Inspired by the, now iconic, graph released on the 4th of July 2012, I began a sonic exploration of the data by sweeping the value of mass from 100 to 150 (at a rate of 1 GeV per second) and listening to the distribution of values for one parameter at a time over that range of masses.
- Can you hear a change in the distribution of any of the parameters around 126-128 GeV? (And perhaps even more interestingly, can you hear unusual distributions at other masses?)
- In all of the files you can hear “clusters” of commonly occurring values and rare “outliers” that pop out now and then. Do you think the outliers are noise? Or is it more likely that the clusters are the background noise and it is the outliers that point to the interesting events?
- Could the sound sweeps be used to scan for other “interesting” areas in the range of masses?
Listening hint: Try fixing your gaze on the horizontal fader showing the mass value while you use your ears to listen for changes in the distribution of pitches. Whenever you hear a change or an unusual distribution, make a note of what the mass was when you heard that sound.
GeV: This one is a trivial mapping of mass itself, just to help you get oriented. As expected, you hear the pitch rising linearly with time (with mass).
delR: This one starts out with a triangular shape and has a tendency to move higher and flatten out for higher values of mass. On the second playing, the higher values are filtered out in order to focus on the lower values to hear if anything special happens around 126-128 GeV.
pTt: On the second playing, the range of values is ‘filtered’ to eliminate the large cluster of values and leave the more rare values. Some large ‘outlier’ values start popping up around 139 GeV and higher.
g1pt: Three times through the data, the second time with a higher base pitch, the third time with quarter-step spacing between bins.
Is there anything significant about 103 GeV, 120 GeV, 131 GeV?
g2pt (and g2pt with a lower bass pitch): In this one, you can hear an overall tendency for the pitches to get higher with increasing mass. The distribution of values also seems to get flatter with larger mass.
Sounds for Gilles Jobin
CERN choreographer-in-residence, Gilles Jobin, requested some sounds for his presentation in which his dancers were illustrating movements inspired by symmetry and non-contact forces.
delR: 827 events ranging from 125 to 127 GeV (1/100th GeV per 1 second of time).
Stereo Gamma: Gamma-1 in the left channel and Gamma-2 in the right: 827 events ranging from 125 to 127 GeV (1/100th GeV per 1 second of time).
How was this sonified?
One way to think of these sounds is as “animated” or dynamic distributions of values with respect to GeV. On each time slice, you hear the distribution of values for one of the parameters at that GeV level.
There are 21768 lines in the data file; each line contains 5 different parameters: mass (in GeV), delR, pTt, g1pt, g2pt. Let’s say you were curious about the values of delR in only those events where mass = 126. You would scan through the file looking for lines where the first column (mass) was 126 and, on those lines, read the value of column 2 (delR). You’d end up with a range of values from a minimum to a maximum.
Next, take that range from min to max, divide it into 100 equal-sized “bins” and drop the values one-by-one into those bins. By the end of this process, some of the bins have more numbers in them than others. The bins corresponding to the most commonly occurring values have lots of numbers in them and the bins corresponding to rare values have fewer numbers in them (and some of the bins are empty).
Now imagine assigning each bin a different pitch and using the number of values in each bin as an amplitude. You end up with a ‘chord’, a ‘spectrum’ or a ‘timbre’ of 100 pitches, some louder, some softer, and some completely silent. Next, imagine repeating this process for each value of mass from 100 to 150. At the end you have a 50 second sound where the spectrum changes once per second (incrementing by one second each time you increment the value of mass).
It might be interesting to map one of the other parameters (other than mass) to time in order to listen for possible relationships between pairs of parameters at a fixed or small mass range). For example, here are some visual graphs of the values of pairs of parameters, restricted to the range of mass = 126-127.