© 2001 CMP Media LLC.
7/1/01, Issue # 1807, page 30.
Reproduction and distribution of material appearing in PCD and on the www.pcdmag.com Web site is forbidden without written permission from the editor.
Contact information for Reprints.
NOTE: The articles presented here contain only the text originally published in Printed Circuit Design magazine. Any accompanying graphics and illustrations have not been recreated here. You may view the article in its entirety in each printed issue of PCD.