ICECS 2010 at a glance

Approximate distribution of topics among ICECS 2010 conference contributions.

Whether you haven’t yet decided if you should go to ICECS 2010 or not, or you are simply getting hungry for some pie (Converter Passion style … ), this post may be for you. In case it helps for any of the above, I took a quick look in the ICECS 2010 technical program to see how well your data-converter passion will be served at the conference. Since the number of papers (308) is much larger than for NORCHIP, I decided to reduce my survey to only ADC and DAC contributions, and not be as thorough as I was when doing the same for NORCHIP.

My short investigation of the program revealed that, out of the 308 papers, 27 (8.8%) were ADC-related, and only one (!) about DACs. I’m certainly curious to hear what my friend and trusted DAC expert Dr. J Jacob Wikner has to say about that. Because if “ADC research in Europe is about to die, what does this meagre paper count say about DAC research?

Is there nothing more to research on DACs? Was this a one-off? Are DAC professionals more keen to publish at other conferences? What’s your view? Not just Jacob – I’m asking all of you readers.

Regardless of this fact, I’m looking forward to go to the ICECS. [After all, I’m quite ADC-biased …]

See you there!

BTW: The pie is served at the top, and don’t forget to say hello while at the conference.

8 responses to “ICECS 2010 at a glance

  1. Gabriele Manganaro

    Regarding the low number of papers on DACs: it is not an issue with ICECS. Unfortunately, the number of people doing research on DACs is considerably less than those focusing on ADCs. So there are, in general, many less papers in this topic also at other conferences (or journals). For example, at ISSCC, out of two data conversion sessions every year (approximately 15 papers in total), there have only been, in average, two papers on DACs (of which, often, one being an audio DAC) for the past 5 years or so. This should not lead to the conclusion that there isn’t interesting work in this area. On the contrary, the technical challenges are very serious (I am referring, for example, to the high frequency distortion and noise performance of current-steering DACs) and the drive to solve them is as important as for ADCs (very same reason: move the analog-digital boundary closer and closer to the signal source or to the final load).

  2. @Gabriele Manganaro:

    Great to have your input right here at the blog. Thanks a lot!

    I finally got myself to the ICECS WiFi spot, so i could moderate – any future comment of yours should now pass through without delay. (New commentators need to have their first one approved). For some reason the WiFi provided by the conference dropped my connection when trying to post a reply, so that had to wait until now that I’m back home again.

  3. So, I guess what you’re saying is that I should take exceptionally good care of friends like J. Jacob Wikner and other DAC designers I might know 😉

    Regarding the typical compositions of conference programs, I have the same impression as you. Usually no more than 1-3 DAC papers at almost any conference. Even ADCs seem to have dropped at the most recent ISSCCs – perhaps they are moving into the SoCs now.

    I recently surveyed the high-frequency performance of a set of commercial high-speed DACs, and I was quite surprised to see that the typical roll-off from DC performance is so strong across the range of offerings. You don’t usually see a similar roll-off for an ADC already in the first Nyquist. Perhaps it is inherently so in a IDAC, and I shouldn’t have been surprised, but during my entire “ADC upbringing” the view was always that “the DAC is not as difficult as the ADC”. And perhaps the wide spread of that view has caused the fewer number of people involved in DAC, and thus the fewer number of papers.

    No doubt there remains work to be done. Interesting question is if it can be done on current architectures, or completely new ones are needed … and if so, what they might look like?

    Anyway, the ICECS conference was a very good experience indeed. I met a lot of nice and interesting people – both mixed-signal folks and others.

  4. Gabriele Manganaro

    About the comment “the DAC is not as difficult as the ADC”: interesting how I have too heard this comment in the past. In fact, this was relatively common belief in the late nineties when DACs operating at sample rate of the order of 100-200MSPS with output signals in baseband (1/4 of fs or much lower than that) were the state of the art.
    I am not sure how true that was at the time and I will leave the riddle to others. The fact is things have changed considerably. The state of the art is now an order of magnitude higher in output rate fs and with desired output signals possibly in second and higher Nyquist band.
    And guess what? In the meanwhile, advanced process technologies have helped a bit but not that much.
    So, I think that those who believes that designing DACs for this type of specs is easier than designing ADC should really give it a try.
    If they are right then they can easily enable “fully digital transmitters” (DAC driving the power amplifier directly for, say, multi-carrier GSM or LTE) making a lot of money and/or publish groundbreaking papers. But since this hasn’t exactly happened yet… 😉

  5. Yes, I see what you mean.

    The way you put it (“making a lot of money and/or publish groundbreaking papers“) certainly has a sweet and tempting sound to me … 🙂

    Do you think the way forward is to optimize analog design further, or mainly on invention new architectures or circuit topologies?

    Not being as firmly rooted in DAC as in ADC, my impression is that adding mixing modes (i.e. architectural innovation) has been one of the key advances in enabling better high-frequency performance lately. With all the “grinding” of the circuit-level design that must have been done over the years I assume that the analog optimization is pretty well lined up with its practical limits by now. Leaving circuit topology as the only option to architectural trickery. But with scaling of voltage swing there isn’t too much room for playing with topology anymore.

    Indeed, saying that “the DAC is easier” is not to paint the full picture.

  6. So I missed the ICECS conference… I would have needed some Greek sun considering the snow we have here now… As you say, you wouldn’t paint the whole picture by stating that “the DAC is easier”, but I think you would do quite a lot of it, at least the frame.

    I cannot really giev a much better view on why not more is published on the DAC side. After all, there is (still) quite some interesting things going on on the direct digital RF converter/modulator side. Both in terms of advances due to design process improvements and architectural aspects. A bit puzzling simply… well, I know a couple of Ph.D. students that will soon help to improve this ratio!

  7. @jjwikner:

    You shouldn’t have any regrets with respect to the Greek sun that you might have missed. I didn’t see any sun during the whole stay – perhaps in part due to attending the sessions held in windowless conference rooms. But when eventually venturing outdoors, the weather didn’t exactly make you think “where’s the nearest beach” either. The conference was good, though.

    It will be interesting to follow the output from your group. I’ve met quit a few of your students by now – all genuinely positive experiences from a personal point of view – and their scientific results shall be interesting to monitor.

  8. Pingback: Back from ICECS 2010 | Converter Passion


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s