After we correct all the Y2K computer glitches on Jan. 1, what should we expect from instrumentation manufacturers in the next 12 months? Will improvements continue at the current pace in 2000, or is it time for a slowdown?
We wondered about this, so we went to five individuals who have proven their ability to lead in planning, developing, and delivering new and innovative equipment:
- Jim Truchard is president of National Instruments, a supplier of versatile software and computer-based measurement and automation products.
- Rick Wills is CEO of Tektronix, a test and measurement company.
- Tom DeSantis is president of IOtech, a manufacturer of data acquisition hardware and software.
- Tom Saponas is the chief technical officer of Agilent Technologies, formerly known as Hewlett-Packard, a supplier of test and measurement instruments and systems.
- Joe Keithley is president of Keithley Instruments, a company that specializes in test and measurement instruments and systems.
Our conversations revealed that 2000 will be a banner year for instrumentation manufacturers and users. We can expect an exciting period of product development and enhancement. Over and over, we heard these leaders use expressions like faster, more powerful, lower-cost, easier to use, smaller, and solutions-oriented as they told us where their companies are headed.
Dr. Truchard (right) has coined an interesting phrase to describe the environment that will exist in the instrumentation community in 2000. He called it the return of Edison. “When Thomas A. Edison was in the laboratory inventing all those products, he virtually did all of the work with just a few assistants,” he said.
“Life got more complicated after Edison’s day, and eventually we needed a small army of engineers for the typical product design. Each person specialized in one small aspect of the job, took all the measurements manually, and did all the calculations by hand. For all these reasons, it could take a decade to finish a job,” Dr. Truchard continued.
“Now, the power of a PC, the versatility of a laboratory system, and a handy set of software tools will restore small group creativity. Maybe we won’t get as much from a modern-day designer as we did from Edison, but it’s a big improvement over the small-army approach,” he concluded.
Users Initiating Instrument Designs
While it may not be obvious, manufacturers get most of their new product ideas from prospective users. Mr. Wills explained: “Our good product ideas come from our customers. They may not always know what they want, but they know the problems they’re trying to solve. If we can connect our creative people with customers who have problems, that’s where product designs originate.”
Expanding on this theme, Mr. Wills (right) continued, “I have heard two predictions regarding developments in the industry that will have a profound effect on test and measurement and the way we live. First, in four or five years, the number of PCs on the Internet will increase from 150 million to 750 million or possibly to one billion. Second, in four years, the number of mobile phones will grow from possibly 250 million in today’s technology to one billion. If these predictions are anywhere near correct, there will be an explosion of new consumer product designs to be evaluated in 2000.”
As the five industry leaders look at testing these new products, the theme “we sell solutions” was used many times to explain the direction of new instrument development in 2000. “The semiconductor suppliers and communications companies are in a race, each pushing the edge of known technology,” observed Mr. Wills. “Their design and development cycles are much shorter than they were five years ago. We have to develop our test products in parallel with their product development cycles. It is a team arrangement where everyone benefits. The solution to their testing needs is what we will sell next year.”
For IOtech, part of the solution is software that makes it easy to set up the company’s products. “In 2000, we’ll expand the out-of-the-box software that can be used with just about any PC operating system,” Mr. DeSantis said.
For Agilent Technologies, the solutions will focus on hardware. Mr. Saponas (right) said that in 2000, “We will continue to see many products closely integrated with the PC. They will not be mounted in the PC chassis itself because the chassis generally is almost full now. The VXIbus is one of the practical and efficient tools for extending the accessibility of the computer buses.
“The VXIbus, already very widely used, is still growing because it gives tremendous versatility in an open hardware/software environment. This represents a significant trend in 2000,” he explained.
Despite the emphasis on computers, Mr. Saponas noted that not everything is tied to the PC. “Many stand-alone boxes are still stand alone. Some people went crazy about PC-based instruments several years ago. There was going to be a revolution. I had people telling me that in a year or two you wouldn’t see stand-alone instruments,” he said.
“Now, 15 years later, we still expect maybe $200 million in sales of instruments with front panels. One reason is because some RF and microwave generation and measurement instruments just don’t lend themselves to easier control with the computer. Also, users are just not ready to give up the stand-alone DMM or the low-end oscilloscope or counter.”
Two more factors-size and price-deserve some attention according to Mr. Keithley. “One trend that will be enhanced in 2000 is the combination of several test functions into one instrument so you don’t have to find space and money for a whole rack of test equipment when one instrument will do the job.”
While Keithley Instruments focuses on size and price, National Instruments is looking at increasing the speed of testing. “PCs have gotten 100 times faster, yet many tests are the same speed as before because the traditional test instruments haven’t changed. Our goal is to tap into the PC speed to get what we call ’10x for test.’ To do this to a greater degree, we need to do more jobs on the PC instead of on slower instruments,” Dr. Truchard explained. “We also must optimize triggers and timing and use software that coordinates events tightly.
“One big step in getting the extra speed on a test stand is the real-time version of our LabVIEW software,” he said. “The user writes a program the regular way and then uses the real-time version of the program on the test system to get optimum sequencing at high rates. Developments in 2000 will enhance this capability.
“As an extension of this concept, we are creating mini-ATE with the PXI package to bring the capabilities of big ATE onto the desktop with easy-to-use software,” Dr. Truchard continued. “We do other things to optimize speed, such as using faster algorithms, newer and faster ASICs, and some of our higher-performance modules. The same PC software, tools, drivers, and I/O capabilities can be migrated to the small system without modification.”
Other Industry Developments
While none of the companies represented in our interviews make PCs, semiconductors, cell phones, or fiber-optic systems, each leader is on top of happenings in the fields that can make big differences in his company’s direction. “Those portions of the industry are very volatile,” commented Dr. Truchard, “and we know that we can’t exert any control over them. However, we follow their movements and provide equipment as quickly as we can when a new technology is accepted.”
This is typical of the positions of all five of the industry leaders. For example, the big drivers for Keithley Instruments will be miniaturization, increased bandwidths, portability, and lower power consumption. “From a product-development standpoint, we are somewhat dependent on semiconductor design enhancements,” Mr. Keithley (right) said. “The availability of new components is a good example of this progress. The expanded acceptance of Windows NT is another factor in our plans for next year.”
According to Mr. Wills, “Technology drivers for Tektronix in 2000 will be the internet, faster semiconductors, Rambus memories, and higher-speed optical lines. Users also will begin to see more video on the internet which presents some new challenges.”
Challenges also will unfold as companies determine what PC buses and communications channels are viable for products emerging in 2000. Two-thirds of his customers still buy equipment to use in ISA bus computers, according to Mr. DeSantis. This ratio will change some in 2000, but he still is selling to many people who can’t use equipment built around Firewire, the universal serial bus (USB), the USB 2, or even the PCI bus.
“The market for USB products is still evolving for us,” said Mr. DeSantis. “Many of the computers out there are not present-day technology but they still work, and the users want to spend their money on something else rather than on new computers. USB is part of our game, but it will not be a major player in 2000. Firewire is an even greater uncertainty for us in 2000 because the industry isn’t supporting it wholeheartedly yet. We feel that USB 2 may undermine the growth of the Firewire bus.”
Mr. Keithley echoed the sentiment: “Replacement of the ISA bus by the PCI bus is coming about much more slowly than we anticipated. When you take several years to perfect a measurement system in a factory application, you will not uproot it without a good reason. System developments will use the new technologies in 2000, and we will respond to this market, but we are not about to tell ISA and GPIB good-bye.”
Dr. Truchard agreed, “Look at Firewire. It’s a wonderful standard, and a year ago it seemed to be the wave of the future. Then it got caught up in politics and may turn into a pumpkin. We have a few products in that technology, but the jury is still out on its future. The USB is more interesting to us for 2000, especially with the prospect of increased data transfer rates.
“With those higher USB rates, vision technology will get more attention in 2000,” he continued. “Cameras are not expensive, software now is affordable and easy to use, and new techniques such as pattern recognition are advancing rapidly. Vision will become another measurement tool for us.”
The subject of data rates also is popular with Mr. Saponas. “We believe it is not going to be very long before people will stop using 10Base-T or 100Base-T and move to 1-Gb/s fiber optics. The 1-Gb/s Ethernet is getting less expensive at a very rapid pace, and we don’t think 10 Gb/s is going to be very far behind. When you think about every computer and every instrument having a 10-Gb/s port on it, all of a sudden, communication is free.”
“Probably the biggest trend we see,” said Mr. DeSantis (right), “is the multiplicity of tasks that can be done simultaneously on the PC. A person can surf the internet, write a letter, participate in a chat session, and follow the stock market all at the same time. This leads to the same expectations for instrumentation products. In 2000, our products will be more synchronous to allow simultaneous input and output of a multitude of signals. Technology makes this possible.”
Other Sources of Instrument Development
Many plans for 2000 seem to be based on the ingenuity of design engineers. For example, Mr. Keithley plans to put more intelligence into instruments: “People generally think of computer control as a smart PC operating a dumb instrument. We are challenging that way of thinking. In the next year we will offer the smart instrument that can accept a whole set of actions at one time and then push its own buttons to accomplish its assigned tasks independently.”
Mr. DeSantis also discussed an interesting design. “We have a product that logs data directly onto the internet. We are eager to get feedback from our customers to see if they want us to expand this concept. Another strategy in 2000 is to proliferate the idea that one plug-in board can perform multiple functions so you don’t need to buy a unique board for each task,” he said.
Much of the impetus for new instrumentation development in 2000 is the insatiable demand for more bandwidth, according to Mr. Saponas. “A big part of the movement will be in the optics realm, and this fast-growing field is pushing harder and harder on the instrumentation manufacturer,” he said. “In fact, you can think of instruments as the enablers for this growth.”
Mr. Saponas continued, “Communications will change the environment of the instrumentation community. It also will represent a challenge, measuring 1 Gb/s, 10 Gb/s, or 40 Gb/s. Wave-division multiplexing is another very important trend as more and more information will be carried optically. To go faster, you’re going to see optical backplanes on computers and big switches.
“All of those things must happen because it is becoming harder and harder to move information fast with conventional electronics,” he said. “Right now, the state-of-the-art people talk about running 40 different-colored lights down the same piece of fiber, each running at 40 Gb/s for a total of 1.6 Tb/s.
“The physics are identical to wire transmission. It’s just that light travels much better in glass and doesn’t interfere with people in the next county,” Mr. Saponas concluded.
Mr. Keithley shared his plans for 2000. “We will continue to expand our layered software for data acquisition so that, instead of board drivers, you get true plug-and-play capability from one board to another. We will expand our setup package so you can collect good data faster and more accurately, he said.
“We see our data acquisition products coming out of the PC and taking new form factors in 2000, a trend we already started with our SmartLink miniature instruments. This will blur the distinction between instruments and data acquisition systems as it streamlines data collection,” he continued.
Globalization in 2000
Continued expansion into the global community is the next challenge for test and measurement companies. We become aware of the international aspect of instrumentation when we see similar needs on six continents. Users look for the same performance, and design teams work toward the same goals.
Mr. Wills commented on this. “We have been purchasing companies and expanding existing divisions. We bought a protocol-analysis group in Berlin and a network-monitoring group in Italy. There are hundreds of engineers in our India office plus a location in China and one in Japan. We’re working on a product now that was defined in the United States, the hardware was designed in Japan, and the software was developed in India,” he explained.
Dr. Truchard noted that National Instruments has two divisions in Germany plus a small research center in China. Mr. Saponas reported that the number of Agilent Technologies employees in other countries is approaching 50% of the company’s total.
As for income, about 50% of sales at National Instruments, Agilent Technologies, and Tektronix are exports. IOtech and Keithley Instruments also report substantial international sales.
Conclusion
As we move into 2000, we can expect to see these five companies listening to users and watching the PC, semiconductor, and closely related industries. Based on what they hear and see, they will develop new products and enhance the ones that already are on the market. We will see smaller products with greater speed, simpler operation, and more versatility. Happily, we can expect increased performance for each dollar we spend for solutions.
Welcome to another banner year for instrumentation.
Return to EE Home Page
Published by EE-Evaluation Engineering
All contents © 1999 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
December 1999