Today at lunch with Gary Smith of Gary Smith EDA it was great to hear someone with not only a positive outlook for electronic design, but an enthusiastic one!  Of course the caveat - you have to be looking at the situation in the right way.

Challenges = Opportunities…and electronic design has a lot of challenges ahead of it.

This was too good not to share.

The conversation wound all over the landscape of the industry, dropping in to focus on some key points: EDA, ESL, and the need for integration with the embedded systems part of the supply chain.  Here are some of the main points of our discussion from memory.  Hopefully I captured them accurately and trust Gary to correct me if I go astray.  I’ve then expanded on some points that we didn’t have time to get to.

As you can imagine, there was a lot to capture and rather than turn a blog-post into a dissertation I’ve split it up.  Come back and see the rest in subsequent posts.

Globalization, Pace and Product Evaluations

The conversation started with a description of how the Xuropa Platform came to be, which lead to a discussion of some important fundamentals.  The electronic design industry is more global and dynamic than it has ever been.  Thus there needs to be a different approach to getting products in front of engineers to educate.  Gary split semiconductor vendors into essentially two camps - those that require a “high-touch” process and those that can be served with a “low-touch” process.

The “high-touch” vendors are the top “30″ that are well known and understood.  The challenge is not so much knowing who to call, but engaging at the right time within their development process.  There’s essentially a three month window when design teams can evaluate new technologies - if that window is missed, you’re done.

I’d add another logistical inhibitor to a smooth process here - group buy-in, or “socialization”.  As tools or IP are acquired by a design group, it is now likely that the group itself is split over multiple geographies.  There are also up-stream and down-stream teams that will be impacted directly or indirectly.  Therefore, we need to accommodate engaging with multiple teams spread out all over the globe as we work to get a new technology to be adopted by a company.

We probably don’t even know most of the “low-touch” semiconductor vendors.  They’re distributed all over the world, and more are coming up all the time.  These smaller providers find new niches to exploit, leverage existing and open source tools and the semiconductor infrastructure that has been built up over the past 50+ years.

Building out an expensive global technical field organization or relying upon regional power-player reps is either not an option for many and not scalable for all.

In summary, we need a new model that meets the following objectives:

1. Enables busy engineers to experiment with new technologies at their work-stations without any hassle.
2. Is globally ubiquitous, ready to go when the engineer is - regardless of location or time-zone.
3. Is scalable and economical.  The new model must operate independently of human resources.  Crucially, it must break the dependency upon the availability and location of Applications Engineers for as much of the process as possible.

This is where Xuropa comes in - as a platform to connect engineers with products globally, 24/7.

Xuropa Online Labs

1. Enable an engineer to use an actual product with no downloads, installations or configuration.
2. Provide access to software tools securely via any web browser, and so by definition the product becomes ubiquitous.
3. Enable remote global training, demonstration and pre-evaluation of technologies.

By making technology available via a Xuropa Online Lab engineers are able to access products in a cost effective manner.  Resources are leveraged globally and focused on those design teams that require a “high-touch” process.

Having said that, the web is not a cure-all.  We still need to have solid business processes to accommodate the new tools.  Also, there still needs to be an on-site evaluation - however when it does occur, the engineer will be better prepared for a smoother and quicker evaluation process.

There was one point in this part of the discussion where we may have a different perspective - how much engineers use the web.  Gary wondered if engineers are not too busy even to use the web.  From what we’ve seen on sites such as LinkedIn (who have over 70,000 semiconductor related members at last count), Design and Reuse and other news sites, engineers are using the web for research, they may be more focused in their activity and not just browsing though.

In part 2: The split between ESL and RTL responsibilities.

Posted under industry