harry ... the ASIC guy

Honey, I Tattoo’ed The Kids

harry — Thu, 22 Oct 2009 16:56:59 +0000

I may be the engineer in the family, but my wife is the inventor.

Being an engineer is easy. You go to school to learn well established laws and methods. When you get a job, there are others who can mentor you and show you the ropes. If you need to know more, there are training classes to take. Someone else, usually marketing, decides what needs to be to be designed. All you have to do is follow the rules.

Being an inventor is hard. There is no school for inventing, nobody to teach you the ropes, and no classes to tell you how to do things. You come up with your own idea that has never been done before. You learn on the fly what you need to know from a variety of domains you know nothing about. You build it yourself. There is no such thing as first pass success. Experimentation and refinement are a constant process.

For those of you who know Joyce and have been privileged to receive one of her hand-made personalized holiday “greetings”, you know that she is one of the most creative and gifted people you could ever meet. I’ve had people tell me that they look forward to the holidays just to find out what she is going to send. Most Novembers, our garage looks like a scene out of Honey I Shrunk The Kids as Joyce invents a new holiday greeting. Each of these inventions takes countless hours of brainstorming, planning, experimenting, assembling, and tweaking to get it “just right”.

There was the mobile of our kids made with clear fishing line tied off. The fishing line, being so slick, would always untie until Joyce figured out to add a drop of glue to the end to keep it in place. Problem solved.

A few years ago Joyce made a lantern with pictures of our children on the outside. It took a lot of research and experimentation to make sure the cardboard comprising the lantern was sufficiently coated so as not to catch fire but still provide a cozy glow. Good to go.

And probably the most interesting of all, there was the snow globe which contained an acrylic encased photo of our daughter Kiara reaching for the stars. I remember Joyce’s frustration trying to pry the small Lucite blocks from their molds until she realized that sticking them in cold water would loosen the blocks from the molds due to the CTE mismatch between the materials. Then, these Lucite blocks were placed in baby food jars that had to have just the right mixture of water and propylene glycol to be viscous and not grow algae. Perfect.

About a year and a half ago, we were visiting Legoland in Carlsbad, CA. My kids (then 4 and 6), especially my son, have a gift for getting “sidetracked”. So, Joyce thought it would be a good idea for our kids to have our cell phone numbers conspicuously on them in case they got lost. That way, it would not depend on their memory of our cell numbers for us to be contacted by a helpful stranger if they wandered away. That’s when the idea for the “WhoTat” was born.

Joyce decided to start a business called Aware Gear that would provide safety products for young children. The first product was the WhoTat which is a temporary tattoo that is personalized with a phone number to call in case of emergency. Kids like to wear them and parents know that their kids don’t need to remember their phone numbers. Now, I’ll admit that I was personally dubious that this was needed. After all, how hard is it for a kid to remember a phone number? The problem is that most of us have several phone numbers (home, work, cell) and a scared lost child of any age can get confused and even mix the phone numbers together. That’s what my son did, mix the numbers together.

As with the other holiday greetings, this was an invention of its own. At first it seemed easy. Just buy these tattoo sheets that could be fed into a printer, design some neat tattoos, add the phone number, and print. But it wasn’t so easy. First there were moisture issues where the printing process wouldn’t work on humid days or the ink wouldn’t stick. Then the Tats were kind of sticky, so baby powder was used to remove the stickiness, but not too much so as to blur the image. It was probably 6 months before she got the process down to something that would repeatably produce a high quality product.

The second offering was a skin sticker where you could write in your own phone number or message on the fly. Again, material problems. Most recently, she is producing glow-in-the-dark skin stickers called gid-its. With Halloween coming up, and all the inherent dangers of visiting strangers while trick-or-treating, this should help kids stay safe. Once again, this has proven to be a real lesson in materials science, specifically how to get paint with glow-in-the-dark powder to adhere evenly to a non- porous acrylic surface. Who knew? Oh, and 2 days ago she told me about another invention idea and I already see the beginnings of prototypes around the house.

Since Halloween is coming around and child safety is a big concern for a lot of parents, I’d like to ask a favor … ummm … I mean offer you an opportunity. Actually, 3 opportunities.

First, as I mentioned, my wife is a creative genius and you have the opportunity to acquire some of there original work. Her products really do help to keep kids safe and are fun and you have the opportunity to get them for your kids. If you don’t have young kids, tell a friend who does. You can order them here. An if you use the checkout code ASICGUY10 you will get a 10% discount.

Second, you have an opportunity to support a good organization. In addition to the 10% discount mentioned above, Joyce will donate 10% directly to the National Center for Missing and Exploited Children. The NCMEC acts as an information clearinghouse and resource for parents, children, law enforcement agencies, schools, and communities to assist in locating missing children and to raise public awareness about ways to prevent child abduction, child sexual abuse and child pornography. John Walsh, Noreen Gosch, and others advocated establishing the center as a result of frustration stemming from a lack of resources and coordination between law enforcement and other government agencies.

Lastly, if you have gotten some value out of reading this blog over time, then here is your chance to thank me. I’d really appreciate it.

harry the ASIC guy

ShareThis

Synopsys Synphony Synopsis

harry — Mon, 12 Oct 2009 16:52:17 +0000

I was contacted a few weeks ago by Synopsys’ PR agency to see if I’d be interested in covering an upcoming product announcement. I usually ignore these “opportunities” since the information provided is usually carefully wordsmithed marketing gobbledygook and not enough for me to really form an opinion. However, it turned out that this announcement was on a subject I know a little bit about, so I took them up on their offer.

The announcement was “embargoed“, that is, I was not to make it public until today. Embargoes are a vestige of the days when traditional journalism ruled the roost and when PR departments thought they could control the timing of their message. I don’t think embargoes benefit companies anymore since news is reported at light speed (literally) and people will write what they want when they want. Still, I consider it a sort of gentleman’s agreement so I’m not writing about it until today.

I also waited a little bit until the “mainstream press” wrote their articles. That let’s me point you to the best of them and conserve the space here for my own views, rather that regurgitating the press release and nuts and bolts.

(Update: Here is a very good description of the Synphony flow from Ron Wilson).

Today, Synopsys announced a new product called Synphony High Level Synthesis. You can read about this here. Basically, Synopsys is introducing a high level synthesis (aka behavioral synthesis) product that takes as its input Matlab M-Code and produces RTL Code, a cycle accurate C-model, and a testbench for simulation. Since I have not used the tool, I cannot comment on the capabilities or the quality of results or compare it to other tools on the market. However, I have had some past experience with tools like Matlab (specifically SPW) and Synphony (specifically Behavioral Compiler). So, here are my thoughts, observations, opinions that come to mind.

Synopsys, once the leader in behavioral synthesis, is now the follower - When Synopsys introduced Behavioral Compiler over a decade ago they were the first to preach the gospel of high-level synthesis and all the associated benefits. Architectural optimization. Faster simulation. Bridging the gap between system design and ASIC design. Smaller and easier to understand code. Dogs and cats living together. The promises never fully materialized and Synopsys seemingly moved out of the market. Meanwhile, Mentor introduced Catapult C, Cadence introduced C-to-Silicon, and several others including Forte, Agility, Bluespec, Synfora, ChipVision, and AutoESL introduced their own high-level synthesis tools. Now, after acquiring Synplify DSP through Synplicity, Synopsys is finally re-entering the market (at least for ASIC design) with Synphony. The hunted have become the hunters.
Synphony takes M-code from Matlab as its only source - Whereas most (but not all) other high-level synthesis tools input C like languages, Synopsys has chosen to input M-code only, at least for now. According to Chris Eddington, who is Director of Product Marketing for System-Level Products at Synopsys (according to his LinkedIn profile), approximately 60% of those who say they do “high-level design” are using M-code or some form of C (ANSI C, C++, System-C) for some portion of their design activities. Of those, slightly more use the C variants than M-code, which means that somewhere close to 25% of all ASIC designers could be a possible market for this tool.
Synopsys can try to leverage the Matlab installed base - As mentioned above, Synopsys estimates that 25% of high-level designers could use the Synphony tool which is a pretty big market. By targeting mainly algorithmic design, not control logic, Synopsys can try to serve the Matlab installed base with a more narrowly targeted offering which should make it easier to support. It also allows Synopsys to avoid a bloody battle over C dominance and to pursue a blue ocean strategy with Matlab’s installed base. Interestingly though, there is no partnership with MathWorks implied by this announcement.
Synphony leverages existing IP libraries - Libraries already exist for many common functions that were available for the Synplify DSP tool. The library elements are available as well for Synphony, allowing the designer to specify his functionality using this library or using M-code as the source.
An FPGA tool is being adapted for ASIC - This is probably one of the first times that a tool initially developed for FPGAs (Synplify DSP) is being adapted for ASICs. It’s usually the other way around (e.g. FPGA Compiler grew out of Design Compiler). It should be interesting to see if the FPGA tool can “cut-it” in the ASIC world.
Ties to implementation are seemingly tenuous - A tool that can take M-code as its input and produce RTL and C and do all the other things is all fine and good. But for Synphony to become more than an experimentation tool, it has to produce results (speed, area, power) as good or better than hand-coded RTL. However, the ties to the implementation tool (Design Compiler) are not as direct as even Behavioral Compiler was a decade ago. It seems that Synphony takes an approach where it pre-compiles and estimates timing for various blocks (kind of like building DesignWare libraries), but it assembles the design outside of DesignCompiler without all the associated timing views and engines necessary for true design and timing closure. It’s hard to understand how this can reliably produce results that consistently meet timing, but perhaps there is something that I am not aware of?
Focus on “algorithmic design”, not control - As mentioned above, Synopsys is going after the folks using Matlab. And those designers are developing algorithms, not state machines. In essence, Synphony can focus on the fairly straightforward problem of scheduling mathematical operations to hit throughput and latency goals and not deal with more complex control logic. Much simpler.
Conversion from Floating Point to Fixed Point - Anyone who has designed a filter or any DSP function knows that the devil is in the details, specifically the details of fixed point bit width. One choice of bit width affects downstream choices. You have to decide whether to round or truncate and these decisions can introduce unexpected artifacts into your signal. Synphony converts the floating point Matlab model into a fixed point implementation. Supposedly, it then allows you to easily fiddle with the bit widths to tweak the performance. Some earlier Synopsys products did this (Cossap, System Studio) and it’s a nice feature that can save time. We’ll see how useful it really is over time.
Synphony produces real RTL, as well as C-code and a testbench - One of the drawbacks of Behavioral Compiler is that it never produced a human readable form of RTL code. This made it hard to simulate and debug the RTL. Synphony supplies readable RTL (or so I am told) as well as cycle accurate C-code for system simulation and a testbench for block simulation. This should help facilitate full chip simulations for chip integration, since Synphony will probably only be used on blocks, not entire chips.
Couldn’t Synopsys come up with a better reference than Toyon Research Corporation - No offense to Toyon, but they are hardly a household name. It makes me wonder how many partners Synopsys has engaged in this development and how well tested this flow is. Not saying it isn’t well tested, just that Synopsys is making me wonder. Gimme a name I’ve heard of, please.

Only time will tell if Synphony is truly music to our ears, or if it is just SYNthesis that is PHONY.

harry the ASIC guy

ShareThis

Writing a Fan a $439 Personal Check for a Bad Game - Priceless

harry — Fri, 25 Sep 2009 05:27:27 +0000

To hear long time fan Tony Seminary tell the story, he was embarrassed by the Oregon Ducks performance when they lost to Boise State on the opening night of the 2009 college football season. Not only did they play a sloppy game, gaining only 152 total yards, making only 6 first downs, and committing 2 turnovers, but the whole nation got to see one of their star players punch out a Boise State player on the field after the game.

So what did Tony do? He wrote an email to new Ducks head coach Chip Kelly asking for him to reimburse him for the cost of his round trip from Portland to Boise.

So what did Chip Kelly do? He wrote an email back asking “what’s your address?” A few days later, Tony received a personal check from Chip Kelly for the $439 that Tony had requested.

This happened a few weeks ago, but it just hit the news this week and has gone, dare I say the word, ‘viral’. Among the results of this are the following:

Customer Loyalty - Tony Seminary sent the check back and has been quoted as saying “I think of Coach Kelly as a totally different person now, I have a different bond with him now thanks to what happened. Let’s just say he lost every game as an Oregon coach. You would never hear me calling for his head. It just wouldn’t happen. The guy showed an incredible amount of class”.
Team Loyalty - Said Seminary, “I now know why his kids would run through a wall for that guy, because who does what he did, right? That is simply amazing.”
Personal Reputation - In blogs and articles all over, Chip Kelly is being hailed not only for doing what he did, but for doing it quietly without drawing attention. As one blog said “Chip Kelly, a man of his word.”
School Reputation - I don’t have any evidence of this as yet, it’s too early, but certainly some of this will rub off on the University of Oregon, in a good way.

The good news was that Oregon had put in charge a man with integrity and they gave him the freedom to respond as he saw fit. But how would most schools and companies have handled something like this? It would have probably gone something like this:

Coach receives email and forwards it to the legal department.
Lawyers craft a carefully worded response indicating that Oregon regrets the loss but it is not responsible for incidental damages according to the relevant terms and conditions on the ticket that Mr. Seminary tacitly agreed to and should have read.
Mr. Seminary vows never to go to another Ducks football game again. He then goes online and tells hundreds of his friends who are Ducks fans the story.
One of his friends writes a Oregon Ducks blog and posts the story and the text of the Oregon response email. ESPN picks it up and shows it on Sportscenter.
Hundreds of Ducks fans come to the next game with signs saying something like “Win, or give me my money back”.
Top recruit hears the story and decides that he’d rather not go to Oregon. Chooses USC instead.
Athletic Director resigns.

No matter what business you are in, hire good people with good judgment and give them the freedom to make the customer happy. That kind of service is “priceless”.

harry the ASIC guy

ShareThis

I’m (Not) an IBMer Anymore

harry — Thu, 17 Sep 2009 07:38:04 +0000

I came across a tweet the other night that pointed me to a discussion on the EE Times forum regarding an editorial by Mark LaPedus a few weeks ago. The editorial states that “IBM Corp. has cut nearly 10,000 jobs this year, according to reports, although Big Blue still refuses to fess up to most of the layoffs.” Although IBM denies the reports and claims they are fabrications by union officials, the editorial adds fuel to the fire by stating that “for some time, the union has charged that IBM is cutting and outsourcing U.S. jobs, while quietly hiring in India.”

The comments in the discussion thread are very interesting and worth reading through. If you don’t have time, here’s a synopsis of the various opinions offered there:

We are in a jobless recovery and IBM is not the only company moving jobs offshore.
IBM is a corporation and their only obligation is to their shareholders. Outsourcing to India makes business sense so that is what they need to do.
“The sole function of a union is to keep their own jobs and breed a sense of entitlement and proliferate mediocrity.”
If it’s in the national interest to keep jobs and fabs here, then public funds should fund them.
Instead of blaming IBM, consider the high taxes and burdensome laws that favor outsourcing over keeping jobs in the US.
If you think IBM is doing the right thing, just wait till it’s your job being outsourced.
If this kind of off-shoring continues, we will wipe out the technological advantage we have in the US and you’ll need to move to another country to get a job.
Outsourcing “delivers worse results at lower costs”, and that’s what businesses want right now.
We, as consumers, are more concerned about the low prices we get at Walmart than the notion of “social justice.”
Instead of laying people off, cut out the big executive bonuses and perks.
Companies need to take care of all three groups - employees, customers, and shareholders. Otherwise employees can quit, customers can stop buying, and shareholders can sell.
Do companies owe anything to the communities where they are located or to the nations where they are headquartered?
Yes, they owe taxes!

Personally, I can see both sides, but what do you think? I’m especially interested to hear from any company execs who have decided against outsourcing and why that is.

harry the ASIC guy

ShareThis

The Accidental Blogger

harry — Fri, 11 Sep 2009 06:05:39 +0000

As a kid, I always dreamed of being interviewed after hitting the game winning homerun or jump shot or throwing the game winning touchdown pass. Well, at this point in my life, the likelihood of those dreams is pretty much zilch. But, fortunately, I’ve been able to achieve something almost as great. A one-hour interview on Dee McCrorey’s Big Dreamers! The Reinvent Success Show.

So, after a full weekend of watching college football and then NFL football, and listening to those other guys getting interviewed after the game winning touchdowns, you can sit down at 6pm PDT, and unwind as Dee McCrorey, Risk Guru, Innovation Catalyst, and Business Coach asks me about my career from no-name engineer to “Harry the ASIC Guy”. You can always listen to the recording afterward if the time is inconvenient, but if you tune in live you can actually call into the show and ask questions, make comments, remind me of the $10 I borrowed for lunch and never gave back, whatever.

Honestly, I’m both flattered and embarrassed to have this opportunity. I met Dee just this past July at DAC in the Synopsys Conversation Central booth and we hit it off right away. Even after the sessions were over, Dee stayed and continued to ask questions trying to dig deeper and get at the core the topics we were discussing. She really has a desire to get to the essence of things which is a great asset for an interviewer, so I’m looking forward to some tough questions. She also has a thriving consulting business helping professionals reinvent their careers, both within corporations and individually, so I’m looking forward to working with her professionally as well.

For more information on the show, you can go here. I hope you can join me.

harry the ASIC guy

ShareThis

Two Blog or Not Two Blog?

harry — Tue, 08 Sep 2009 04:44:08 +0000

I got an email last week from one the readers of this blog that observed “it would be interesting to learn how to manage both blogs while doing justice to your readers.” He was of course referring to my new blog on Xuropa that I write in addition to this one.

Indeed, this was a concern of mine that I had considered carefully before embarking on the other blog … or so I thought. The other day I wrote a new blog post about how designers want to actually use tools hands-on rather than just listen to product marketing pitches, or webinars, or podcasts. I originally wrote the post for this blog, then decided that it made more sense for the Xuropa blog, and ended up publishing it there (here’s the link). But it could have really gone on either one with small adjustments. I can see that this is now going to be more difficult than I thought.I did a little research online to see how other bloggers are handling writing multiple blogs. One of the suggestions was to set down the objectives of each of the blogs so I could be more clear in my mind and to the readers. I think that’s a good idea. So here goes:

The Xuropa blog will be focused on ways that EDA companies can do more with less, like cloud computing, online tool access, and software-as-a-service. It will also be written for an audience of EDA sales and marketing professionals. If you are in EDA, you’ll want to subscribe to that blog.
The harry the ASIC guy blog will include lots of other content and is hopefully valuable for people in all aspects of the semiconductor industry. I’ll discuss general engineering trends, quarterly reports from EDA companies, technical topics, and industry news. If you are a designer, you’ll want to subscribe to this blog.

I’m guessing that many of you will be interested in both topic areas and so it is OK to subscribe to the Xuropa blog and subscribe to this blog. You have my permission. After some time you may find that you are only interested in one of the blogs. That’s OK too, just unsubscribe to the one that doesn’t meet your needs.

Another suggestion was to set realistic expectations for how frequently I’d be publishing a new post. I think that is a good idea as well. I will continue to post on this blog roughly once per week as I have in the past. For some time I was actually closer to 2 posts per week but I have fallen back to once a week and that is about what I can handle now. The other blog is shared with some other folks from Xuropa so I will probably publish there every other week. We’ll see how that goes.

I’d like to ask you each a favor as well. Please help me keep to my commitment. I’ve already made this commitment of public record here, so that alone will provide some pressure. But if I start to post too infrequently or the quality slips or goes off track, let me know. Leave a comment or send me an email.

I would also like to make this blog a little more fresh and collaborative. I’ve said in the past that I learn more from you folks than you learn from me. You are working in hundreds of companies with thousands of years of collective experience. I’d like to see if we can tap into that for all our benefit. So here’s the deal:

If you have an idea for a blog post, let me know. Leave it as a comment or send me an email. I’ll make sure I give you full credit (unless you want to be anonymous) and link back to your website or LinkedIn profile.
If you’d like to write a guest blog post, I’m open to that as well. The more viewpoints the better.

Of course, not every suggestion will be used and not every offer of a guest blog post will be accepted. I’ll still make that decision to make sure the content is of high quality. But I won’t censor anything just because I disagree.

Well, I guess that’s it. We’re going to try this 2 blog thing and see how it goes. Wish me luck.

harry the ASIC guy

ShareThis

Who Ya Gonna Trust?

harry — Thu, 27 Aug 2009 15:03:11 +0000

Our summer has been pretty hectic and full of uncertainty, so we put off planning a short vacation until just this past weekend. We usually go up to Big Bear and stay at this one place that is dog friendly and has a pool for the kids and is close to town. We’ve stayed there 3 times before and have always been very happy.

This past Saturday morning, I Googled the name of the resort in order to get the web address when I noticed that there was a Trip Advisor listing for the place. So, I thought I’d check it out. Much to my surprise there was a slew of negative reviews. I dug a little further and found that many of these reviews were placed around the same date (since we had been there last) by people who only rated this one place and who had very similar complaints. These reviews seemed suspicious, but who knew, maybe some had merit. These could be legit or they could be someone posting them on the behalf of competing resorts to discredit their competitor.

As I surfed a little more, I found comments on some other pages indicating that this sort of negative posting on rating sites had become epidemic for Big Bear. Who knew that the lodging industry in this cozy little town in the mountains was so cutthroat? It’s a good example of a lose-lose strategy. Now I can’t trust any of the ratings!

In the end we ended up booking at a different resort, mostly due to other factors, but admittedly also due in part to the FUD (fear, uncertainty, and doubt) caused by these reviews.

On Sunday morning I came across an article that describes how one PR firm allegedly hires interns “to trawl iTunes and other community forums posing as real users, and has them write positive reviews for their client’s applications.” Now, I knew that this sort of mischief happened, but I thought it was all amateurish behavior on the part of overzealous business owners and their fans. I did not realize it was an actual service one could select from a PR firm. How brazen!

On the other hand, maybe this article I read was actually secretly sponsored by a competing PR firm in order to discredit the PR firm being decried in the article. Who is to believe whom? Hmmm…..

Before you say that I am naïve about all this behavior, I’m not. The verification methodology survey I posted back in February was vandalized by VMM and OVM fans. And more than a year ago, someone copied a blog post of mine onto comp.lang.verilog for the sole purpose of posting in response a personal attack on my credibility. I’ve seen this stuff first hand.

A big part of the problem is anonymity and impunity. When someone uses a fictitious name and email address to post such a review as the one described above, we never know who that person is and he never suffers any consequences. After all, who is Vactioner287 after all? However, let’s say that one could only leave a comment by using his LinkedIn profile. I bet that would kill 99% of the issues right there.

(Actually, it would probably result in a proliferation of fictitious LinkedIn accounts, but then you could tell pretty well from those accounts that they are fakes since they’d be very bare. To some extent, like metastability, you can never totally get rid of the problem … you can only make it less likely.)

Most websites that accept reviews require registration. Although the hassle of registration deters some legitimate people from leaving legitimate comments, it also beneficially deters those with malicious intentions to a great degree. Almost all the online communities in EDA require some sort of registration, the Synopsys blogs being the only one that I can name that does not.

So, who ya gonna trust?

Personally, there are 3 types of people who I trust on the internet and they are as follows:

People I already know and trust - These are people who I know personally. Maybe they are current or former colleagues or customers or suppliers or partners or friends. I have reason to trust them because I know them.
People I’ve come to trust - These are people whom I have come to know through the internet who have demonstrated over a period of time that they are trustworthy. Maybe it’s a blogger who has proven to be right most of the time. Or whose advice rings true. Or who provides me with valuable information and insight. Hopefully, I am one of those people for you.
People I’ve been told to trust by others I trust - This is where social capital and influence come into play. If someone I trust links to someone else, then I gain trust in that person to whom he is linking. If he’s on his blogroll. If he’s a guest blogger. If he’s written a book that is referred to. Not that everyone that is referenced is automatically trustworthy, but it helps.

If you were to look at my Google Reader and see who I subscribe to, they pretty much fall into the 3 categories above. That gives me plenty to read.

Unfortunately, this doesn’t help too much with the situation I originally described, because Vacationer287 doesn’t fall any of these categories. What do you do then? Ask yourself the following:

Did he write anything else under this name or did he just join to post this one review. If the former, then he may be legit (you need to look at what they wrote). If the former, that’s suspicious.
Did he use a real name? Vandals often hide behind fictitious and non-descript names.
Does it pass the smell test? I can smell bad milk without a lab test and you can too. Does it all make sense or does some of the writeup just seem too good or bad to be true?

I don’t know if this post helps you or confuses you more. Probably, it confuses you because now you have to consider why and how you come to trust some people and not others on the internet. That’s good. From reconciling confusion comes understanding.

Trust me, you have my word on it.

harry the ASIC guy

ShareThis

The Road Not Taken

harry — Fri, 21 Aug 2009 06:26:04 +0000

I’d like to offer you the opportunity to help someone out who needs to make a key decision in her life.

As I’ve written about and spoken about recently, the economic woes of the past year have impacted many of my peers and I’m sure yours as well. Especially hard hit seem to be those in the middle of their careers, a group that I count myself a part of. For those of us who have faced or are facing these uncertainties, I think it’s only natural to second guess the key decisions we made in our careers and wonder if we made the right choices. Some may have decided to take a chance on a new opportunity only to have it evaporate. Others may have passed on that opportunity only to see their current “safe” position turn out not to be so safe after all.

It’s with this in mind that I received an email from a young woman at a crossroads in her career, having to make just such a decision, one that she prays she will cherish but fears she will regret. She has the opportunity to move from her current “safe” position of many years to a new opportunity filled with uncertainty. In order to afford her the best possible insight and advice, I’d like to open this up to you (with her permission) since you all collectively have a ton more experience than I will ever have.

As you read her email, you’ll realize that she is facing several smaller decisions as part of this one big decision, namely:

ASIC vs. FPGA
ASIC Design vs. IP development
Existing company that she knows vs. a new company that she has to learn
Comfort zone vs. Temporary Incompetence
Hands-on Technical Work vs. Management
Expert vs. Generalist

Each of these decisions could be the basis for a debate on its own. Feel free to comment on any of these or all of these or on other aspects that you find important. If you can take some time to respond, I think this will not only serve to advise this woman, but will also be a great guide to anyone looking to make a career change.

__________

I am a Lead ASIC Designer with 13 years of experience in front end ASIC design and have worked on multiple ASICs to this date at a company in India. Everything is fine here, just that the work is getting very repetitive lately. I have an offer from a IP development firm and need to decide soon. The following things come to my mind when I think about the offer:

1. The work would be mostly on FPGAs (no ASICs involved).

2. I won’t work with the Physical Design guys anymore.

3. I may get good exposure on different networking IPs.

4. I am currently leading a sizable design in a big ASIC. Though this position is glamorous and coveted by many, there is nothing new to learn since I have been doing it for the past several years.

I have the following queries,

If I join the new company and start working on FPGAs, will it take away something from me, e.g., my “ASIC Gal” tag?
Will taking up the manageress role and doing project management ‘formally’ be better that working as a Lead Engineer, from a long term employability perspective? or will it be detrimental?
Will it be a one way path with little chances to come back to ASICs without a compromise? (after, say, 4-5 years).
I want to move towards system design/architecture in the future and am thinking that the more IPs I work on, the better it will be for me. Is this assumption correct?
Overall, any other advice as to what I should consider and whether I should take this position.

I would appreciate your reply.

ShareThis

DAC Theme #3 - “Increasing Clouds Over SF Bay”

harry — Mon, 17 Aug 2009 03:48:29 +0000

It was easy to spot the big theme’s at DAC this year. This was the “Year of ESL” (again). The state of the economy and the future of EDA was a constant backdrop. Analog design was finally more than just Cadence Virtuoso. And social media challenged traditional media.

It was harder to spot the themes that were not front and center, that were not spotlighted by the industry beacons, that were not reported by press or bloggers. Still, there were important developments if you looked in the right places and noticed what was changing. At least one of those themes came across to me loud and clear. This was the year that the clouds started forming over EDA.

If you’ve read my blog for a while, you know I’m not talking about the weather or some metaphor for the health of the EDA industry. You know I am talking about cloud computing, which moved from crazy idea of deluded bloggers to solidly in the early adopter category. Though this technology is still “left of chasm”, many companies were talking about sticking their toes in the waters of cloud computing and some even had specific plans to jump in. Of note:

Univa UD - Offering a “hybrid cloud” approach to combine on premise hardware and public cloud resources. Many view this as the first step into the cloud since it is incremental to existing on premise hardware.
Imera Systems - Offering a product called EDA Remote Debug that enables an EDA company to place a debug version of their software on a customer’s site in order to debug a tool issue. This reduces the need to send an AE on site or to have the customer package up a testcase.
R Systems - A spinoff from the National Center for Supercomputing Applications (best known for Telnet and Mosaic), they were wandering the floor pitching their own high performance computing resources (that they steadfastly insisted were “not a cloud”) available remotely or brought to your site to increase your computing capacity.
Cadence - One of the first (after PDTi) to have an official Hosted Design Solutions offering, they host their software and your data in a secure datacenter and are looking at the cloud as well for the future.

And then there’s Xuropa.

Before I cover Xuropa, I need to take a brief digression. You see, July 27th was not just the first day of DAC. It was also my first official day working for Xuropa as one of my clients. I’ll be doing social media consulting (blogging, tweeting, other online social community stuff) and also helping their customers get their tools on the Xuropa platform. This is very exciting for me, something I’ll blog about specifically on the Xuropa Blog and also here. In the meantime, under full disclosure, you’ve now been told. You can factor in the appropriate amount of skepticism to what I have to say about cloud computing, hosted design, Software-as-a-Service and Xuropa.

Xuropa - Offering to EDA companies and IP providers the ability to create secure online labs in the cloud for current and prospective customers to test drive a tool, do tool training, etc. They also have plans to make the tools available for “real work”.

These companies and technologies are very exciting on their own. Still, the cloud computing market is very new and there is a lot of churn so it is very difficult to know what will survive or become the standard. Perhaps something not even on this list will emerge.

Even though the technology side is cloudy (pun intended), the factors driving companies to consider using the cloud are very clear. They all seem to come down to one economic requirement. Doing more with less. Whenever I speak to people about cloud computing (and I do that a lot) they always seem to “get it” when I speak in terms of doing more with less. Here are some examples:

I spoke to an IT person from a large fabless semiconductor company that is looking at cloud computing as a way to access more IT resources with less of an on premise hardware datacenter.
Cadence told me that their Hosted Design Solutions are specifically targeted at smaller companies that want to be able to access a complete EDA design environment (hardware, software, IT resources) without making any long-term commitment to the infrastructure.
EDA and IP companies of all sizes are looking to reduce the cost of customer support while providing more immediate and accessible service.
EDA and IP companies are looking to go global (e.g. US companies into Europe and Asia) without hiring a full on sales and support team.
Everyone is trying to reduce their travel budgets.

Naysayers point out that we’ve seen this trend before. EDA companies tried to put their tools in datacenters. There were Application Service Providers trying to sell Software-as-a-Service. These attempts failed or the companies moved into other offerings. And so they ask (rightly) “what is different now?”

There is certainly a lot of new technology (as you see above) that help to make this all more secure and convenient than it was in the past. We live in a time of cheap computing and storage and ubiquitous internet access which makes this all so much more affordable and accessible than before. And huge low cost commodity hardware data centers like those at Amazon and Google never existed before now. But just because all this technology exists so that it can be done, doesn’t mean it will be done.

What is different is the economic imperative to do more with less. That is why this will happen. If cloud computing did not exist, we’d have to invent it.

harry the ASIC guy

ShareThis

DAC Theme #2 - “Oasys Frappe”

harry — Mon, 10 Aug 2009 07:21:46 +0000

Sean Murphy has the best one sentence description of DAC that I have ever read:

The emotional ambience at DAC is what you get when you pour the excitement of a high school science fair, the sense of the recurring wheel of life from the movie Groundhog Day, and the auld lang syne of a high school re-union, and hit frappe.

That perfectly describes my visit with Oasys Design Systems at DAC.

Auld Lang Syne

When I joined Synopsys in June of 1992, the company had already gone public, but still felt like a startup. Logic synthesis was going mainstream, challenging schematic entry for market dominance. ASICs (they were actually called gate arrays back then) were heading towards 50K gates capacity using 0.35 uM technology. And we were aiming to change the world by knocking off Joe Costello’s Cadence as the #1 EDA company.

As I walked through the Oasys booth at DAC, I recognized familiar faces. A former Synopsys sales manager, now a sales consultant for Oasys. A former Synopsys AE, now managing business development for Oasys. And not to be forgotten, Joe Costello, ever the Synopsys nemesis, now an Oasys board member. Even the company’s tag line “the chip synthesis company” is a takeoff on Synopsys’ original tag line “the synthesis company”. It seemed like 1992 all over again … only 17 years later.

Groundhog Day

In the movie Groundhog Day, Bill Murray portrays Phil, a smug, self-centered, yet popular TV reporter who is consigned by the spirits of Groundhog Day to relive Feb 2nd over and over. After many tries, Phil is finally able to live a “perfect day” that pleases the spirits and he is able to move on, as a better person, to Feb 3rd.

As I mentioned in a previous post, I’ve seen this movie before. In the synthesis market, there was Autologic on Groundhog Day #1. Then Ambit on Groundhod Day #2. Then Get2chip on Groundhod Day #3. Compass had a synthesis tool in there somewhere as well. (I’m sure Paul McLellan could tell me when that was.) None of these tools, some of which had significant initial performance advantages, were able to knock off Design Compiler as market leader. This Groundhog Day it’s Oasys’ turn. Will this be the day they finally “get it right”?

Science Fair

A good science fair project is part technology and part showmanship. Oasys had the showmanship with a pre-recorded 7-minute rock medley featuring “Bass ‘n’ Vocal Monster” Joe Costello, Sanjiv “Tropic Thunder” Kaul, and Paul “Van Halen” Besouw. Does anyone know if this has been posted on Youtube yet?

On the technology side, I had one main mission at the Oasys booth … to find out enough about the RealTime Designer product to make my own judgment whether it was “too good to be true”. In order to do this, I needed to get a better explanation of the algorithms working on “under-the-hood”, which I was able to get from founder Paul van Besouw.

For the demo, Paul ran on a Dell laptop with a 2.2 GHz Core Duo processor, although he claims that only 1 CPU was used. The demo design was a 1.6M instance design based on multiple instantiations of the open source Sparc T1 processor. The target technology was the open source 45nm Nangate library. Parts of the design flow ran in real time as we spoke about the tool, but unfortunately we did not run through the entire chip synthesis on his laptop in the 30 minutes I was there, so I cannot confirm the actual performance of the tool. Bummer.

Paul did describe, though, in some detail, the methods that enable their tool to achieve such fast turnaround time and high capacity. For some context, you need to go back in time to the origins and evolution of logic synthesis.

At 0.35 uM, gate delays were 80%+ of the path delay and the relatively small wire delays could be estimated accurately enough using statistical wire load models. At 0.25 uM, wire delays grew as a percentage of the path delay. The Synopsys Floorplan Manager tool allowed front-end designers to create custom wire load models from an initial floorplan. This helped maintain some accuracy for a while, but eventually was also too inaccurate. At 180 nM and 130 nM, Physical Compiler (now part of IC Compiler) came along to do actual cell placement and estimate wire lengths based on a global route. At 90 nM and 65 nM came DC-Topographic and DC-Graphical, further addressing the issues of wire delay accuracy and also layout congestion.

These approaches seem to work well, but certain drawbacks are starting to appear:

Much of the initial logic optimization takes place prior to placement, so the real delays (now heavily dependent on placement) are not available yet.
The capacity is limited because the logic optimization problem scales faster than order(n). Although Synopsys has come out with methods to address the turnaround time issue, such as automatic chip synthesis, these approaches amount to not much more than divide and conquer (i.e.budget and compile).
The placement developed by the front-end synthesis tool (e.g. DC-Topographic) is not passed on to the place and route tool. As a result, once you place the design again in the place and route tool, the timing has changed.

According to Paul van Besouw, Oasys decided to take an approach they call “place first”. That is, rather than spend a lot of cycles in logic optimization before even getting to placement, they do an initial placement of the design as soon as possible so they are working with real interconnect delays from the start. Because of this approach, RealTime Designer can get to meaningful optimizations almost immediately in the first stage of optimization.

A second key strategy according to van Besouw is the RTL partitioning which chops the design up into RTL blocks that are floorplaned and placed on the chip. The partitions are fluid, sometimes splitting apart, sometimes merging with other partitions during the optimization process as the design demands. The RTL can be revisited and changed for a new structure during the optimization as well. Since the RTL partitions are higher-level than gates, the number of design objects in much fewer, leading to faster runtime with lower memory foot print according to van Besouw. Exactly how Oasys does the RTL partitioning and optimizations is the “secret sauce”, so don’t expect to hear a lot of detail.

Besides this initial RTL optimization and placement, there are 2 more phases of synthesis in which the design is further optimized and refined to a legal placement. That final placement can be taken into any place and route tool and give you better results than the starting point netlist from another tool, says van Besouw.

In summary, Oasys claims that they achieve faster turnaround time and higher capacity by using a higher level of abstraction (RTL vs. gate). They claim that they can achieve a better starting point for and timing correlation with place and route because they use actual placement from the start and feed that placement on to the place and route tool. And the better placement also runs faster because it converges faster.

What Does Harry Think?

Given the description that I got from Oasys at DAC, I am now convinced that it is “plausible” that Oasys can do what they claim. Although gory detail is still missing, the technical approach described above sounds exactly right, almost obvious when you think about it. Add to that the advantage of starting from scratch with modern coding languages and methods and not being tied to a 20 year old code base, and you can achieve quite a bit of improvement.

However, until I see the actual tool running for myself in a neutral environment on a variety of designs and able to demonstrate faster timing closure through the place and route flow, I remain a skeptic. I’m not saying it is not real, just that I need to see it.

There are several pieces of the solution that were not addressed adequately, in my opinion:

Clock tree synthesis - How can you claim to have a netlist and placement optimized to meet timing until you have a clock tree with its unique slew and skew. CTS is not address in this solution. (To be fair, it’s not addressed directly in Design Compiler either).
A robust interface to the backend - Oasys has no backend tools in-house, which means that the work they have done integrating with 3rd party place and route has been at customer sites, either by them or by the customer. How robust could those flows be unless they have the tools in-house (and join the respective partner programs).
Bells and whistles - RealTime designer can support multi-voltage, but not multi-mode optimization. Support for low power design is not complete. What about UPF? CPF? All of these are important in a real flow and it is not clear what support Oasys has.
Tapeouts - This is probably the key question. For as long as EDA has existed, tapeouts have been the gold standards by which to evaluate a tool and its adoption. When I asked Paul if there are any tapeouts to date, he said “probably”. That seems odd to me. He should know.

However, if Oasys can address these issues, this might actually be the game changer that gets us out of the Groundhog Day rut and onto a new day.

harry the ASIC guy

ShareThis