October 31, 2005
Automotive Solutions from Mentor
Please note that contributed articles, blog entries, and comments posted on EDACafe.com are the views and opinion of the author and do not necessarily represent the views and opinions of the management and staff of Internet Business Systems and its subsidiary web-sites.
| by Jack Horgan - Contributing Editor
Posted anew every four weeks or so, the EDA WEEKLY delivers to its readers information concerning the latest happenings in the EDA industry, covering vendors, products, finances and new developments. Frequently, feature articles on selected public or private EDA companies are presented. Brought to you by EDACafe.com. If we miss a story or subject that you feel deserves to be included, or you just want to suggest a future topic, please contact us! Questions? Feedback? Click here. Thank you!
In April I wrote an editorial pointing out the significant opportunity for electronics in the automotive industry and therefore the opportunity for EDA vendors to offer tools to support that market sector. Undoubtedly as a result of that article the very next month Mentor Graphics acquired Volcano Communications Technologies AB (VCT), a provider of network design tools, in vehicle software and test and validation products for the automotive industry, for a purchase price of $23 million. VCT customers include AUDI, BMW, Daimler Chrysler, Volkswagen, Volvo, Ford, Fiat, PSA, Bosch, Delphi, Magneti Marelli, Siemens VDO and Visteon. VCT was founded in 1998 and was headquartered in Gothenburg
Sweden. VCT's product portfolio includes network design tools, embedded software and test and validation tools for major automotive networks.
In March 2000, VCT joined Motorola Audi, BMW, DaimlerChrysler, Volvo Car Corporation, and Volkswagen to form the LIN consortium that defined a cost-competitive, sub-bus network solution. LIN complements Controller Area Network (CAN) solutions by providing customers with lower cost connections within local network clusters.
Volcano is also a member of Autosar (AUTomotive Open System Architecture) partnership which was formed in September 2003 by BMW Group, DaimlerChrysler and Volkswagen, and automotive system suppliers, Bosch, Continental and Siemens VDO to jointly develop and commercially release a standardized Electrical/Electronic (E/E) architecture concept. The goals of this partnership include the standardization of basic system functions and functional interfaces, the ability to integrate and transfer functions and to substantially improve software updates and upgrades over the vehicle lifetime. The AUTOSAR scope includes body electronics, powertrain, chassis and safety as well as multimedia systems,
telematics and man-machine-interface.
Oct. 6, 2005 Mentor announced it had significantly extended its set of electrical and electronic design solutions for the automotive market with the recently acquired Volcano automotive networking solutions and SystemVision, a new modeling and simulation tool for automotive mechatronics sub-systems.
Before this announcement I had an opportunity to talk with Larry Anderson, Director of Marketing for Mentor's Automotive Networking Business Unit.
What is going on in the automotive sector at Mentor Graphics?
What we are seeing in this release are two things, a kind of automotive initiative of multiple products that address the electrical and electronic design process in automotive and the second is specifically two new items to that portfolio in the areas of automotive networking tool and mechatronic simulation.
We are doing more product development focused on emerging challenges, new markets such as transportation which will break down into auto, aero and rail. Obviously one of the things we have been doing is acquiring new technology from outside companies. When we do make acquisitions with companies that typically have small sales and marketing organizations, we marry them with our global sales and marketing distribution. It is really a win-win situation.
What is happening in the automotive industry to create so much interest with Mentor Graphics?
The bottom line is that automotive has become a sort of distributed computing challenge, distributed computing on wheels. Some of the high end luxury vehicles have upwards of 80 to 90 electronic control units, ECUs, all networked together controlling all of these distributed mechatronic systems like airbag inflation, antilock breaking, door lock, you name it.
We do see some differences in that time to market is not always the critical factor in this industry. It is really time to market plus warranty costs. Avoiding warranty cost in many cases is approaching even out stripping development cost and risk management and liability.
Mechanically dominated process in the past is becoming more and more electronic dominated: electrical distribution systems, automotive networks, the actual design of electronic control units, and embedded software development - making sure it all works together. Mentor is uniquely positioned in that we do have tools in all these areas to address the electrical and electronic design process. Many people have known Mentor from a PCB and FPGA standpoint. The design of an electronic control unit is similar to the design of a single board computer. We also have system integration, electrical distribution tools, the hardness tools and we have now added tools for network design and
mechatronics simulation. Mentor is the only EDA company with its own embedded software division. We also have a host of embedded IP that is targeted to automotive over and above the normal markets we serve like consumer electronics. A host of tools there.
You know Mentor from doing a lot of business in telecommunication, having customers like Ericsson, Nokia, Lucent etc. A lot of business was done in automotive was to tier two suppliers, the big semiconductor houses: Infineon, Freescale. We have done a lot of business especially with our cable harness tools in the major OEMs. We have a lot of major OEMs that are Mentor customers along with their tier one suppliers who are doing complex subsystem design. We have moved Mentor up in the value chain so to speak in automotive. Our goal would be that with these new tool additions to have an even more robust portfolio for these types of customers.
The trend in the nineteen seventies was for pretty simplistic electronic systems, maybe 3 or 4 systems, if you were lucky enough to have air conditioning. Pretty simple networks. Not a lot of traffic on these networks between the systems.
Since that time there has been an explosion in the number of electronic systems. A typical vehicle today might have a couple of dozen systems.
Each of these could be comprised of multiple ECUs that are networked and running a good amount of embedded software. To make it more complicated they're all talking to each other. Lane departure warning and adaptive cruise control are interacting. Body electronics in the man-machine interface are interacting. They are all basically sending signals over the same network. It's really become a difficult process because the OEM is getting a lot of these subsystem designs from their tier one suppliers and they are trying to integrate all of them to work at a system level. The tools we are going to talk about really address that system level design.
Continuing the trend there is a lot of simultaneous things that are happening with these controllers. In automotive you have much longer cycle times. In the mid eighties to mid nineties circa three years. We have vendor OEMs that are saying for a new platform there's a 2 year turn around cycle today. I was in Japan a couple of months ago talking with Densu who is a major supplier to Toyota. They were saying Toyota is striving for 12 month development cycles by the year 2010 that is approaching what we saw in consumer electronics and telecommunications in years past. So a lot of pressure there, the same as we have seen in other industries to get product out the door faster, to adapt
to changing markets like the emerging market in China or for example a new hybrid vehicle for fuel efficiency.
At the same time you have to try things out faster but the electronic content is growing dramatically. In this case a car in the 1970's 2% of the total car price was attributable to electronic development and content. In 2004, today's timeframe, we are seeing about a quarter of the total car price due to electronics and growing to 35% in 2010. I've had an executive at BMW tell me that in the seven series it could be as high as 40% to 50% by 2010. A lot of electronic content and again mechanically dominated and differentiated design process is moving more and more to electronics. Within electrical and electronic we are also talking about embedded software. OEMs are using better
software to differentiate their products.
At the same time they are being challenged to get more and more vehicle launches out, a lot of these are because of specialty brands like high end sports brand. Some of these brands are due to fuel efficiency like hybrid. Some brands are being developed for new and emerging markets. There's a huge opportunity that a lot of vendors see in China.
The vendors are trying to get out more products with more variants and options. Unfortunately they are doing this with less budget. R&D budgets remain flat at best. They need to develop more with less, be more efficient and meet all the constraints we just talked about.
Warranty costs are pretty staggering numbers as a percent of sales. GM and Ford in 2003 had ~$8B of warranty expense. The impact this would have had on their profitability is huge. Trying to avoid or minimize warranty costs is really a big thing. What it comes down to is that the OEMs are just not doing enough upfront validation of their product prior to going to market. These warranty claims all reflect field failures. Albeit some of them are mechanically related, Ford Motor Company in the year 2004 attributed $1.12 billion of their warranty costs to embedded software issues and that would be about 1/3 of their 2003 numbers. A huge amount of that is related to electronic content.
What is Mentor Graphics' response to these trends?
Mentor's response to decreasing cycle times is to focus on design reusability and specifically the reusability of embedded software in the system. The Volcano tools for network design and verification. Correct by construction, design rules, top down design methodologies and data and change management. Be able to manage the huge amount of options and variants in these vehicles.
Mentor's response to increased electronic content is complexity management. They have to deal with a lot more complexity which we help them manage. They need to virtually prototype more of the system than in the past that they physically prototyped not only the hardware and software but also the mechatronics, the mixture of mechanical and electrical systems that exists in the vehicle. More and more design automation. This is an industry that doesn't have a lot of experience with embedded software development and finally system integration and component design.
Mentor's response to doing more with less is network design automation. We have interesting technology in our wire harness of composites wire synthesis. We support the automatic generation and viewing of system diagrams. A lot of this is related to the physical design side of our Capital Harness System that allows you to develop the electrical distribution system and create all the detailed harnesses. In many cases the entire harness infrastructure can be one of the most expensive components of any vehicle.
The obvious way to avoid warranty and testing costs is to do more and more robust system simulation as opposed to physical prototyping in the field where you can't do as many what-if analyses. Again automation of design and verification, analysis for decision support. Network and verifying the network has a lot of warranty costs attributed to embedded software but more particularly it is due to traffic on the network and messages that were lost or delayed that did not meet their timing specification of the consumers of those messages.
Another trend is embedded software growth. It is truly staggering how much of a percentage of total production costs this is and actually increasing. A lot of the differentiation of the vehicle is done through embedded software. Not only in the release of the vehicle but this maps out into the service bays. In a lot of cases when you have vehicle defects, you are not getting new mechanical components, you are simply getting new software being reflashed into the ECUs. Obviously you need good and better software development environment. OSEK is the standard for real time operating systems in automotive. CAN is the control area network. The industry whether it likes it or not is being
exposed to a lot more to embedded software and needs to become more sophisticated and management of that software.
Where does this fit into the overall electronic design process?
The V diagram is often used to describe the process in automotive and aerospace.
It starts with requirement definition in the upper left, moves all the way down to the detailed component design and then as you move up on the right hand side component test, integration other components, system test, verifying that those requirements that were specified were met. Mentor has been known in the component design area predominantly in the past. We've obviously been working with a lot of the tier one suppliers for PCB design, harness design for tier one and OEMs and OEMs for embedded software. That allows us to get the ECUs, electronic control units, developed. What we are really doing is moving up to areas of higher impact, more into the system design and system test and
integration process. The products we are introducing are in this area of vehicle networks and network verification, mechatronic analysis. Obviously, the embedded software and electrical distribution system that physically maps all these signals to the different areas of the vehicle.
What about specific new products?
We acquired automotive network tools from Volcano. This was a Volvo Venture technology firm that we purchased in May of this year. They had a lot of customers, domain expertise in automotive. The founder and his design team did the network design for the Volvo S80 platform. They commercialized their consulting efforts into products. Basically started going mainstream with it. Volvo at the same time said that they didn't want to have this stuff in house. Their competency is the design and manufacture of automobiles and not development of tools. We looked at this technology and said we don't really have this particular domain expertise inside the company as we do in other areas. So
we made the acquisition. We are happy to have these tools and domain expertise in house.
The reason is that we have customers today that are experiencing and as a consumer of the vehicle you have probably experienced this yourself when you get into a high end vehicle and start it up and get an airbag fault or traction control failure or a car going into safe more where you lose part of your cockpit interface functionality. The car tells you to drive to the dealership. Again in the end you get better software in the vehicle. A lot of that is because the bottoms up approach in network design. A lot of the automotive network design today is done in Excel spreadsheets for controlling the network and the LIN (local interconnect network) which is a more cost sensitive version.
There are producers of signals like engine speed, temperature, rpm and there are consumers of these signals. Designers look at the mapping of the ECUs that would talk to each other the most, like transmission control and engine control. They would go through several scenarios. For example, this transmission control needs rpm data every 5 msecs. Then they would design the network based upon those requirements. They would do some limited testing. Very inefficient network communication would happen. In many cases there will be a lot of frames. These messages basically are delivered in what is called CAN or LIN frames that are sent along the network that have only a few new updated
You are seeing a kind of movement from this backend-loaded process where you do some limited design, definition and simulation, and a lot of integration and testing
where you suffer a lot more in warranty recall towards analysis and upfront design where you do this correct by construction, integration and validation where you try to minimize warranty and recall. The way the Volcano tools do this is that they have a tool called Volcano Network Architecture.
A typical automotive network will contain subnets, it's a hierarchical network. There are gateways between CAN and LIN. There's high speed CAN and lower speed CAN based on the requirements. For example, there are high speed CAN engine controller and a low speed LIN for body electronics. What the tool will do is based upon the topology of the network and the requirements, i.e. so many producers and consumers of a signal. The system consumers of signals need to have data typically at a periodic rate and at a minimum timing what we are calling the maximum age of the signal. Transmission control which needs rpm data would use the maximum age of the signal is 5 msec. It would need to be
refreshed in less than 5 msecs. Knowing that information, the tool automatically packs the CAN and LIN frames and optimizes the communication matrix. The tool will do that not just with the frequent paths of communication but it will go through all paths. If you have a 40 ECU network, it would perform an analysis of ECU 1 talking to ECU 2, ECU 1 talking to ECU 3 all the way through to ECU 40 and all the permutations so that it can verify that the maximum age of a signal will never be exceeded and it will calculate the utilized network bandwidth.
Volvo was experiencing about 20% to 35% network bandwidth utilization using a bottoms up approach. They would design the network and build in extra headroom because of the corner cases which they can't test virtually and because they don't have the proper tools and they don't feel they can cover in field testing. Of course that overhead is represented by more expensive microcontroller, more memory .. that drive up the cost vectors of the car. That's not a good thing for the OEMs. In the end it turns out they still have issues because there were spots in the network where they didn't have enough headroom to cover.
This tool with its correct by construction approach will verify that with that network topology and requirements that all the signals will be able to be scheduled. If not, you have to go back and adjust the network. In some cases, if for example you are putting in adaptive cruise control, you have to put in a higher speed CAN network and going the other way, if you are reducing some subsystems for a more cost sensitive model you may be able to actually take a gateway out and re-simulate and find out that you still meet the original network requirement. It verifies that the maximum age of a signal will not be exceeded. Using this approach Volvo was able with the F80 to move up to as high
as 80% bus bandwidth utilization. Much more efficient network design and much more reliable. This is the approach we are seeing with the OEMs. The interest here now is because they have to integrate all these subsystems and they are really finding that a lot of their electronic content warranty is related to networking issues.
The Network Architect is only one of the three tools that we have from Volcano. Over and above the ability to architect the network, if you enter the topology, enter the requirements and verify that everything is correct by construction then you can automatically generate all the low level drivers for the CAN/LIN network. This is actually embedded software that goes into the ECUs. You can think of it as a protocol stack basically for CAN or LIN device drivers that is really a comprehensive target package. It allows the application through a signal based API to interface with the CAN or LIN box. All the application has to worry about is getting the engine speed. It doesn't care how I
get that off the CAN or LIN bus. The embedded software device drivers will take care of that. It is ISO9000 certified. Very low memory footprint which is important for cost perspective, very high quality software.
The other thing that the Network Architect does is custom validation tool. We have a hardware interface to the network called the Tellus network interface that sits on the CAN or LIN network. It has 2 CAN ports and 2 LIN ports. It also has 1 MOST (multimedia related bus) port used for entertainment. In the simplest form it is an FPGA based box that's firmware configurable, for example just a data logger in a prototype vehicle. The next level up is a network analyzer. You can think of it as a logic analyzer specific for CAN or LIN networks. Because it knows the information from the Network Architect tool, it knows good network behavior, it can flag errant frames on the CAN or LIN bus
to the user. In its most advanced form it can be used for emulation. It can emulate a known good network so it can interface the virtual prototype world into the physical prototyping world. You can actually have it talk to a prototype or a legacy ECU on the CAN or LAN network through that interface. It is a multi-protocol, very tight time resolution product. So we are pretty excited to have that in our product portfolio.
Does the test and validation functions end up in the consumer's vehicle for diagnostics?
No it would not. It could end up in the service bay but not in the end vehicle. It would be basically sitting there looking for network behavior and analyzing that. So without question that could be applicable to a service bay that would plug into the CAN or LIN network and look at for errant frames on the network. Down the road I suppose some of the technology could be deployed into the vehicle but obviously the cost sensitivity of diagnostics on board is pretty high. It definitely can be used for the design of the network.
What about the new product System Vision?
There are many tools out there that do high level modeling like MathLab Simulink for requirement specification and system partitioning. There are also many tools that allow you to detailed component design and modeling like PSPICE. What System Vision does is to bridge the gap by supporting multilevel abstraction. You can have something as highly abstracted as a S&V control diagram interfacing with a transistor, motor, spring; something as simple as a resistor. Something very, very detailed. Multi-abstraction is appealing to a Simulink user that wants to move down and do more detailed modeling or to a PSPICE user who wants to move up and model more disciplines. The unit supports the
analog mixed signal extensions to VHDL, i.e. VHDL-AMS, but also supports SPICE, C and C++. That allows us to support anything they can be modeled with differential equations, hydraulic, mechanical, thermal, magnetic etc. For the user who wants to stay with Simulink for their control algorithm, we have an interface to Simulink. You can have part of the design with Simulink and part of the design like detailed components and sensor/actuator in SystemVision through a co-simulation interface.
What about pricing?
Very competitively priced! We have set this tool up to be scalable to the user starting out. More SPICE centric version of the tool with limited VHDL-AMS capabilities that would be appealing to a SPICE user that might eventually move up to VHDL-AMS capability for modeling more disciplines including advanced analysis for multi-run analysis like Monte Carlo, worse case sensitivity and the Simulink interface.
The low end product (limited in the number of quantities that can be simulated) is priced below $10K. The full blown product which has the full VHDL-AMS capabilities, advanced analysis and Simulink interface is priced at $38K. You are in the range of desktop pricing rather than Enterprise pricing which is typically $25K to $35K departmental signoff. The user has the right amount of simulation capability.
Where does this fit with other Mentor tools?
It is also integrated into all the Mentor flows. It uses the same front end - the schematic capture system of our PCB tool, DXDesigner, it maps into our harness system, Capital Analysis, which allows you to do qualitative or go-no-go analysis on the harness. If you want to do more detailed analysis, for example did a fuse blow or not, it's more a question of why did it blow, where did it blow in time, what voltage level did it blow at, then you can plug SystemVision in and it would give you that numeric or quantitative analysis capabilities. It also interfaces to the HDL Designer and ModelSim. It also supports DX Databook library from our PCB tools. Multilevel abstraction modeling,
multi-technology modeling with VHDL-AMS to fully model a complex electrical-mechanical system. High model reuse and model exchange through standards. We are working with universities and industries to get more VHDL-AMS standardization out there. We are seeing a lot of interest in this language as a standard language to exchange models or components in a supply chain.
We talked about Volcano target packages as embedded software. You need more than these target packages; you need tools and embedded IP. We have full blown Nuclear Bench, an Eclipse based integrated development environment for the development of embedded software: compilers, linkers, loaders, debuggers, you name it. We also have Nucleus Bridgepoint, the executable UML high level modeling for which we see interest in automotive.
Nucleus SimDX is used for graphical prototyping the embedded software applications in your systems. We have used it in telecommunication with companies like Nokia, where you have a graphical mockup of keyboard and display, i.e. man-machine interface. You can see how your embedded software application interacts with that interface. In automotive you would have the cockpit, the actual man-machine interface of the vehicle to see how your embedded software application would interact with that.
What about AUTOSAR?
AUTOSAR is automotive open standard architecture that was really driven from the European automotive marketplace a few years ago. Its goal quite simply was to abstract away the application from the details of the electronic control unit, the hardware, the microcontroller, the network such that embedded software could become interchangeable or reusable IP just like a mechanical or electromechanical component can today. In doing this the industry is striving to interchange, e.g. a wiper control function, that they can bookshelf, have better software and actually retarget that to different ECUs to do cost tradeoffs or maybe do a cost reduction for a lower cost model. Through the standard
API, the AUTOSAR runtime environment, they can really look at embedded software as an interchangeable component. It is populated by all the major OEMs and tier suppliers as far as membership. With this announcement we are also announcing that Mentor is a premium member of AUTOSAR through our acquisition of Volcano Communication Technology. We are obviously interested in some of the basic software areas and in sitting on those subcommittees. We have been donating technology for the communication areas of CAN and LIN. We are going to continue to make the investment. We really believe that there is going to be an inflection point in the industry. It's going to aggregate the industry
somewhat because it will allow the tool vendors to look at the market not as a niche market where I can only create tools specifically for GM or for Toyota or for BMW. Now I can create tools for any AUTOSAR compliant application and that opens up the entire OEM and tier one marketplace to me. A key standard being driven very aggressively for an industry that moves relatively slowly because this whole issue of embedded software is such a big issue for the OEM.
Are there links to other systems?
The SystemVision product captures the schematic, produces a mechatronics model and allows you to you look at the waveform analysis. The control algorithm interacts with the mechatronic model and through the link to MathLab Simulink. The control algorithm is broken down into functions running an ECU model. Here's where we can start looking at the future where we can bring in some of the hardware/software co-verification modeling technology from products like Seamless, for more detailed ECU modeling. The ECUs do not live by themselves. You have network tools and that's where Volcano tools play a part. For a complex vehicle this could go on for 40 to 60 ECUs could be networked. You have
very complex overall network. You need to look at a system level analysis. Our goal is not just to provide design creation simulation and analysis tools but to virtually prototype the entire control algorithm to mechatronic interaction and the network that is between all these control systems and mechatronic control systems. What we are really striving for is to have a standard that allows us to retarget function to an ECU. You now have a set of tools that allow you to do architectural exploration.
Now we can create an environment that allows you to retarget the application to the ECU. I can look at system optimization by moving functions among the ECUs. For example I might want to take a function that exists on one ECU and move it to two smaller ECUs that exist in the right and left hand door assemblies because they are closer to the body electronics that they control and because that move has a positive impact on the cost of the harness system. Or on the other side for performance reasons I can take 2 or 3 functions that are distributed across multiple ECUs and consolidate them in one higher performance ECU for a more intense application like drive by wire. This what-if analysis
at the architectural level and architecture optimization is something the OEMs are striving for, this is effectively what their dream is. We believe that we have the tool technology combined with AUTOSAR as a true standard to allow the application to be retargeted. The AUTOSAR basic software plus the network abstraction we get from Volcano really gives us a pretty comprehensive tool set.
Would you summarize all of this?
A very comprehensive product portfolio for the design of electrical/electronic design challenges in automotive a lot of domain expertise we have not only in our factory organization. I mentioned the experience out automotive networking team has. But we also have people in our field organization that are coming from automotive to work with customers deploying technologies with new projects. Mentor is now a premium member of AUTOSAR. We are pretty excited about that standard. We will continue to make the technology donations and staff of engineers. We believe that there will have a major inflection point in the industry for embedded software. We are very focused on this industry. We
believe it's a very exciting industry. It has complexity challenges that consumer electronics and telecommunications have had in years past. The networks are very specific, the microcontroller are very specific, the electromechanical the focus on warranty costs, risk management and liability. If we can help with those OEMs and tier one suppliers in the end avoid any of that warranty cost, it's a huge opportunity for us.
The major automotive companies are heavy users of mechanical CAD systems like CATIA and Unigraphics. Are there any integrations or interfaces with these systems?
There is a very important integration between the mechanical tools and the Capital Harness System tool for the routing of the electrical distribution system. This is not a simple 2D challenge for board layout. You really do need to bring in the 3D model of where there is space to route the wire harness and how much space actually exists. There is a very robust interface between UGS and CATIA tools just for that purpose. Then there is also the very important item of full product data management which these mechanical systems offer. We have a subsystem of that on the electrical side within our tools that also has an interface to PDM (product data management) like UGS Teamcenter. These
interfaces are available. Coexistence with mechanical CAD is extremely important and the functional and physical side is bring those two worlds together is also important because you could make an impact on electronics and have a very negative impact on the mechanical side and that could be a huge cost factor.
You purchased the technology from Volvo not too long ago. Is this announcement essentially their technology integrated with Mentor's environment?
It was not so much Volvo per se. It was a Volvo venture firm as Volvo has its own VC arm and had this technology in incubation. The technology actually comes from outside, from the team that basically comprises the Volcano tool. Volvo was the first commercial customer they worked with to really perfect that technology but it is not vendor specific by any means. We are working with other major OEMs. We have brought that technology into Mentor and are in the process of integrating that in with our other tools and also into our sales and marketing organization because Volcano Communication Technologies from a sales and marketing standpoint had very little resources. We have already seen
some major success in markets where they couldn't reach. We will be making some announcements in the near future.
The Volcano architecture, embedded software and a predecessor of the Tellus product called Inspector were also used by Volvo and its tier one supplier.
Who do you see as competition?
In the area of embedded software there are several vendors out there. Vector is probably one of the more prominent ones. It's a German based company well thought of in the automotive and embedded software space. 3Soft is another firm. In test and verification Vector is the major competition. But with our multiprotocol support and out Tellus interface and time resolution we have a very competitive product.
Vector is a manufacturer of software tools and software components for networking of electronic systems based on CAN, LIN, MOST, FlexRay as well as multiple CAN based protocols. Founded in 1988, Vector Informatik currently has - together with Vector Consulting GmbH - 540 employees and annual sales of 79 million euros (2004). Besides its corporate headquarters in Stuttgart, Vector also has subsidiaries in the USA, Japan, France and Sweden. The firm has customers in the automotive, commercial vehicle, transportation and control engineering industries. Vector is a premium member of AUTOSAR
3SOFT was founded in 1988 by three software engineers. It is headquartered in Erlangen, Germany and has branches in Braunschweig, Stuttgart, Tokyo and Detroit. At the end of June 2005 the firm had 299 employees. In March 2004 3SOFT became a company of the Finland-based wireless engineering company Elektrobit Group Plc. 3SOFT is a premium member of AUTOSAR.
The top five articles over the last two weeks as determined by the number of readers were:
Mentor Graphics Announces Third Quarter Results; Reaffirms 2005 and 2006 Guidance
Mentor had record revenue for a third quarter of $164.8 million. Earnings were break-even on a GAAP basis, and $0.04 on a non-GAAP basis. Bookings increased to the highest level ever for a third quarter, and were up about 15% on both a year-on-year and a sequential basis.
In a counterclaim suit Magma asserts that the applications Synopsys made to the U.S. Patent and Trademark Office knowingly concealed relevant prior art that described the inventions claimed in the applications. This would constitutes a fraudulent application, and that Synopsys' Sept. 26 lawsuit against Magma was therefore an attempt to enforce a fraudulently obtained patent and constitutes a violation of Section 2 of the Sherman Antitrust Act.
Mentor Graphics Releases Expedition Enterprise - The Next Generation PCB Systems Design Solution for Enterprise Electronics Companies
The new flow encompasses DxDesigner for design definition, CES for constraint editing, Expedition PCB for layout, and DMS for data management. Also included in this flow is ICX Pro Explorer, which is used to evaluate and verify constraints defined early in the process. It is built on Mentor's AutoActive technology, and uses the same database and environment as Expedition PCB.
Assertain success boosts investors' interest in TransEDA Assertion-Based Verification strategy
The new funding brings investment in the company's coverage and formal technologies to over $18M. The company offers Coverage and Verification Measurement solutions for Assertion Based Verification (ABV) design flows. Assertai was announced at DAC 2005 in Anaheim.
Implementing Low-Power Configurable Processors - Practical Options and Tradeoffs - Technical Paper from DAC 2005
This work explores the tradeoff and analyzes the results: the design priority tradeoff, the process technology impact, and implementing low-power configurable processor using commercial scaled-VDD cell libraries compatible with mainstream SOC practices.
Cadence Reports Q3 Revenue up 12% Over Q3 2004
Third quarter 2005 revenue of $337 million, an increase of twelve percent over the $302 million reported for the same period last year. On a GAAP basis, Cadence(R) recognized net income of $21 million, or $0.07 per share, in the third quarter of 2005, compared to $20 million, or $0.07 per share, in the same period last year.
Other EDA News
Other IP & SoC News
You can find the full EDACafe event calendar here
To read more news, click here
-- Jack Horgan, EDACafe.com Contributing Editor.