harry … the ASIC guy

Every year on March 19th, the swallows wing their way back to San Juan Capistrano. Just up the road in Anaheim, designers from around the world will fly in for the 45th Annual Design Automation Conference, held June 8th - 13th. How appropriate will it be then, when EDA and ASIC design bloggers flock to the 1st annual DAC Birds-of-a-Feather session on blogging?

Perhaps you are a blogger or are thinking of becoming a blogger or know somebody who is a blogger. Perhaps you are a marketing director or just curious. Whatever your interest, you’ll want to come meet and engage with the bloggers who are growing in quantity, quality and industry influence:

• Quantity: A quick check of the DAC home page shows 17 industry blogs, and there are at least a dozen more, including yours truly.
• Quality: Topics as diverse as verification, design for test, analog design, semiconductors, memory, good design techniques, and standards, are discussed by design experts who actually use the tools and do the work.
• Influence: These blogs have influenced standardization efforts and challenged the status quo.

This event will be held in Rooms 201B and 201C at the Anaheim convention center on Wednesday, June 11 at 6pm.

I am helping to coordinate this session, so if you are planning to attend, just drop a quick email to harry {at} theASICguy {dot} com so we can get an idea for how large a group we will have. If you are a blogger and would like to present or be part of a panel, please let me know as well.

I hope to see and meet many of you there.

harry the ASIC guy

On Wed April 30th Hewlett Packard announced that they had fabricated a device previously only theorized. Known as a memristor, this is the 4th basic type of passive circuit element, joining its brethren the resistor, capacitor, and inductor. The device’s unique property of “memristance” is equal to the rate of chance of flux with respect to charge, hence it has been commonly compared to the mythical “flux capacitor”, popularized in the original Back to The Future movie. Now, in order for us to achieve time travel, someone simply needs to invent the Mr. Fusion Home Energy Reactor to generate the 1.21 GigaWatts needed to power the memristor. Anybody …. Anybody … Anybody …

The memristor is said to be absolutely unique because it is the only one of the basic circuit elements to exhibit the property of memory. According to UC Berkeley Professor Leon Chua, who first postulated the existence of the memristor in 1971, “the memristor is our salvation, because it works better and better as you make it smaller and smaller. The era of nanoscale electronics will be enabled by the memristor. This is not just an invention, it is a basic scientific discovery. It has always been there — we just had to face these nanoscale problems to realize its importance.”

The applications being described include ultra-dense (100Gbit) non-volatile memory, ultra-high-density crossbar switches, and brain-like neural networks. One comment on a blog site this week from a John Conner even said “This is the beginning of SkyNet. Soon the T101 model will be developed, based on this design.” And many new applications will emerge once that the technology achieves production quality.

In a word, Wow!

Personally, I’m not at all familiar with the science behind this discovery, so I decided to do some research of my own. Since this was such an important discovery, where would I go? Wikipedia, of course.

By the end of the next day after the discovery was announced, the Wikipedia article for Memristor has been updated 110 separate times by 53 different authors. Here is the article before the announcement, and here is the article at the end of the next day. This is the 1.21 GigaWatt power of Wikipedia .. harnessing the efforts of scores of volunteers worldwide.

To be fair, I decided to try to search for “memristor” in the other three online encyclopedias.

• The time honored Britannica.com returned “sorry, we were unable to find results for your search”.
• Encyclopedia.com (aka Columbia Encyclopedia) returned a blank page except for Google Adsense ads for 3 art sites, including one for Dog Art (hmmmm)
• Encarta returned “No results were found for your search in Encarta. Did you spell your search words correctly?”

Anybody …. Anybody … Anybody …

Like Ferris Bueller, I guess they were taking the day off.

harry the ASIC guy

Last weekend I got to do something I had not done for a long time.

Go to camp!

The local YMCA Adventure Guides organized a weekend camping trip up to a place called Camp Whittle outside of Big Bear Lake, CA. So Joyce and I packed up the kids (actually, Joyce did most of the packing), dropped Mookie off at the kennel, and drove the 3 hours to the camp site on Friday afternoon. I expected a very rustic non-heated cabin with outdoor “facilities”, but it turns out they had remodeled and the cabins were more like small cottages with central heat and indoor plumbing. Things were looking good already

At 5:30 AM on Saturday morning I was awakened by the voices of several 6-8 year olds.

“I saw my dad’s private parts”.

“I saw my mom’s”.

“I saw my sister’s”.

OK…time to get up. We had split into boys and girls groups, so I got to spend time with our 4-year old son Nate while Joyce got Kiara, our 6-year old first grader. There were planned activities, but campers were also free to do anything they wanted. Here’s just a little bit of what we did all in one day…

petting zoo … breakfast … archery … arts and crafts … foosball … ping-pong … basketball … lunch … built a dam on a muddy stream out of branches … hay ride … climbing wall … baseball … football … dinner … skits … pudding

Phew!!! By this time it was 8:30 and we were all very tired, so we started heading back to our cabins to collapse.

“Snipe Hunt!!!”

About 30 screaming kids and their exhausted parents gathered in the open area between the cabins while one of the Adventure Guide leaders described the lore and rules of the snipe hunt.

“You gotta be real quiet … the snipe like to hide in the bushes … when you see one, yell out and everyone shine your light on him … look for a small yellow and purple furry animal about the size of a small dog … there are two types, the long-tailed snipe and the short-tailed snipe … we almost caught one last year, but it got away”.

Armed with flashlights, we trekked out into the woods in search of the elusive snipe. “There’s one”, shouted one of the parents, pointing furiously towards a bush. The kids swarmed to the bush like bees on a hive, shining their flashlights as they tripped along. “Darn, he got away”.

“I got one here”, yelled a father. Again, the kids swarmed over to catch the snipe, but he got away again.

This frenzy continued for about half an hour (and we didn’t even lose a single child in the pitch black). Several of the campers saw the snipe and even touched it as it ran by, including Kiara and her new friend CJ. We came teasingly close, but we never did catch the snipe. “I’ve been doing this for 28 years, and this is the closest we’ve come”, said Chad, the veteran leader of the snipe hunt. “You guys are the best snipe hunters I’ve had the pleasure of snipe hunting with”.

Sunday morning nobody got up until about 7:30.

“Ethan farted”.

“No I didn’t”.

After breakfast we gathered together as a group and they had the parents go around and tell what they thought of the weekend. When it came to me, I recalled how much this reminded me of when I was a small kid at camp growing up. All the simple innocent games and how much fun it was just to be in the fresh air and make new friends and have all this unstructured time.

These days, our kids’ time is totally structured. A friend in Silicon Valley, Carolann, who is a 3rd grade teacher, told me that kids in her class go to Kumon tutoring right after school. They take classes all summer. She told me of one instance where a little girl came to her crying because another boy told her, “you’re grammar is bad so you’re not going to get into a good college”. Whatever happened to childhood?

At least for one weekend … we got a little of that childhood back.

Please tell me about your camp memories……..

harry the ASIC guy

Last Saturday night I went to a family Seder to celebrate the first night of Passover. You know, like in The Ten Commandments with Charlton Heston. As part of the Seder, we read a story of 4 sons, one wise, one contrary, one simple, and one unable to ask a question.

This got me thinking about some of the contrary ASIC designers I’ve worked with through the years … you know the type:

1. If everyone else wants to take road A, he wants to take road B.
2. If everyone else wants to take road B, he wants to take road C.
3. If you’ve got a plan, he’ll tell you why it won’t work.
4. Once he takes a stand on an issue, he’ll never give up.
5. He doesn’t really care what others think about him.
6. Every battle is worth fighting … to the death.

The contrarian ASIC designer can sap the energy and optimism out of a design team with all his negativity. Obviously, not good. So, why would anyone want to work with a contrarian?

Well, I’m here to tell you that the contrarian gets a bad rap and he can be a critical member of the team. First, some background…

Most law schools use a method of contrarian argument based upon the Socratic Method, that goes something like this:

• A legal decision to consider is chosen
• One student or the professor argues one interpretation
• Another student is assigned to argue the opposite position.

It does not matter what the individuals actually believe. They need to argue their assigned position as vigorously as they can. The goal is not for there to be a winner or loser in the argument. The goal is for the students to get the most complete and thorough understanding of the issue under consideration as possible. And only by giving both sides equal status can this be done. In the end, the law students emerge better prepared.

So, again, why would anyone want to work with a contrarian? In short, because the contrarian keeps the rest of us honest.

Consider the 6 behaviors of a contrarian that I mentioned earlier. Viewed within the context of law school argument, the contrarian is simply holding up his end of the bargain, to represent the opposite viewpoint. He’s the one most likely to find the holes that would otherwise eventually kill the project. Sure, he may find 9 holes that are not real for every real hole. But the one real hole he finds probably never would have been found by anyone else. In that sense, the contrarian is actually the ultimate optimist, because he’s the one trying the hardest to protect project success.

So, when you see that Contrarian on your project the next time, give him a hug…well, maybe not.

Back in the early 90s, I worked for TRW on a project to develop new technology for airbag sensors. The airbag sensor is the device in the car that decides whether or not to fire the airbag. Obviously, this is a pretty important, potentially life-saving decision. False negative … someone might die. False positive … someone might get unnecessarily injured by the airbag, not to mention the cost of reloading, which was about $2000.

The goal of the project was to migrate from mechanical sensors in the bumpers to electro-mechanical sensors under the gear shift to save wiring cost and improve reliability. An HC11 Motorola processor with a built-in accelerometer and A/D was used to measure the deceleration. The processor would evaluate the acceleration data and decide whether or not to fire the airbag.

Sounds simple enough. Since the integral of acceleration is velocity, add up the acceleration data and that is your change in velocity. The greater the change in velocity, the more likely a crash worth firing the airbag.

The guys in Michigan crashed up several different classes of cars to provide us with raw acceleration data files. Some of the raw data files had names like “30 MPH 50 lb Pig Left Front”. I can only imagine what they used to get this data . In any case, we had to run this data through our algorithm and make sure the airbags fired on real crashes but not on the others.

Easier said than done.

As it turned out, hitting a 50 lb. pig at 30 MPH is a pretty decelerating event. In the short 30 milliseconds required to make this decision, the algorithms had a hard time deciding what to do correctly.

It was only after 60ms or more that the algorithms started to reliably distinguish between a real crash and roadkill.

 

The critical issue was this … the timeframe was too short. There was no reliable way to differentiate all the possible “fire” scenarios from the “no fire” scenarios with just 30 ms worth of data. You just have to get more data. As it turned out, they redesigned the airbag to inflate faster, thereby allowing 60 ms to make a decision, which was just enough time.

It seems to me that we’re in the same boat with Global Warming.

We have a few decades of temperature data for an Earth that is 4.6 billion years old. Looking within those few decades, it looks like we might be heading for catastrophe.

Or, it might just be a normal temperature cycle.

Just as with the airbag sensor, we don’t have enough data to make a reliable decision one way or the other. And just as with the airbag, the cost of being wrong is great. False negative … we experience catastrophic climate shifts. False positive … we unnecessarily impose costs upon ourselves and developing countries that can cause social catastrophe.

Personally, I don’t know which is the right answer. And I don’t know how anyone else can know with certainty that they have the right answer. I wish that both sides in this discussion had a little more humility and they would acknowledge that they don’t know more than they do know. And that they could be wrong.

Or maybe these experts are just like the airbags … full of hot air.

harry the ASIC guy