Where we’re at…

I thought maybe a ‘year in review’ as regards to listening to music might be in order. It has been almost a year since I wrote and started using the ‘Absolutely Baching Music Player’ or AMP, beginning on January 9th 2021. On June 1st, AMP was superseded by Giocoso. But, using one or the other, I’ve been playing my music in more-or-less randomised form for the best part of a year.

I say ‘more or less’, because -by means of Giocoso’s run-time switches- I’ve been tweaking and influencing what ‘random’ means. For example, when I started playing things in Giocoso on 1st June, it turned out that some 87% of my music collection was listed as having not been played in either player. That is, AMP or Giocoso between them only had records of having played a mere 13% of my music collection.

But that’s quite true not as far as my ears are concerned: I’ve been playing digital music for about 20 years before AMP or Giocoso came along, so a good proportion of that ”87% unplayed’ has indeed been played (in things like Foobar2000 or MusicBee), but there’s simply no record of them having been played. But the 87% figure nevertheless bugged me and I therefore decided to set out, deliberately, to reduce it… by making sure Giocoso only played music in future which was not recorded as having previously been played by Giocoso or AMP. (In the process, I should point out that I have encountered the delights of Holmboe and Vasks, Schulhof and Antheil, and a load of other ‘obscure’ stuff I’ve long-owned but seldom played, in any player: the past few months have been a wonderful ride through the unfamiliar).

Thus, for the past few months, I’ve run Giocoso with the ––unplayed switch, to ensure that each new play is of music that neither AMP nor Giocoso has a record of having played before.

Accordingly, here’s a graph of the number of unplayed recordings in my collection, since June 2021, together with some other data I’ll explain as we go along:

The left axis is ‘percentage of collection not yet played’, recorded by the blue line of dots: it starts on 3rd June at the 87% mark; it decreases rapidly until, at around November 1st, it’s showing only 54% of the collection has not yet been played. Put another way, it shows that I’ve spent five months listening only to ‘previously unplayed’ recordings, and managed to reduce the amount of my collection that is not recorded as having been played by around 33%.

The right axis of the graph is ‘number of new recordings added (per day) to the collection’, recorded by the height of the orange bars.Thus, you’ll see that on August 1st, I managed to add a new 64 recordings to my collection; on 27th August, I managed to add another 67 new recordings; on 14th September, a further 64 were added; and so on. I think the average added over the entire period covered by the graph is around 11 new recordings per day -but, clearly, I was doing stuff-all cataloguing in June and July, but got rather busy thereafter!

As you can see, the blue line started June and July by plunging nicely! That’s because I was not only using the ––unplayed switch when running Giocoso, but also specifying the ––maxduration=10 one, to ensure that Giocoso only selected to play the shortest previously-unplayed recordings: naturally, when you only play short pieces, you can play a lot of them per day. Hence, the blue line of unplayed recordings plummets in a satisfyingly near-vertical manner. With maxduration set to 10, you can get at least 6 new pieces played per hour.

At the beginning of August, however, I’d run out of very short pieces to play, and thus eased the maxduration switch back to 20 minutes -which means we’re down to playing only 3 new pieces per hour. Naturally, the slope of the blue line is likely to flatten out to have a slope about half that which applied during the ‘up to 10 minutes’ times. Accordingly, you see the blue line has a distinct ‘kink’ in it, around the start of August.

The straight orange line doing a diagonal descent that disappears around the start of November is a ‘line of best fit’ for the blue line. It says that, based on the entire number of recordings I’ve listened to since June 1st, I should be down to 50% unplayed versus 50% played by around November 3rd. Well: it’s now November 5th and clearly I haven’t reached that point yet! The trouble is that the line of best fit isn’t aware that there’s really two lines in blue: the one up to August where I’m playing all the sub-10 minute recordings and thus churning through the plays at a rate of knots; and the other line from the start of August onwards, where I’m listening to recordings up to 20 minutes in length and thus getting through maybe half the recordings per day I was managing to play previously. One line of best fit applied to both playback modes is bound to be unduly influenced by the artificially high rate of plays recorded in June and July.

So here’s a more realistic update:

This is the same data, but only displaying the post-September music plays and catalogue additions. Since everything played during this period matched the ‘under 20 minutes’ rule, the slope of the blue ‘plays’ line is reasonably consistent over the entire time of the graph. The orange diagonal ‘line of best fit’ accordigly now seems to predict that I’ll finally have played 50% of my music collection by December 3rd or 4th -and I suspect that’s reasonably accurate.

Once I finally hit the ‘50% of your collection has been played’ mark, I think I’ll take off all time constraints and simply say to Giocoso ‘play me things previously unplayed, regardless of length’. It will mean that I might suddenly find myself in the middle of Wagner’s Ring (playtime=~16 hours!) and the slope of the blue line will be practically horizontal in consequence… but I don’t think, at that point, it will really matter. I mean, by the time I die, I’d like to have evidence that I’d actually listened to all my music collection, at least a couple of times! But I’m fairly happy that the last 50% takes care of itself over time, without me forcing it into specific duration slices.

The other factor that might yet push out the ‘when will I reach 50%’ point is the fact that I continue to acquire unfeasibly large amounts of new recordings to add to my collection! As of now, I’ve added a cumulative total of 1,665 new recordings to my collection, since June 1st. Remember that in these parts, a “recording” isn’t the same thing as a “record” or a “CD”. One CD might contain half-a-dozen “recordings”, or even more. Thus, a new CD of Monteux’s Symphonie fantastique will actually be catalogued as four recordings, because the relevant CD also comes with recordings of an overture to Les Troyens, the Damnation of Faust and the Benvenuto Cellini overture. So, 1665 new recordings probably represents roughly 400 CDs (but I’m guessing). I have maybe another 300 CDs to catalogue before I’m all caught up (mostly new Haydn symphony recordings) -but, obviously, the addition of 300 new unplayed CDs and their various ‘recordings’ to the collection will push out the date by which I may be expected to have listened to half of my entire collection!

There’s also an additional factor which will muck up the maths: November 22nd is Benjamin Britten’s birthday and the tradition in these parts is to spend the entire day playing nothing but Britten, by way of acknowledgement of one of the towering geniuses of 20th Century Music. That is a break from the ‘only previously unplayed’ rule and could thus result in previously-played music being re-played. Alternatively, Giocoso might choose to play some of his operas, which last an hour or more each… which will therefore result in a reduction in the unplayed ‘batting average’ for the day.

Anyway: long story short, I suspect I may finally and properly hit a 50% mark sometime in the early new year, provided only I stop buying new CDs in the meantime!

I realise that analysing one’s listening habits in this way might not be for everyone! The nice thing I find about Giocoso, however, is that it at least allows this sort of analysis if you’re in to it. If you’re not, it just plays music and you needn’t ever know it can report listening statistics 🙂 Giocoso is all about giving you what you want from your music collection, light or heavy, as you choose.


Leave a Reply

Your email address will not be published. Required fields are marked *