Suggest A Topic


Suggest any ADC/DAC-related topic you wish to discuss on the blog, and I might add it as a post. If not, those interested can always continue to discuss it here.

5 responses to “Suggest A Topic”

  1. Praveen | June 14, 2012 at 23:00 | Reply

    Glad to see an exclusive blog for data converters . It will be great if we can have a discussion on the

    “Best universities/academic institutions with strong data converter research” .

    Would like to know if it is dominated entirely by a few just like data converter industry ?

    From a student’s perspective , it will be good to know where exactly is data converter research happening than going on with general electrical engineering ranking given by sites like “us news” 🙂

    • Converter Passion | June 20, 2012 at 15:29 | Reply

      Thanks for the suggestion, Praveen. I’ve been contemplating some visualization of scientific output from different sources/regions/organizations. Given how prestigious it would undoubtedly be to end up among “the best”, I’ve been a bit careful about publicly going out with such data. Even a small mistake could be very sensitive to those involved. It is also complicated to define the criteria for “best”.

      It is however a great suggestion, and I can understand why it could be of interest to many, so I’ll look into it and try to figure something out that won’t make me hated by just about every professor in the world 😉
      I guess you saw the “no-affiliation-names-given” power-law plot in https://converterpassion.wordpress.com/2010/11/06/who-has-done-all-that-ad-converter-research/ which illustrates that most of the output comes from a handful of organizations.
      But, like I said, I’ll try to come up with something.

      BTW: How would you prefer to measure “strong data converter research”? Implementation paper count? Some performance criteria?
      Myself I’m currently leaning towards looking at those with more than x publications total, then perhaps do some profile of how their publications distribute in the {fs, ENOB} plane so that you get a rough idea of what kind of ADCs they are focusing on. Fast/slow, hi/low res. Perhaps combined with an architecture profile.

      • Praveen | June 20, 2012 at 20:35 | Reply

        Thank you for kind response and time . I am eagerly waiting for your survey results.

        I certainly agree that measuring “strong academic research” or defining a criteria for “best” is difficult . However as you suggested , number of publications ( probably at top journals and conferences like JSSC , ISSCC etc ) and the reported performance ( may be in terms of fs, ENOB , FOM ) of those publications do indicate a qualitative measure as to how the universities are performing in data converter research .

        Though steady publication count indicates a focus on the research, after a critical mass has achieved , I think its the quality of those publications that will truly differentiate their research . Also it will be interesting to see if those top universities are still the dominant players or their position is being constantly challenged by the ones that are ramping up very fast.

        Thanks,
        Praveen

  2. Mike Russell | June 20, 2012 at 14:19 | Reply

    Sir,

    You have a treasure trove of information available on your site. It appears you update the ADC FOM section based on technical publications, but do you also provide a market analysis of parts available to consumers?

    I am specifically interested in tracking what devices are leading the pack with respect to lower power consumption, higher rates (>125 Msps), and smaller form factors. I believe Analog Devices may be the present leader with the best balance between high rate, low power consumption, and small size.

    I believe DARPA had a multi-phase project a year or so ago that aimed obtain performance that at the current rate of improvement wouldn’t be available for another 20 years. I lost track of this project and its status, it may be a worthwhile discussion topic.

  3. Converter Passion | June 20, 2012 at 16:04 | Reply

    Thank you for the kind words Mike!

    In part, similar to the above answer, I have been very careful so far to publicly compare commercial parts. The reason is that – although I currently have the performance and design data for over 600 commercial parts logged into the database – it is not yet an exhaustive survey as in the case of the academic data. With the academic data, I am almost 100% certain when I say “author X published the best FOM so far”, simply because I surveyed just about every paper published. With the commercial data, I’m not there just yet. When I am, I might start publishing comparisons including such data. On the other hand, it is going to be notoriously sensitive to list a single part or (worse) company as the “best”, so I may actually choose to not do that.

    All the above said, though, the data I already have is highly useful. It just doesn’t reach my geek-level 100% exhaustive over time and performance space, but has almost complete coverage from telecom speed and upwards due to my personal preferences. For a client I can (and do) include this data in my analysis.

    Similar to the answer to Praveen above, it is always difficult to give an objective definition of “best” balance of design and performance parameters, but Analog Devices is always a good bet. They are, after all, the market leaders. But their main competition is always snapping at their heels, and sometimes they leapfrog each other as it seems. I’d have to do a very specific investigation to say exactly who’s the best at this moment. It’s worth looking at all the big ones (and perhaps the smaller ones too):

    Reader question: Why is the data converter market dominated by six US companies?

    The DARPA project would definitely be interesting to know more about. If you, or anyone else reading this, could find out more about it I’m interested to look at it. Best of all if anyone involved could fill us in.

Leave a reply to Mike Russell Cancel reply