Breaking News
Blog

EDA bashing

NO RATINGS
View Comments: Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
Ispeakfrench
User Rank
Rookie
re: EDA bashing
Ispeakfrench   2/11/2009 4:50:27 PM
NO RATINGS
Hmmm, I must be qualified to bash EDA since I have no idea what it means. Electronics Distributor Association?

ekne
User Rank
Rookie
re: EDA bashing
ekne   2/8/2009 5:15:20 AM
NO RATINGS
EDA tool vendor survival strategy ?? Will the EDA tool industry put the life back into SFLM ? Looking at the type of companies around the business concept of selling a tool for a long term fee but the usage is a few weeks a year is a dying concept, and surly the tool vendors sees that... So why not revive the concept of pay-per-use ? And at the same time also consider hardware/software bundles.... Well, some of you in accounting will turn around a couple of times, since now you can't have a predicted income to book, well welcome to the 2:nd millenia.... The technology is there and the "accounting" can be utilizing the many different technologies such as "prepaid" accounts.... Smaller companies will twist and tweak for any number that is in the 6 figures, for a tool license and a short usage.. Just let this fly around your head a bit...... You get a "pre-paid" account with the possibility of recharging it. a la phone cards, yes you heard it for the first time here, and I will let the idea fly around You get the license when you need it and then no more charge, instead you move on to another phase and another license. You also need some more compute power during 1 month... well bundle it with a license. So what is preventing this, well some legals and where the data resides when it comes to hardware... Well you can deliver hardware onsite, just like you deliver a photocopier or rent-a-tv/sofa so it's not really that hard to do it.... and licenses are still part of the software space so managing the licenses a la SFLM or local license servers where you go to an offsite to download a license file has been around for many years so all bit's and pieces are there... So why isn't the EDA tool vendors looking at this, to get some market from smaller design houses ? Even medium and big size would be interested .. So my fellow EDA Experts, why isn't this an already working concept in this industry ? Bengt-Erik Embretsen

SDBrorson
User Rank
Rookie
re: EDA bashing
SDBrorson   2/6/2009 5:57:39 PM
NO RATINGS
HWisnotSW says: "Sure, I've taken classes in C/C++, and I can occasionally do something useful with it (& Perl) when I get the opportunity, but I call that what it is: hacking....." So I guess I am a little confused about what point you are trying to make. Is your point "I am a HW engineer, so I can't use open-source software"? If so, then that's absurd. You just download an application, install it, and use it. For the vast majority of cases, no hacking is necessary. Or is your point "I am a HW engineer, so I can't hack/modify open-source software"? If so, then fair enough. You can judge your own interests/talents. But a couple of points: * With open-source, even if you don't want to hack, others are not prohibited from modifying or improving the code to their own benefit. And the design teams I have been on have a wide variety of talents, including some folks who *do* have programming skills and like to use them. We all recognize those guys as being particularly valuable. * More importantly, at least some customizations are pretty trivial. For example modifying the format of a report for better searchability. That's just some simple mods to a printf statement. Not all useful tweaks and mods require a full understanding of the program. So in summary, open-source hands *you* control over your design tool. Whether you exercise that control or not is up to you. Commercial software has its advantages, but it never gives you complete control. Anyway, I have made all the useful points I can make on this thread, so I'll try to exercise self-discipline and not comment any more.... :-)

HWisnotSW
User Rank
Rookie
re: EDA bashing
HWisnotSW   2/6/2009 6:00:03 AM
NO RATINGS
"The user has to be fairly clueful about how to use the tool, and he may need to made some mods to the tool to customize it to exactly his purposes. ..... However, most chip designers are pretty computer savvy. ..... Also, when considering "make vs. buy", if you have to do a little hacking with a tool to make it work for your purposes, well, great! You can do that with open-source!" Unless I have a really narrow exposure to the breadth of what the typical ASIC/FPGA designer can do, this strikes me as a hopelessly optimistic view of the kind of skill & experience it takes to master another's profession. Sure, I've taken classes in C/C++, and I can occasionally do something useful with it (& Perl) when I get the opportunity, but I call that what it is: hacking. To be expected to be software-savvy enough to muck around in hundreds of thousands of lines of code to tweak the one thing you need different in a complex EDA tool whose success your company is counting on for a multi-million dollar chip project .... and be a grade-A whiz with the HDL languages to design the chip itself? I hope that person is getting paid well! I'm reminded of all the analog & RF engineers who see what I do and casually opine "Gee, I'd like to do an FPGA some day ....". One can almost see the thought-balloon above their head (".... after all, it's just ones and zeros") as they jump in to tackle a design I know I'm going to be asked to rescue in a few months. "Yes, you're paying your chip guy to fiddle with Icarus, but what's the difference between that and paying Synopsys to fiddle with VCS?" The difference between paying an Army medic vs. a board-certified cardiac surgeon to do your heart bypass operation. Engineers (of whom I am one) are famous for knowing the cost of everything, but the value of nothing. "On the other hand, with open projects, many hands make light work, ...." But is it WELL-DOCUMENTED work? Didn't think so. ".... so over the course of time, features are added, and the program evolves." ... and spaghetti code is created.

SDBrorson
User Rank
Rookie
re: EDA bashing
SDBrorson   2/5/2009 2:52:35 PM
NO RATINGS
Seibe asks "Why is it that the penetration for say open-source/free simulators is less than 1%..." I have also wondered this for a long time. Here are my thoughts: * First off, there is a mind-share problem. Lots of hardware designers are simply unaware that there is an open-source ecosystem out there. They are unaware because the EE press does not write about it, so they don't hear about it. Indeed, with the exception of a couple of articles by Richard Goering, the EDA press has taken great pains to *ignore* the movement. (OTOH, the Linux press has written about it, but hardware designers apparently don't read those mags.) I know for a fact that they have ignored the press releases issued by Fedora, gEDA, etc. I see two reasons for this: 1. The open source folks don't purchase advertising, so why cover them? 2. There is open hostility on the part of some EDA journalists against open source EDA because it is perceived as a threat to the EDA industry's profitability. And EDA journalists identify themselves with the industry. * Secondly, there is the FUD problem. Just as nobody was ever fired for buying IBM, nobody is fired for paying $100K for a seat of VCS. On the other hand, if you use an open-source Icarus (or ngspice or gnucap), chances are good that your boss will ask you if you've gone insane -- rightly or wrongly. Why stick your neck out having to defend the choice of using an open source solution? FWIW, folks who do use, say, Icarus, in a commercial often use it as a supplement to their one copy of a commercial simulator. They can take a design under construction, and run a whole bunch of different variants of that design (or different corner cases) on different computers nightly using multiple copies of Icarus. That way they can speed up validation of their design by checking many different design scenarios without the bottleneck imposed by EDA licenses. Then, when it comes time for final design check, they can run the final design using their *single* licensed copy of VCS as a kind of insurance policy. That way you reap the benefits of multiple simulators *and* get the warm & fuzzy feeling provided by your $100K license fee. There may be other reasons too; I'd be interested in hearing them.

seibe
User Rank
Rookie
re: EDA bashing
seibe   2/4/2009 6:38:58 PM
NO RATINGS
I have been asking this question for a long time! Why is it that the penetration for say open-source/free simulators is less than 1% (at least according to some older data I got from Deepchip)? Icarus & Co. have been around for a long time, long enough for people to have picked them up and develop into something competitive. My conclusion is that ultimately people crave the (arguably false) sense of security given by a support organization even if it costs them hundreds of thousands of dollars. But then why didn't somebody create that organization (a la RedHat) and charge a fraction of the cost of a VCS/NC/MTI license? Anybody has a good answer?

SDBrorson
User Rank
Rookie
re: EDA bashing
SDBrorson   2/3/2009 6:06:17 PM
NO RATINGS
Just wanted to riff on a comment made by Joe Hupcey III: "This is only 2/3 true -- open source is not actually zero cost. There are very real costs when you go down the "make" route of the "make vs. buy" decision when you decide to use open source. I'm not saying open source is a bad thing -- far from it. I'm saying it's a fallacy to say open source is free." I think we can agree on the 2/3 ratio! :-) Open source EDA is not zero cost in many cases. The user has to be fairly clueful about how to use the tool, and he may need to made some mods to the tool to customize it to exactly his purposes. Time is money, and designer time is expensive. However, most chip designers are pretty computer savvy. They are used to tying a bunch of tools together using Perl or TCL, so programming doesn't scare them. Also, they're already on staff, so there's no additional overhead to the company. Time *is* money, I agree, but when making budget decisions, the business of no overhead is attractive to managers. Also, when considering "make vs. buy", if you have to do a little hacking with a tool to make it work for your purposes, well, great! You can do that with open-source! If you buy VCS (to pick on an arbitrary tool), and you want to customize it, well, you're SOL unless you want to pay Synopsys big bucks to do it for you. With, say, Icarus Verilog, if you want to customize it, well the source code is right there for you to do whatever you want! Yes, you're paying your chip guy to fiddle with Icarus, but what's the difference between that and paying Synopsys to fiddle with VCS? Open source *does* have it's own issues. Since it's hard to turn an open-source tool into a revenue stream, the tools tend to lag behind their commercial equivalents. The reason is simple: With a revenue stream, a tool company can afford to pay a team of developers who will spend 40 hours/week (minus meetings and other time-wasting corporate overhead) working on improving the tool. Development is therefore continual. With open-source, the developers tend to be engineers who have a particular need, and spend nights and weekends implementing their required feature. Absent of corporate backing (i.e. like Sun's support for OpenOffice), the developers treat their projects as a hobby, or perhaps as a career booster. They can't write software full time since it doesn't pay the bills. Therefore, progress (new features, etc.) on any particular tool is slow. On the other hand, with open projects, many hands make light work, so over the course of time, features are added, and the program evolves. You also say: "Furthermore, one way that EDA vendors really add value, i.e. how EDA vendors "compete with 'free'" and/or fight commoditization, is in the methodology guidance and support we provide our customers, in tandem with a willingness to assume some of their risk." We probably agree about 50/50. Another way to say it is: There is room in the world for low- to mid-level CAD for electronics, which is where the open-source stuff tends to be used. The big EDA houses are therefore forced to continually push the high-end, as well as offer services and support so that their markets are not completely commoditized. However, I am not certain that services and support are winners against open-source in the long run. Why doesn't somebody create a services and support business for open-source EDA? After all, it seems to have worked for Red Hat in the market of operating systems.....

Lou Covey
User Rank
Rookie
re: EDA bashing
Lou Covey   2/3/2009 5:16:59 PM
NO RATINGS
Michael makes some good points. Traditional advertising has not been working because the web has made our marketing goals much more granular. It's the EDA companies that began saying 10 years ago that there were only 10 companies they cared about, so why did they need to reach large audiences. The web gives the granularity EDA marketers want. The funny thing is that 60 percent of the decision makers in electronics and technology as a whole, use social media to make their buying decisions, yet less than 20 percent of the marketers in those niches use social media. Generic social media platforms like, Linkedin, Facebook and YouTube, have been been used with some success, but they are not modifiable for direct applications like EDA. Ning offers the ability to do a one-off social media platform for individual companies, but is still too generic to reach the EDA customer base. I'm hoping that something like Xuropa becomes the "Facebook for EDA." that the industry needs. Oh, and a historical note: The cuts in advertising for electronics publications began in 1999 with the public decision of the EDA big three to halve their advertising budgets. That got the ball rolling. Within two years, Electronic News, which was already struggling, went all online. EDA has followed suit every year by cutting advertising. So we can't say EDA isn't a trend setter. Look what their marketing decisions started.

jvh3
User Rank
Rookie
re: EDA bashing
jvh3   2/3/2009 3:59:01 AM
NO RATINGS
Gabe: Thank you, thank you, thank you for writing this! To make my motivation for this praise clear: Myself and my colleagues are more than willing and able to take any criticism directed our way. In fact, we actively seek it out so we can improve our products, methodologies, engagement practices, operations -- whatever has been identified as being in need of attention. We don't always have the resources to act on feedback as soon as anyone would like; but nevertheless such feedback is much appreciated. However, as you correctly identified in your article, blanket "EDA bashing" is of no use to anyone as it needlessly displaces discussion of actionable issues and solutions. --- Regarding the economics point made by SDBrorson above; specifically: "And the open-source solutions give customers exactly what they want: Flexible design flow, control over their designs, and zero cost." This is only 2/3 true -- open source is not actually zero cost. There are very real costs when you go down the "make" route of the "make vs. buy" decision when you decide to use open source. I'm not saying open source is a bad thing -- far from it. I'm saying it's a fallacy to say open source is free. Furthermore, one way that EDA vendors really add value, i.e. how EDA vendors "compete with 'free'" and/or fight commoditization, is in the methodology guidance and support we provide our customers, in tandem with a willingness to assume some of their risk. To paraphrase a famous EDA-turned-semiconductor executive I used to work for (not Jack Harding), "Ostensibly we sell products, but the truth is that EDA is actually a services business where we must proactively work to make the customer successful or they will kick us out to the curb the first chance they get!" At Verisity, and now at Cadence, by taking these words to heart in both our initial engagements and our ongoing consultative partnerships, we have sustained our success despite the unrelenting competitive pressure from all quarters. Joe Hupcey III Cadence Direct Blog: http://www.cadence.com/Community/posts/jvh3.aspx Team Specman blog: http://www.cadence.com/community/posts/teamspecman.aspx

msanie
User Rank
Rookie
re: EDA bashing
msanie   2/2/2009 10:38:23 PM
NO RATINGS
Nice article -- you hit a lot of the points right on. I did want to comment on your first point on "traditional" advertising being dead. I have to say: it is! Traditional advertising is based on a message-broadcast model: company broadcasts message, and target audience hears it, and if repeated enough they'll act on it. That model does not work any longer, and same for the traditional communication channels. Marketing and advertising has shifted to an user-engagement model: company engages in direct communication with users, message is subtly passed from users to peers, and users/peers act on it. That is why companies (and even independent individuals are turning into blogs, forums, and professional networks). Social media is very powerful, and it certainly isn't only "social" and for kids. EMC and Cisco are among companies who are using social media effectively to increase information sharing, team building, and removal of silos within the companies. EDA marketing and advertising is not dead, but needs a new lifeline. It needs shift its methods around the user-engagement model. Synopsys and Cadence have already done a good job moving in that direction (still far from where they should be). Even this article, affording folks like me (the "users" of your message) with an opportunity to add comments, is a good form for user-engagement model. You mentioned yourself that only a "small portion" of your social network cares about EDA. Here's the answer: Xuropa (www.xuropa.com) is providing the right approach to take the first step. They have created a Web 2.0 platform which, among other things, includes a professional network for the electronic design folks ... who do care a greatly about EDA.

Page 1 / 2   >   >>
Flash Poll
Radio
LATEST ARCHIVED BROADCAST
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Top Comments of the Week