Thanks for bearing with us – the data on which these calculations are based is correct, with the factor of ’27-times’ valid. The figures in the infographic have been corrected.

The report’s primary purpose is to focus discussion and (hopefully) stimulate much-needed and more extensive research into how we might meet industry forecasts of accelerating data traffic which, some say, will be exponential. It also considers future new consumer storage solutions and the challenges this might give the industry.

Regardless of increasing technological efficiency, the report questions whether newer energy efficient servers and transmission technologies can forever compensate predicted data traffic growth which many sources predict will become exponential – therefore it cannot be assumed that the increase in data traffic for the next 10 years can be compensated by more energy efficient networks.

The main issue – the energy/lightbulb hour calculation – is based on energy data published in the ‘Shipping To Streaming’ paper, duly referenced in our own report. This is one of the most authoritative studies available and serves as a reliable source of baseline data.

You can access that report here (http://dl.acm.org/citation.cfm?id=1851304&bnc=1); the download energy calculation used data in Table 3, p64 (Energy Costs: Non-Energy Optimized Streaming Method), the comparison with physical (CD) used data from Table 2 p63 Energy Costs: DVD Shipping Method.

Our report also references the fact that the energy consumption figures (figure 3, p13) show an energy consumption rate 10 times lower than those found in a previous and again widely acknowledged study conducted some years earlier (Koomey in 2004 and updated again in 2008 by Taylor & Koomey). In that respect, our report’s energy consumption figures are truly conservative and supports your view that technology is becoming more efficient every year.

Inclusion of uncompressed files in our analysis is, in our view, essential – as the report notes on p14, advances in network infrastructure speeds in excess of 100 mbps (wireless) mean that audio files would no longer need to be compressed in future, including new HD formats, which may be even bigger the current CD wav files.

Data used throughout the report was the most up-to-date at the time. The bulk of the research was carried out as noted in the report’s Assumptions, remembering that it was based on a small part of extensive three-year future-looking project that began in 2009.

You raise some interesting points, not least in your second response concerning alternative energy consumption figures based on Google research. Whilst we aren’t in a position to question these (nor would it be appropriate to), it highlights the lack of sufficient baseline data, for what is a very complex topic with multiple touchpoints. Such studies are entirely dependent on a large number of variables, some of which you allude to in your post, with other studies (e.g. Google) receiving any number of interpretations and challenges – from the FT to the paper industry (http://www.paperonline.org/uploads/MR_bad%20for%20climate.pdf).

We’re pleased and grateful that you’ve raised important questions – our hope is that it will fuel further analysis and inspire debate in addressing these future challenges.

Kindest regards on behalf of the MusicTank team.

Jonathan Robinson
Programme Director, MusicTank

Interesting – thanks for the feedback and note about the typo. The numbers do square better at 88 minutes, although it does perhaps leave a question over the general “light bulb” measurement. It’ll be interesting to see the sources for those figures.

For example, the quoted time of 38 light bulb hours to manufacture and ship a CD suggests an energy requirement of 380 Wh, which seems low.

The closest comparison I could find is Google’s figure that it takes 3KG of CO2 to produce and ship a DVD. Turning their figure back into watts using the DECC standard of 0.542 kg CO2 per kWh we get 5.5 kWh.

I only mention that because, turning their figure back into light bulbs, it would suggest 550 hours to produce their DVD against 38 hours to produce the MusicTank CD. Perhaps you have a more efficient disc press or distribution network?

Either way, it puts the difference between physical and virtual distribution back into the 300-400 times less efficient range, rather than in the tens.

Cheers,
Grant

Thanks for taking the time to go through this. You have picked up a typo in the report which we are in the process of amending. You are absolutely correct, an uncompressed 12 track WAV download consumes 88 “light bulb” minutes, not seconds.

One of our team will be in touch again with full references for the numbers used, and an answer to your other points. Please bear with us on this, but for the time being be assured the comparison is correct. Thanks very much!

Best

Sam
Associate Director, MusicTank