Tuesday, December 19, 2006

Get a (better) job

I've been reading Ask The Headhunter for years, since Nick Corcodilos wrote a column in EE Times. It's always provocative, and worth a read if you may be looking to get hired or do some hiring in the future. (And who isn't?)

Poke around his web site to get a feel for what he writes about. It's a good reference. And, sign up for his newsletter to get his periodic thought-provoking missives. Happy Holidays!

Friday, December 01, 2006

EDA is Child's Play

One of the more common parts of my job is to figure out why something stopped working, or why it is suddenly producing different (namely worse) results than before, when "nothing's changed". Things like

  • Why is the synthesis tool crashing now when it ran fine before?
  • What changed so that I'm not making area/timing any more?
  • Why is equivalence checking failing now?

For all of these sorts of things, I end up comparing logs, netlists, scripts, etc. I am the tkdiff master! With my eagle eye and understanding of the personalities of EDA tools and companies, it's a sleuthing challenge to find the culprit.

If I tried to explain this to a layman, I'd have to say that part of my job is playing the grown-up version of one of my daughter's favorite games, Spot the Difference. Yes, that's what I went to all those years of college for. :-)

Friday, November 10, 2006

The New Way of Graphics, and More

NVIDIA launched their new graphics chip architecture, the chip code-named G80. Check out these reviews of how the performance blows away yesterday's best graphics.

This page has the G80 Hardware Specs. How about 681 million transistors?

The architecture is very interesting and different from past GPUs designed by NVIDIA or ATI (AMD). Rather than dedicated compute engines for different stages of the "graphics pipeline", this new architecture has scores of 1.35 GHz streaming processors that can be assigned to the most demanding graphics tasks for a particular game or scene. Because of this generality, it can be applied to completely new problems, as described in ExtremeTech's article:

As you might imagine, this will be aimed at highly parallelizable compute tasks, such as scientific computing, financial analysis and oil exploration. Nvidia will actually be offering a C compiler to facilitate thread computing, so programmers interested in general-purpose programming with the GPU don't have to be locked into Direct3D semantics. -- GeForce 8800 GTX: 3D Architecture Overview

Thursday, October 26, 2006

The Longest Quick Reference Ever

This week I wanted to run an EDA tool that I don't run every day. I needed to look up a command line option, and found that, inexplicably, the option wasn't listed by the program's help option.

So, I trudged over to the online documentation and was happy to find the Tool X Quick Reference manual. I figured it'd be a couple of pages long and I'd print it out and keep it next to my computer.

Well I was shocked (shocked, I tell you!) to find that the "Quick" Reference manual is 86 pages long. Something is seriously wrong with your product or your tech pubs when this happens.

What do you think? Are you surprised that a tool could be so inscrutable? Is it just that the writers don't understand what Quick Reference means? And, can you guess which tool I'm referring to? (I'll post the answer later.)

Tuesday, October 24, 2006

Just how unsexy is EDA?

April Fools in October? I thought I must be reading a joke: in Life without Dataquest, Gabe Moretti reports that research firm Gartner Dataquest has dropped coverage of the EDA industry, and laid off reknowned analyst Gary Smith and his colleagues.

This is disturbing. I don't like changes like this! Dataquest (and Gary) have been covering the EDA industry for many years, providing analysis and prognostications, that, agree with them or not, were a starting point for debating the direction of EDA technology.

Let's hope that Gary and crew resurface to share their wisdom and opinions again. Best wishes to them.

10/25 Update: see the "backstory" and lots of comments from EDA industry personnel in John Cooley's article on Gary Smith's departure.

Over 500 Million Served

The production volumes described in NVIDIA AND TSMC CELEBRATE NEW MILESTONE:500 MILLION PROCESSORS are mind-boggling! What a profitable business model for both TSMC and NVIDIA, and what a contrast to the "old" ASIC model.

At my previous company, the lifetime projected volume of the ASIC we were designing was around 50,000 units. Yet, the ASIC vendor had to provide significant complex IP and especially engineering headcount to shepherd the chip to the prototype stage. Unfortunately, the market changed and that chip never made it to production. Sorry, ASIC vendor!

No wonder that most ASIC vendors are having a very tough life, with rising costs and declining design starts (and especially, fewer designs going to mass production). Contrast this with foundries such as TSMC, which get the customer to do all the IP integration, implementation, and verification, and the foundry can just focus on selling wafers. No wonder this model is eclipsing the ASIC model.

Wednesday, October 18, 2006

ICCAD Early Reg. Deadline Today

ICCAD-2006 is coming to San Jose this November 5-9 and the early registration deadline is TODAY, October 18.

Registration reveals a major benefit of IEEE or ACM membership: you get a discount of over $100 registering for the full conference!

Tuesday, October 03, 2006

Wondering about the IEEE

I feel like I should be a member of the IEEE. I am, this year. Throughout my career, it's been an on-again, off-again relationship. Even when my company reimburses me, sometimes I feel that it's not worth it.

What I want it to be

  • Teach me about new developments in my specialty. There are some gems, but it's harder to find from IEEE publications vs. trade publications like EE Times or EDN. Part of it is because I can't find a Society/Publication that aligns well with my specialty (see Gripes, below).
  • Teach me about interesting developments in other specialties. The articles need to start from a basic level, not trying to prove how sophisticated and complex the author can be. I am not finding this from the IEEE. Something like Scientific American or Discovery magazines might be a more accessible resource.
  • Represent my professional interests. The IEEE-USA tries to do this, and I'm a member. But membership is optional, and I wonder how many USA EEs belong or are even familiar with it?

What IEEE should do

  • Give a primer to members on how the IEEE works. I have sort of figured these things, but why doesn't the IEEE come with a "user's manual" to address these dead-on?
    • What is a Region and what's a Chapter?
    • What is the difference between their publications Spectrum, Proceedings, and Transactions?

Other gripes

  • There's no Society for Digital IC (RTL & Gates) engineers! I just can't figure this one out. There are certainly Societies for semiconductor process, circuit design, and packaging. But what Society should a plain-ol' ASIC engineer belong to? Why isn't there a perfect fit for such a common EE discipline?
  • The price/performance of an IEEE membership really seems out of whack. Why does it cost me (or my company) over $140/year, when the main benefit is a magazine subscription or two? Why, I can subscribe to EE Times for free and get far more practical news and technical information! Where is all the IEEE money going? To some huge bureaucracy in New Jersey?

What do you think about the IEEE? Do you belong? Why or why not?

Tuesday, September 12, 2006

EDA standards development within the IEEE is mismanaged

Come on, Gabe, tell us what you really think about the IEEE DASC: EDA DesignLine | EDA standards development within the IEEE is mismanaged

It's quite an indictment about what ought to be one of the most prestigious standards bodies related to EDA.

I know that most EDA vendors have groups whose full-time responsibility is the care and feeding of standards groups. In an ideal world, they'd be ensuring the most elegant and implementable open standards. The Internet standards bodies seem to be a fine example of this, and have delivered interoperable standards that spurred an incredible information revolution over the past decade. Why can't the EDA industry have that, too?

In reality, the EDA vendors do quite a bit of political maneuvering and intrigue. For example, we might only have to deal with one HDL if Cadence had opened up Verilog before VHDL emerged. Imagine the tremendous resources spent on VHDL that could have been saved. We almost had two library standards because Synopsys wouldn't open up the "Liberty" format until there was a revolution brewing to move the industry to DCL/ALF formats. You've never heard of DCL and ALF? Well, Synopsys finally opened up .lib and no one needed those alternate formats.

Anyways, getting back to Gabe's post, it's sad how quickly the proposed low-power standards are degenerating into political battles between EDA companies, much like the CCS vs. ECSM library standoff. Cadence's initial proposal sounded pretty reasonable, though perhaps I'm naive to think that they and Synopsys could have agreed on a common standard.

Friday, September 08, 2006

A Good Chip News Site

Geek.com's ChipGeek page is a good source of semiconductor news items. The analysis and commentary by the editor adds helpful "color" to the pure news. If you dive down into the readers' comments, they range from wonderfully instructive to embarassingly sophomoric.

By the way, the EDA blogs I follow seem awful quiet! Are people burned out after DAC? Or is summer over and it's time to get back to work?

But, on the professional blogging front, a shout out to Gabe Moretti, who is editor at EDA DesignLine, a new web site of the CMP tech publishing empire.

Tuesday, August 29, 2006

YAST - Yet Another STA Tool

I'm initially puzzled when an EDA vendor fields a new tool to take on Synopsys PrimeTime, e.g., Incentia Timing Analysis and Constraint Management Software Adopted by Ambarella.

PrimeTime is one of Synopsys highest-quality, most trouble-free tools. The main gripe I have with PT is that it can only tell you about timing problems -- it doesn't fix them! Hence, a cottage industry of home-grown scripts and supplementary tools exists to take PT reports and generate ECOs to fix the violations.

However, some of the features in this TimeCraft tool sound pretty attractive, and give me second thoughts about assuming nothing can beat PrimeTime. Attractive features include

  • MSV support
  • "multi-task capability for multi-corner and multi-mode analysis"
  • Advanced OCV features
  • Constraint Manager

Friday, August 18, 2006

Charter for a Corporate CAD Group

I discovered another EDA-related blog, Cool Verification. Cool Verification: Creating Corporate Standards? Beware... is an excellent post outlining the considerations in creating and managing a "Corporate CAD" department.

I've taken this group's mission for granted, but my company didn't even have such a department four years ago, even as it was cranking out huge industry-leading chips. What finally pushed us to create a central group was the proliferation of derivative chips and especially the diverse chips being designed by new business units. It no longer made sense to have a few "designers with a tool fetish" responsible for creating and maintaining all the flows. Hence, a CAD department was born. The referenced post touches upon the dynamics of trying to please all groups in the face of different requirements and priorities.

Monday, August 14, 2006

EDA Confessions, Part I

After my prescient DAC trip report, I've been thinking of a couple of times in my career when I've really missed the boat on EDA companies. I have a couple of examples that I can't forget. I'll write about them one at a time. Confession is good for the soul. ;-) The first is a company that I ignored, yet in retrospect could have done a great jobs for us.
The Time
Late 1990s
Product Type
The Company
Silicon Perspective (SPC)
The Problem
My company was working with Synopsys on "Chip Architect" and "Floorplan Manager" to close timing on "deep submicron" designs. These were around 0.35 micron, IIRC. I never used Chip Architect, but my colleagues were beta testing and helping to debug it. I spent a lot of time with Floorplan Manager; it was pretty inefficient to back-annotate tons of layout data onto Design Compiler, and for DC to optimize designs in a way both effective and feasible for physical design. We were struggling to close timing on the design.
The pitch
An SPC Sales guy kept calling me, asking us to evaluate FE. He made extraordinary claims about its performance, capacity, and quality of results. He wanted us to try it on a multimedia design one of our groups was working on.
What I did
I couldn't believe that this little startup could deliver such amazing results that Synopsys and Cadence couldn't! Clearly, he was just making wild claims to get our attention. So, I never did bring him in for an evaluation.
What happened
Our design project took so long and it was eventually cancelled. And I ended up leaving the company around the same time. Later, SPC was acquired by Cadence for a large sum and today is the foundation and namesake of "Encounter", their whole family of Digital IC implementation products!
Lesson learned
Don't dismiss startups lightly! Consider them skeptically, but give them a chance if their story seems plausible.

Thursday, August 10, 2006

Graphics Beginners' Guide | Tom's Hardware

Graphics Beginners' Guide | Tom's Hardware is a great introduction for understanding (and purchasing :-) today's graphics cards.

It's a series of three articles, all of which are accessible from this page.

Thursday, August 03, 2006

After the AMD / ATI merger: Will Nvidia GeForce still compete with ATI Radeon?

After the AMD / ATI merger: Will Nvidia GeForce still compete with ATI Radeon? | TG Daily is the best in-depth analysis I've seen of the implications of the AMD acquiring ATI. It lays out a number of scenarios, but the end game is far from clear.

Monday, July 31, 2006

2006 DAC Report


The "hot" areas of EDA these days are Electronic System Level (ESL) and Design For Manufacturing (DFM). Indeed, there were numerous companies offering things such as SystemC design or yield-aware layout tools.

I attended DAC for one day, Wednesday. I focused on low-power and variation-aware tools, and compiled a list of companies participating in these areas. I had time to see about half the companies I identified. Below is a short summary of my impression of those companies.

Companies that Impressed

  • Azuro. Low-power clock design. They optimize clock gating and physical clock tree for low power. They claim 15-25% less power than using today's tools. This is a niche tool that is worth looking at. The most impressive thing I saw at DAC was their clock tree analysis/visualization tool. At a glance, the design can visualize the structure and quality of the clock tree. It's better than any analysis I've seen before.

  • Altos. Library characterization, including statistical effects. Their characterization system, Liberate, includes their own very fast SPICE-like engine. They claim ~10X faster than today's tools. They can generate .lib and also the latest CCS and ECSM models. They appear to have a solid grasp of what's needed for statistical characterization, and their high performance will be very welcome for that. They're working to support SSTA from Extreme DA. The founders are from CadMOS (SI tool acquired by Cadence) and seem sharp. They recently published an article in EE Times. [Note: Magma just announced an upgrade to their characterization system.]

  • Proficient. Low-power design. Intriguing. It seems like a pretty small company.

  • Prolific. Post-layout power & timing improvement. Their tool may be useful and the approach is sensible and simple (bolting onto PrimeTime). Designers could do something like this themselves, but if their tool is reasonably priced and effective, it may be worth it.

  • TSMC. They conducted a series of advanced methodology presentations. It is impressive how much they've thought through the whole methodology and worked with EDA vendors on their "Reference Flow 7.0".

Companies that Didn't Impress

Censored! I'm don't yet have John Cooley's nerve to dis companies in public. In my case, it wouldn't really be fair to opine on them after a half-hour presentation or demo, anyway.

Other Notes

  • "Power Forward Initiative" lunch hosted by Cadence. This is a consortium initiated by Cadence to develop a standard "CPF" Common Power Format description of low-power design intent. CPF would drive implementation, checking and analysis tools. Something like SDC is for timing. This may bear fruit in a couple of years, but of concern is that Synopsys is not on board. Will it be another political fight like CCS vs. ECSM? [Yes, the politics have already started with a Synopsys counter-proposal.]
  • Coincident with DAC, Synopsys announced enhancements to PrimeTime and Star-RCXT with statistical capabilities.

Best Freebie

There didn't appear to be major schwag as I remember from years ago. I remember when you could always count on Viewlogic for a bat or a hockey stick, or Altera for a soccer ball. However, one gift at this year's DAC impressed me. It was the hand-held fan from Bluespec. It appears to be your standard simple battery-powered fan. However, there are LEDs embedded in the fan blades, and when you turn it on and the fan spins, the lights spell out promotional messages for Bluespec. Pretty "cool"!

Other Coverage

Friday, July 28, 2006

Cooley's Must See List for DAC 2006

It's unfortunate I didn't see this before going to DAC, but EDA Gadfly John Cooley posted an extensive Must See List for DAC 2006.

John's list is quite long, spread out over 24 categories. Rather than being very discriminating, it's more of a concise summary of all that's new and significant at DAC.

Companies Common to Cooley and Busco Lists

  • Altos - Statistical timing characterization
  • Apache - IR drop analysis.
  • Athena - simultaneous timing closure (post-route)
  • Azuro - low power clock implementation
  • ChipVision - Low-power ESL design
  • Extreme DA - Statistical STA
  • Forte - SystemC synthesis
  • Magma - Talus implementation system, Blast Fusion design closure
  • Prolific - post P&R timing optimization
  • Sierra DA - physical synthesis & routing.
  • Zenasis - custom cell generation for timing optimization

Companies Common to Cooley, Smith, and Busco Lists

  • Apache
  • Forte Design Systems
  • Magma DA
  • Sierra DA

Tuesday, July 25, 2006

Gary Smith's Must-See List

Coincidentally, just after posting my DAC must-see list, I see the list of veteran EDA industry observer Gary Smith. His list is broader than mine with a good dose of "ESL" and "SVP" vendors: EETimes.com - Imperas tops analyst Smith's watch list

Looking for intersections, we both mentioned the following vendors:

  • Apache
  • Forte Design Systems
  • Magma DA
  • Sierra DA
  • Proficient Design

It appears that I don't share Gary's enthusiasm for ESL, and he really doesn't share my enthusiasm for "statistical" or "variation" analysis.

DAC is Here!

EE Times is doing a nice job of covering the Design Automation Conference, which started yesterday in San Francisco. Road to DAC: Complete Conference Coverage has lots of articles and even a set of webcasts that includes an interview with Joe Costello, former Cadence CEO and EDA celebrity.

Thought I'd post the list of companies that I'm interested to visit. It is far from a comprehensive list, but these are companies mostly in the "design implementation" space that piqued my interest based on what I've read.

  • Altos - Statistical timing characterization
  • Apache - IR drop analysis.
  • Athena - simultaneous timing closure (post-route)
  • Azuro - low power clock implementation
  • ChipVision - Low-power ESL design
  • Extreme DA - Statistical STA
  • Forte - SystemC synthesis
  • Magma - Talus implementation system, Blast Fusion design closure
  • Proficient - Low power design
  • Prolific - post P&R timing optimization
  • Sierra DA - physical synthesis & routing.
  • TSMC - Reference Flow 7.0
  • Zenasis - custom cell generation for timing optimization

Wednesday, July 19, 2006

EDA Players Detail Involvement in TSMC Reference Flow

EDA Players Detail Involvement in TSMC Reference Flow - 7/18/2006 - Electronic News details which EDA vendors are providing which tools and technologies for TSMC's recently updated "reference flow".

With respect to statistical STA, an area of interest to me, the article sites support from Cadence and Magma, but not Synopsys. That's funny, I don't think of Cadence as a player in the timing space. They don't even really have an STA tool, do they? I thought all their customers sign off on PrimeTime.

Monday, July 17, 2006

TSMC Flow Adds Statistical STA

TSMC (Waferzilla?) has announced the latest additions to their Reference Flow, version 7.0. The two main additions are Statistical STA and more Power Management features.

Interesting that the Statistical STA flow supports vendors Magma and Synopsys. Magma has been promoting their Quartz STA tool. Has Synopsys announced a Statistical STA tool? I don't recall seeing it.

And, what about library support? How will this be done? I haven't seen any standards offered for representing variation in Liberty format, for example.

Overall, this announcement is a hopeful sign, but I wonder how real it is, or when we might actually see a working flow.

Thursday, July 13, 2006

Statistical Characterization and STA Primer

Cell model creation for statistical timing analysis is a great primer on Statistical Static Timing Analysis (SSTA). It's written from the viewpoint of the challenges facing statistical library characterization. After reading this article, I wonder how we're *ever* going to have a complete set of SSTA-characterized libraries. Fortunately, the EDA startup Altos, which is where one of the authors works, has a solution under development. Unaddressed by this article is how such characterization will be represented for all the vendors' STA tools. As of now, there is no .lib "Liberty" standard for statistical modeling. We've got a long ways to go!

Wednesday, July 05, 2006

Metcalfe's Law is Wrong

IEEE Spectrum: Metcalfe's Law is Wrong gives in-depth analysis to one of those laws that people accepted uncritically.

I don't know what analysis Metcalfe's original "law" was based upon, but it doesn't seem as robust as Moore's law. The paper referenced here goes on to hypothesize what the law regarding the value of networks should have been. It dovetails nicely with another concept, the "long tail", that comes up when discussing e-commerce, blogs, and many other phenomena.

DAC's 44 New Exhibitors

Do you know much about these new EDA companies? 43rd Design Automation Conference to Feature 44 New Exhibitors They're news to me. Which ones will capture enough interest to thrive?

Tuesday, June 20, 2006

Power Primer

Power Management DesignLine | Low Power CMOS Circuit Design is an excellent survey of CMOS power consumption along with the most common techniques to reduce power.

This article is a good quick reference -- why, these questions might even come up in an interview! ;-)

Tuesday, June 13, 2006

Hot Summer for CPU Companies

Hot Summer for CPU Companies is a nice summary of the recent CPU wars, and how Intel Israel's designs may yet save the bacon of "Chipzilla".

I wonder what tricks they're using for power-efficient design?

Tuesday, June 06, 2006

Lies, Damn Lies, and Engineering Manpower

Heard the One About the 600,000 Chinese Engineers? is very interesting for how a set of poorly-vetted data can take on a life of their own.

I've heard the IEEE-USA referring to such numbers, too. I think their interest is a combination of wanting to strengthen American competitiveness, but also to ensure increased funding for engineering education and research. In other words, self-serving for the greater good!

Friday, June 02, 2006

Cadence Powers Forward

I wasn't sure to make of Cadence's announced Power Forward Initiative. This is such a nascent field, do we know what to standardize?

But then I could think of some design practices that would be very good to have a standard way to describe and verify, such as what voltage and power domains various parts of the design should be connected to, what is the design convention for isolation and level shifting, etc.

This could be a useful standard. Let's hope it doesn't degrade into another Cadence vs. Synopsys "we have two standards so we have no standard" like ECSM and CCS libraries!

Monday, May 15, 2006

Semi Business is too good? Allocation, here we come!

The headline EETimes.com - TSMC puts customers on allocation certainly caught my eye and made me think of the good ol' days.

But reading the fine print, this is only for older geometries, i.e., 0.18 and 0.25 micron. Even though there ought to be lots of "second-tier" fabs with capacity, especially in China, the article explains that it's difficult to port mixed-signal design from one process to another. Once you go to production with a particular foundry on these types of "touchy" designs, you are in bed with that foundry for a long time!

TSMC DFM format empowers fabless design

Good news! Fabless semiconductor vendors (e.g., NVDA, BRCM, etc.) won't be up a creek in getting yields at 65nm and below: EETimes.com - TSMC DFM format empowers fabless design

If this works as advertised, this collaboration will enable TSMC + EDA vendors to work like an Integrated Device Manufacturer (IDM), and not suffer from walls and insufficient information sharing.

I'm looking forward to the TSMC Technology Symposium, which is always good for seeing how the foundry world operates, and what interesting issues they address that we seldom hear about in the "front-end" design world.

Monday, May 01, 2006

Startups to watch

This is an intriguing list of startups, though the selection criteria is rather mysterious. Reviewing it gives you a feel for what's coming and what's hot in our industry. I spotted a few EDA companies, several in the "DFM" space and at least one related to "ESL" design.

"editors have selected companies based on a mix of criteria including: technology, intended market, maturity, financial position and investment profile.

Startups on the Silicon 60 list include companies involved in semiconductor chips, memory, fab equipment, packaging, foundry, materials, MEMS and EDA software that made an impression on EE Times editors. They are emerging companies to watch for a wide variety of reasons.

EETimes.com - EE Times updates list of emerging startups

Tuesday, April 25, 2006

UNDERclocking, anyone?

Those darn ATI chips are "hot", alright: Apple purposely slows MacBook graphics chip.

GPUs are just so huge. It's an "embarassingly parallel" application, so ATI & NVIDIA can just keep throwing more gates that the problem, following Moore's Law. The Holy Grail is to render cinematic quality in real time. We'll get there some day, but there will need to be lots of technology and methodology innovations to use power wisely.

Friday, April 14, 2006

Building up a `bump' in the new flat world

I attended an alumni event, Berkeley in Silicon Valley, earlier this week. It was at the Computer History Museum in Mountain View, which is definitely worth a visit for the audience of this blog. Ah, the memories! They have all the early PCs, and really old mechanical or analog computers, slide rules, etc.

The featured speaker was Professor Richard Newton, Dean of the UC Berkeley College of Engineering (my alma mater). I was going to post a write-up of his talk, but Mike Langberg from the San Jose Mercury News already did a very professional job at MercuryNews.com | 04/13/2006 | Langberg: Building up a `bump' in the new flat world. So go read that!

Some comments/notes:

  • Synthetic Biology is "the next big thing". Remember "Plastics" from the movie The Graduate? Well, this is like that for the next century.
  • The questions from the audience were surprisingly pragmatic:
    1. How can we invest in Synthetic Biology? Prof. Newton didn't have specific opportunities. It's an extremely new and risky field, more "research" than "development" at this point. I wonder if Biotech mutual funds would cover this space? It seems like they would.
    2. What about the problem of H1B visas and how that will devalue US engineering careers? I don't think a university professor can really appreciate the anxiety that some engineers in industry feel. Professors see the newest, sexiest technologies and get to work with really interesting emerging companies. Personally, I've become less concerned with H1B than with "offshoring". At least the visa-holders bring their high skill levels and tax-paying ability into the US!
    3. With demographic changes in California, the questioner suggested that the future majority ethnic groups (i.e., Latinos)do not value education as highly, and does this pose a challenge to Prof. Newton's call to make the Bay Area a "bump in the flat world"? Wow, this seemed like a politically incorrect question, and the questioner must have had guts (or been oblivious) to ask it. Prof. Newton didn't cut him any slack, saying that he found the Latinos that he encounters in his work to be just as passionate about education as anyone else.

Monday, April 10, 2006

The State of 3D: Economics of 3D

The State of 3D: Economics of 3D is an excellent history of the graphics chip industry, i.e., NVIDIA vs. ATI, past, present, and future!

Furthermore, as you dig into the article, you can learn a lot about semiconductor economics, die size, binning, and the huge impact that design and fabrication issues can have on economics and business success.

Wednesday, March 29, 2006

An Awesome DATE Report

Peggy Aycinena has written a very detailed and well organized report on DATE, which is the European equivalent of DAC. Read it all here: EDACafe Weekly : DATE 2006: Between Gem�tlichkeit und Angst - March 13, 2006

Furthermore, she refers to the very impressive keynote address by Mentor Graphics' CEO:

** The keynote address delivered on Tuesday morning at DATE by Dr. Wally Rhines, Chairman and CEO at Mentor Graphics and Chair of the EDA Consortium (EDAC) was nothing short of a magnus opus. Rhines simply said it all - simply.

As an aside, I've never met Peggy, though I recognize her name as a long-time EDA analyst and commentator. This column was the first time I've seen her picture, and I can't help but be struck by her resemblence to Arianna Huffington! (And I mean this in a complimentary way.) Separated at birth? What do you think?

Monday, March 13, 2006

IT Jobs Migration: Fears vs. Reality :: AO

IT Jobs Migration: Fears vs. Reality :: AO is a thoughtful, even-handed review of the job dynamics of the Information Technology field. It refers to an in-depth report by the Association for Computing Machinery (ACM), "Globalization and Offshoring of Software".

I'd assume that the analysis of the IT field applies to the closely related engineering fields like hardware and software engineering. Let's hope that the conclusion is correct, that an open market benefits both developed and developing nations. I'm sure this is true on the macro level, but we must keep an eye on those directly affected by outsourcing -- do companies invest in "new business opportunities to generate more new job"?

Wednesday, March 08, 2006

RapidChip R.I.P.

Being modest, I hate to say "I told you so", but LSI scraps ASIC line, plans to sell DSP unit seems consistent with the doubts about RapidChip that I expressed in this post in July 2005. Is there a stampede for the doors? A pioneer in the field of "structured ASIC", Lightspeed Semiconductor, exited the market just last month.

Too bad, I was sort of longing for the simple days of yesteryear when we did Gate Arrays, and these products resembled re-warmed Gate (or Embedded) Arrays.

Tuesday, March 07, 2006

SystemVerilog, a better RTL

I'm looking forward to being able to use SystemVerilog for Design (synthesis). Looks like this effort is picking up according to EETimes.com - Vendors warm to SystemVerilog

I haven't done a thorough survey of the language, but a few of the things that will help Design/Synthesis are:

  • Interfaces (objects that define ports (and even protocols) between designs)
  • Explicit treatment of intent for case statements
  • Explicit treatment of intent for always blocks
I'm sure there are many more. I look forward to trying it.

Is Synopsys undermining synthesis standardization? Gee, they'd never do that. ;-) Looked at from the perspective of protecting their franchise, they have the most to lose from making designs portable. We've the same types of battles over library standardization (finally, Liberty) and a synthesis subset for Verilog HDL.

Friday, March 03, 2006

The 9-in-1 Wonder Chip

I don't get all the hype about the IBM/Sony/Toshiba "Cell" chip: The 9-in-1 Wonder Chip - September 5, 2005

Sure, I can believe it's a very fast, efficient CPU, with provisions for multi-processing. But IBM is betting its whole computing future on it? And how could this be so revolutionarily different from anything that all the other genius CPU designers have come up with, including AMD and deep-pocketed Intel?

I don't get it, but I'd like to. What's so different?

Wednesday, March 01, 2006

Engineers' pay will fall

Talk about "tough love"! This letter to EE Times, Engineers' pay will fall, presents a reasoned analysis of the dynamics affecting an Engineer's career, particularly in the US.

He has a very concise analysis of the reasons that Law and Medicine seem to provide more lucrative, secure careers now that "The World is Flat".

He doesn't dump on engineering and say to never pursue it as a career, but you need to go into it with your eyes open and priorities straight.

Tuesday, February 28, 2006

Computer research due for course correction

Research due for course correction lists several research directions for computer design, including
  • What's the right granularity for parallelism?
  • 1,000 CPU multiprocessor reference system for academic research (RAMP)
  • "CMOS killer" and CMOS rejuvenator technologies
  • wireless convergence and management
  • new software engineering approaches for higher productivity and quality

Fascinating, stimulating stuff. It will keep us engineers busy for a long time. :-)

Friday, February 24, 2006

Analysis & Solution of the American Engineering Crisis

OK, the post's title is grandiose, but it caught your attention. Humorous, satirical, but with a big grain of truth:

"Americans, it seems, have also grown too lazy to bother obtaining an electrical engineering or computer science degree.

How this surprises anyone is beyond me. You can float through four years of undergraduate classes sucking beer out of co-eds' navels, pick up your philosophy degree and then head to law school. Or, if you're one of those pathetic young Democrats, you can pretend to work hard at your politics classes, shoot a few rounds of golf for the school team and then whore yourself for millions in Washington. Less ambitious types can work hard enough in a grade inflation rich system to interest a Fortune 2000 company, flag down an MBA and then spend thirty or so years collecting a nice paycheck.

You'd have to be a real expletive to try and succeed in the difficult engineering and science classes"

America can lick the Asian peril by training Mexican smarties

Wednesday, February 22, 2006

Trends in semiconductor technology | Gabe on EDA

Trends in semiconductor technology | Gabe on EDA

Great summary by Gabe of an IBM talk on the big technical challenges to scaling CMOS further. Some of the points that caught my eye:

  • silicon CMOS is viable for at least 10 years down to around 22 nm range
  • Power & leakage management are essential
  • next generations of CMOS will change to metal gate and high k dielectric materials. Hey, doesn't the "M" in CMOS stand for Metal, anyway? Why did we start using polysilicon gates and still call it CMOS? Should be CPOS!
  • Yield variability is addressed in three categories: lot-lot, regional systematic, and local systematic. ... The real challenge is the local random variations. Because of the atomic scale of the devices, all effects are a function of small number statistics.

Tuesday, February 21, 2006

NVIDIA should acquire ...

And now, for a business interlude in this blog, What would Jen-Hsun Do?

Jen-Hsun Huang is the CEO of NVIDIA. In NVIDIA's last earnings conference call, management announced the intention to make more acquisitions. The article that I've linked to is a survey of the markets that NVIDIA competes in, and offers a theory as to which company NVIDIA should acquire. Interesting reading! Is he right? Check back in 2007.

Thursday, February 16, 2006

One In Two PCs Won't Run Vista's 3D Interface

One In Two PCs Won't Run Vista's 3D Interface, i.e., those with integrated graphic controllers lack the 3D power for Microsoft's new OS user interface.

Sounds like a good omen for discrete GPU vendors like NVIDIA and ATI!

Tuesday, February 14, 2006

Geeksta rappers rhyme tech talk

EETimes.com - Geeksta rappers rhyme tech talk

This is funny! I must be in the 1 percent of 1 percent of the tech community that appreciates this stuff. Jump to page 3 if you want to get to the lyrics.

Thursday, February 09, 2006

Panelists: ESL is promising but still needs work - 2/9/2006 - EDN

Panelists: ESL is promising but still needs work - 2/9/2006 - EDN is a nice survey of what some companies are doing for design above RTL abstraction.

The article acknowledges that the systemC flow has holes. It was interesting to read that people are translating RTL to SystemC. What do you call synthesis in reverse? Abstraction? It sort of makes sense if you get a huge simulation speedup, but robust synthesis and verification from SystemC to RTL would help productivity much more. That is still out in the future for the general case.

Wednesday, February 08, 2006

Digital designer's plight debated at ISSCC

EETimes.com - Digital designer's plight debated at ISSCC is just full of ideas on the future of careers in circuit and logic design.

I was surprised that there doesn't seem to be that large a market for digital circuit design, as "clever" techniques like domino logic prove to be too power-hungry.

There will be opportunities in mixed signal circuit design or for "those engaged in synthesis and place and route". That's fine by me, as the latter is my area of expertise.

Friday, February 03, 2006

U.S. designers still the best (for now), CEOs say

EETimes.com - U.S. designers still the best (for now), CEOs say is a thought-provoking survey of CEOs perspective.

Note the difference between quality and quantity, number of engineers vs. number of good engineers.

Also interesting is the comment that overseas engineers' salaries are catching up to the US. Good for them! (and for us!)

Tuesday, January 31, 2006

Development work is five times cheaper here

‘Development work is five times cheaper here’ is the claim by the CEO of ATI Technologies, commenting on their Hyderabad R&D center.

There are certainly very attractive aspects to setting up shop in India, but this article is a little over the top. I wonder if it's because this story is published for an Indian audience, or if it's because ATI has only been doing R&D there for a year, and perhaps hasn't seen the tradeoffs through a full product cycle?

Thursday, January 26, 2006

The Power of the Xbox 360 GPU

The Power of the Xbox 360 GPU is a very informative interview with a VP of Engineering at ATI. He speaks in reasonable detail about what's in the custom graphics chip of the Xbox 360 game console. He contrasts it with conventional GPU architecture, and of course espouses why it must be better than what NVIDIA will provide in the Sony PlayStation 3 GPU.

The Xbox 360 GPU, "Xenos", is actually a two-chip implementation, and one of the chips has a big embedded DRAM! Although lots of ASIC vendors talk about doing this, you don't see it often in practice. Performance-wise, it makes sense, as you get huge memory bandwidth via the very wide interface to the DRAM.

SNUG: Synopsys Users Group - San Jose Conference at a Glance

SNUG is coming! The flagship San Jose Synopsys Users' Group meeting starts on March 20, 2006. Here is the conference schedule.

I've always felt this is the best conference for the "in the trenches" chip design engineer. It's very practical, with good tutorials, keynotes, and a healthy dose of user papers.

As I was signing up this year, I noticed that almost all the sessions I signed up for were Panels or Tutorials, not User Sessions. I wonder if the conference schedule is shifted away from user papers this year, or if my interests just happen to be with the other topics. It would be interesting to look at the split of types of sessions over the years. Well, not so interesting that I'm actually going to do the study, though. ;-)

Wednesday, January 18, 2006

Library standardization effort takes flight

Library standardization effort takes flight

There is a need for standardization of modeling for Statistical STA, though that whole technology is not well understood. It may be too early to standardize!

It would be great to settle the CCS vs. ECSM divide, however. These are essentially the same modeling approach, yet defined differently by Synopsys vs. Cadence/Magma, respectively.

I have sympathies for Si2 in trying to forge these standards across EDA companies. I was more involved in the days of Liberty, DCL/DCM, and ALF standardization, and it was very hard to get cooperation across all EDA vendors. They see Standards as an area to maneuver for competitive advantage.

Tuesday, January 17, 2006

Leakage takes priority at 65 nm

Leakage takes priority at 65 nm in EE Times is a nice survey of the experience of 65nm IC designs so far.

I noted that Statistical Timing Analysis was called a "must-have" technology, because of all the combinations of operating conditions when you have multiple voltage islands.

Thursday, January 05, 2006

Two Win Prestigious Engineering Award - New York Times

Two Win Prestigious Engineering Award - New York Times

Cool! It's nice to see a couple of great engineers prominently recognized for their great contribution to technology (37 years after the invention of the CCD)!