Modern Luddites

Entire bookshelves have been written about the various human attitudes towards the perspective of change. And it is quite intriguing to watch the well know patterns unfurl in a real-life situation. I am just returning from a series of customer visits in Europe, and as it turns out, during this streak, I got to experience nearly the exact same conversation in two different meetings. 

“High-level synthesis will never be able to do better than what we can do by hand!”

In both circumstances the conversation started with this statement. As I went on explaining how HLS can be driven to achieve production-quality results, an engineer interjects and challenges the various case studies and customer testimonials.

“But you are completely missing the point. We have more silicon real-estate that we can possibly use. Our challenge is not to further optimize for area, we need to cut design time by half to remain competitive!”

In both circumstances the project manager, attending the meeting with the rest of his team, stood up and clarified the situation in a pretty clear way.

“But if we use these tools, the chip will be bigger and it is going to cost the company a lot of money!”

Despite the manager’s explanation, the engineer insists. The emotional nature of his stance is quite perceptible. The conversation has shifted from EDA tools to business goals, individual contributions and fundamental roles.

“Assuming we are slightly bigger area-wise on a portion of the design, the consequences will be negligible compared to the time savings. You know the competitive pressure. We need to tape-out those ASICs faster. This is what the board expects from us…”

Resistance to change is a well understood pattern of human psychology. There are many examples of how innovation can cause disruption and discomfort, the most famous case being probably the one of the Luddites, a social movement of British textile artisans in the wake of the 19th century. Back then, the industrial revolution brought mechanization and automation to assist with what used to be a labor intensive task. Many skilled and passionate artisans feared the new wide-framed automated looms would constitute a threat to them and their way of life. Resisting these advances, Luddites organized in bands and, at night, attacked factories, destroyed machines and sabotaged equipment.

Don’t get me wrong. Coming from France, I love my artisan “boulangerie”, “charcuterie” and “fromagerie”. Craftsmanship is the recipe to the best things out there; and even in electronic design there is – and will always continue to be – a place for such highly-skilled optimization work. But economic reality also has its say and in fierce markets, change is the only constant.

And I don’t doubt that these two design groups will take the right step forward. After all, it’s not about high-level synthesis: it’s about overcoming their critical business issues.

About Thomas Bollaert

imageMy first encounter with HLS, back then behavioural synthesis, dates more than 15 years. Since then my ventures have led me to explore many aspects of the ESL design flow, including HW/SW co-design, architecture exploration and of course, C synthesis. Five years ago, I joined Mentor to develop the Catapult C product line in Europe. Recently, my little family followed me all the way from Paris to Oregon, where I now serve as product marketing manager for Mentor Graphics' high-level synthesis product line. Visit Thomas Bollaert’s Blog

More Posts by Thomas Bollaert

More Blog Posts

Preparing Recommendations

Comments (↓ Add Your Own)

4 Comments on this Post

Commented on 3:54 PM, Mar 31, 2010
By Lou Covey

I'm not sure the comparison to Luddites is accurate here. The Luddites were concerned that jobs would be lost if new technology would be adopted. The issue between the engineer and the manager here have nothing to do with jobs but outcome. The engineer wants to build a technologically superior product, regardless of the cost. the manager wants to build a profitable product. At this point, the vendor needs to provide a valid economic reason to win the day, and it has to be something other than, build a better chip. The vendor needs to consider the needs of both sides, especially when his opposition is coming from the guy with the checkbook.

Commented on 4:30 PM, Mar 31, 2010
By Thomas Bollaert

The conversation is between the engineer and his manager, with the EDA vendor as a catalyst. But while the engineer talks about creating a better (smaller) product, he is really talking about his own case, and his contribution to the organization. And in this case, the unspoken issue is indeed about jobs and changes in the work environment due to technology and economic factors. But unlike Luddites, I don't expect hardware engineers to wreck their labs or fight the British army.

Commented on 12:50 AM, Apr 1, 2010
By RunningMan

I am a designer of multimedia IP. I take pride in what I do: creating optimized RTL. But these days, my management doesn't seem to care about "better". Faster and cheaper are the only two words they seem to know.

Commented on 11:54 PM, Apr 8, 2010
By Gary Dare

Actually, Thomas, the Engineer should have realized that area optimization plays a role (its priority, on the other hand, is balanced against performance and power) since size of die brings packaging issues. For the manager, the issues are cost and time. CatC addresses both of their issues. For Running Man, in the case of the Engineer, there is added benefit in getting the algorithm implemented properly (and verification as well as design) by tackling the problem at ESL rather than RTL, where they must interpret the high level design created in (e.g.) Matlab or Excel! (I have used a Calma layout workstation as a grad student but I see no need to go back there!)

Add Your Comment

Please complete the following information to comment or sign in.

(Your email will not be published)