Altars of Mathness

What could be duller than digits? They just sit there on the page or screen, mindlessly marking mathematics:

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100…

But perhaps they become more interesting as images. Let’s display the final digit of the integers, or counting numbers, on a graph. Running left-right and up-down, the graph represents the final or rightmost digit of 1, 2, 3, … 10, 11, 12, 13, … 100, 101, 102, 103, …, 1000, 1001, 1002, 1003, …:

Rightmost single digit of the integers (click for larger)


No, that’s still dull: the graph just generates endlessly repeating triangles. After all, the final digits fall into a cycle: 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3… So do the final two digits: 1, 2, 3, 4, 5, […] 94, 95, 96, 97, 98, 99, 00, 01, 02, 03… Here they are as a graph:

Rightmost two digits of the integers


Now the triangles look like waves sweeping to shore. That’s a bit more interesting, but not much. So let’s try something different. The trailing digits of the integers generate triangles, so let’s see what the triangular numbers generate. The triangular numbers — 0, 1, 3, 6, 10, 15, 21… — are very simple to form. You just sum the integers: 1, 3 = 1 + 2, 6 = 1 + 2 + 3, 10 = 1 + 2 + 3 + 4, 15 = 1 + 2 + 3 + 4 + 5, 21 = 1 + 2 + 3 + 4 + 5 + 6, 28 = 1 + 2 + 3 + 4 + 5 + 6 + 7, 36 = 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8, 45 = 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9, 55 = 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10… Here are the final digits of the triangulars — 1, 3, 6, 0, 5, 1, 8, 6… — as a graph:

Final digit of triangular numbers in base 10 (click for larger)


Now something interesting has appeared. The final digits form a repeated palindromic pattern (counting 0 as the zero-th triangular number):

0, 1, 3, 6, 0, 5, 1, 8, 6, 5, 5, 6, 8, 1, 5, 0, 6, 3, 1, 0, 0, 1, 3, 6, 0, 5, 1, 8, 6, 5, 5, 6, 8, 1, 5, 0, 6, 3, 1, 0, 0, 1, 3, 6, 0, 5, 1, 8, 6, 5, 5, 6, 8, 1, 5, 0, 6, 3, 1, 0, 0, 1, 3, 6, 0, 5, 1, 8, 6, 5, 5, 6, 8, 1, 5, 0, 6, 3, 1, 0, …

An Altar of Mathness created by the final digit of triangular numbers in base 10


And those palindromic digits create symmetric shapes that remind me of little altars — let’s call them “altars of mathness” in tribute to Morbid Angel’s genre-defining album Altars of Madness (1989). And what about the final two digits of the triangular numbers? Here’s the graph (adjusted so that 99 fits into the same space as 9):

Final two digits of triangulars in b10


Final two triangular digits in b10 (horizontal scale compressed)


The final two digits form palindromes too. And this time we don’t get just triangles, but curves too. But that’s in base 10. What happens with the trailing triangular digits in other bases? Well, here’s the final triangular digit creating more altars of mathness in different bases (note that the altars are more elaborate in even bases):

Final triangular digit in base 4


Final triangular digit in b5


Final triangular digit in b6


Final triangular digit in b7


Final triangular digit in b8


Final triangular digit in b9


Final triangular digit in b14


And here’s the graph for the final triangular digit in base 100:

Final triangular digit in b100


The graph for final single digit in b100 should look familiar, because it’s identical to the graph for final double triangular digits in b10:

Final two digits of triangulars in b10


That’s because two digits in b10 are equivalent in one digit in b100, four digits in b10 are equivalent to two digits in b100, and so on. But b100 can’t capture three digits in b10 (the graph is again adjusted so that 999 fits into the same space as 9 and 99 above):

Final three triangular digits in b10


If you compress the x-axis for that graph, you can see how long the symmetries are:

Final three triangular digits in b10 (x-axis / 2)


Final three triangular digits in b10 (x-axis / 4)


The final four digits of the triangulars in b10 create even longer symmetries:

Final quadruple triangular digits in b10


Final quadruple triangular digits in b10 (x-axis / 2)


Final quadruple triangular digits in b10 (x-axis / 8)


Note how, as the length of the final digits rises, you need to compress the x-axis more and more to see the symmetries. But integer sequences obviously don’t end with the counting numbers and triangulars. What about squares and powers of n? What about primes and Fibonacci numbers? Here’s the final two digits of the squares — 1, 4, 9, 16, 25, 49, 64, 81, 100, 121, 144, 169… — in b10:

Final two digits of the squares in b10


It’s reminiscent of the triangular numbers (so are the final-digit graphs for other polygonal numbers). So what about the powers of 2? That’s 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024… Here’s the graph for final single digits of 2^p in b10:

Final single digits of 2^p in b10


This time there’s repetition, but not symmetry. Here’s the graph for final double digits, or 2-digits, of 2^p in b10:

Final 2-dig of 2^p in b10


Now the graph looks a little like a range of eroded mountains. Now try dig-4, the final four digits of 2^p in b10:

Final 4-dig of 2^p in b10


The patterns are similar to those of dig-2 and don’t need compressing in the x-axis. This similarity and lack of need for compression are true of any number of final digits in 2^p. The final 10 digits look like this:

Final 10-dig of 2^p in b10


And the final 20 and 30 digits like this:

Final 20-dig of 2^p in b10


Final 30-dig of 2^p in b10


Powers don’t behave like polygonals: the finals are fractals. That is, the final digits create similar patterns at all scales: 1-dig, 2-dig, 10-dig, 100-dig, 1000-dig and so on. That’s true in other bases:

Final 5-dig of 3^p in b2


But a glimpse of b2 is all you’re going to get of other bases. There are other fish to fry — Fibonacci fish. The Fibonacci sequence, whose terms are equal to the sum of the previous two numbers (after seeding with “1, 1”), starts like this: 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584, 4181, 6765, 10946, 17711, 28657, 46368, 75025, 121393, 196418, 317811… And what about the graphs for final fib-digits? As you’ll see, final Fib-digits are fractal too. Indeed, Fibonacci final-graphs look like 2-power final-graphs (in a way, Fibonacci numbers are powers of φ = 1.6180339887498948482…). The patterns are similar at all scales. And they remind me of the skyline of a ruined city in an Oriental tale, with collapsed domes and crumbling minarets:

Final 1-dig of Fibonacci numbers in b10


Final 2-fibdig in b10


Final 3-fibdig in b10


Final 4-fibdig in b10


Final 5-fibdig in b10


Final 10-fibdig in b10


Final 15-fibdig in b10


Final 20-fibdig in b10


Final 25-fibdig in b10


So final fibdigs are fractal. But final prime digits aren’t:

Final 1-digit of primes in b10


Final 1-digit of primes in b5


Final 2-digit of primes in b10


Primes aren’t final-digitally fractal like Fibonaccis and powers of 2. But there’s occasional symmetry in the prime fin-digs. I’ve marked some palindromic patterns in red and green:

Palindromic patterns in final 1-digits of the primes in b10 (click for larger)


The palindromic patterns, or pal-pats, in the primes look like the altars of mathness in the triangulars. They’re created by digital palindromes like these:

19, 23, 29 (c=3)
347, 349, 353, 359, 367 (c=5)
937, 941, 947, 953, 967, 971, 977 (c=7)
1951, 1973, 1979, 1987, 1993, 1997, 1999, 2003, 2011 (c=9)
26423, 26431, 26437, 26449, 26459, 26479, 26489, 26497, 26501, 26513 (c=10)


Here are the first few pal-pats in the primes (note that 157, 163, 167 and 163, 167, 173 overlap):

2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 101, 103, 107, 109, 113, 127, 131, 137, 139, 149, 151, 157, 163, 167, 173, 179, 181, 191, 193, 197, 199, 211, 223, 227, 229, 233, 239, 241, 251, 257, 263, 269, 271, 277, 281, 283, 293, 307, 311, 313, 317, 331, 337, 347, 349, 353, 359, 367, 373, 379, 383, 389, 397, 401, 409, 419, 421, 431, 433, 439, 443, 449, 457, 461, 463, 467, 479, 487, 491, 499, 503, 509, 521, 523, 541, 547, 557, 563, 569, 571, 577, 587, 593, 599, 601, 607…

And are there palindromes among the final 2-digits, 3-digits and higher n-digits of the primes in different bases? Yes, you can easily find some. But I haven’t put them on a graph yet:

Base 10 (2-dig)

58789, 58831, 58889 (c=3)
286873, 286927, 286973 (c=3)
360649, 360653, 360749 (c=3)
404851, 404941, 404951 (c=3)
590437, 590489, 590537 (c=3)
623071, 623107, 623171 (c=3)
651517, 651587, 651617 (c=3)


Base 6 (2-dig)

300335, 300401, 300441, 300501, 300535 (c=5) (23459 to 23531 in base 10)
1030255, 1030331, 1030351, 1030431, 1030455 (c=5) (50651 to 50723 in b10)
1140451, 1140501, 1140521, 1141001, 1141051 (c=5) (59791 to 59863 in b10)
1402451, 1402545, 1403031, 1403045, 1403051 (c=5) (78367 to 78439 in b10)
1435431, 1435451, 1435505, 1435551, 1440031 (c=5) (82891 to 82963)
2400505, 2401001, 2401015, 2401101, 2401105 (c=5) (124601 to 124673)
2442235, 2442311, 2442351, 2442411, 2442435 (c=5) (130127 to 130199)
2444215, 2444225, 2444311, 2444325, 2444415 (c=5) (130547 to 130619)
2533105, 2533121, 2533215, 2533221, 2533305 (c=5) (136769 to 136841)


Base 4 (3-dig)

20013013, 20013133, 20020013 (c=3) (33223 to 33287 in base 10)
21031111, 21031303, 21032111 (c=3) (37717 to 37781)
22310011, 22310333, 22311011 (c=3) (44293 to 44357)
33030121, 33031001, 33031121 (c=3) (62233 to 62297)
102031333, 102032131, 102032333 (c=3) (74623 to 74687)
110013121, 110013311, 110020121 (c=3) (82393 to 82457)


Base 3 (3-dig)

112121020012, 112121021211, 112121022021, 112121100211, 112121101012 (c=5) (287393 to 287501 in base 10)
202002212002, 202002212101, 202002220001, 202002221101, 202010000002 (c=5) (395741 to 395849)
1001012111212, 1001012112202, 1001012121022, 1001012121202, 1001012122212 (c=5) (555143 to 555251)
1010112112012, 1010112112201, 1010112120222, 1010112121201, 1010112200012 (c=5) (601079 to 601187)
1011202211211, 1011202212212, 1011202220022, 1011202221212, 1011202222211 (c=5) (625369 to 625477)


Base 2 (5-dig)

101110001111101, 101110010000111, 101110010001001, 101110010100111, 101110010111101 (c=5) (23677 to 23741 in base 10)
10000001000111101, 10000001001011001, 10000001001110001, 10000001001111001, 10000001001111101 (c=5) (66109 to 66173)
10111100110011011, 10111100110011111, 10111100110111001, 10111100110111111, 10111100111011011 (c=5) (96667 to 96731)
11010000111110001, 11010001000001101, 11010001000011001, 11010001000101101, 11010001000110001 (c=5) (106993 to 107057)

And I conjecture that you’ll get palindromes for any number of final digits in all bases. And can these palindromes be of arbitrary length? Again, I conjecture so. There are infinitely many primes and very rare patterns can occur infinitely often in an infinite set of numbers.


Post-Performative Post-Scriptum

Here’s Dan Seagrave’s classic cover for Morbid Angel’s Altars of Madness (1989):


Morbid Angel — official website
Dan Seagrave — official website


Elsewhere Other-Accessible…

Formulas Focal to the Flesh — a pre-previous post paronomasizing the title of a Morbid-Angel album…

The Darling Duds

What is a Darling Dud? It’s my name for a band that meets two simple criteria: 1) I like them (hence “darling”); 2) they aren’t as well-known as I think they should be (hence “dud”). I based the name on…

The Darling Buds

A Welsh female-fronted jingle-jangle indie band who are, for me, the archetypal darling duds. I like them a lot and I think they should have been much more successful. But if they had been, I might not enjoy their melodic music as much.

The Darling Buds at Bandcamp


The Primitives

An English female-fronted jingle-jangle indie band who I like a lot and who I think should have been much more successful. The Guardian said of them: “The Primitives is a great, great name for a group, and barely a day goes by when I don’t lament the fact that it was wasted on brittle little one-hit indie wonders from Coventry with a fifth-rate Debbie Harry wannabe for a singer. There oughta be a law against it.” As so often, the Guardian gets it badly wrong. In fact, Tracy of the Primitives was a third-rate Debbie Harry wannabe. But she was more attractive than Debbie Harry, which perhaps explains the vituperation in the Guardian.

The Primitives


Compulsion

Kinda punk, but with much more musical subtlety and lyrical intelligence than that label usually suggests. Why weren’t they more successful? I don’t know, but two things occur to me. They’re obtrusively loud on record in a way that I think detracts from that subtlety and intelligence. And they looked old in their publicity photos. With less volume and fresher faces, they might have done better.

Compulsion in shades (Wikipedia)

Compulsion


David Tyrrell

Perhaps the most undeservingly unsuccessful of the lot, because you’ve never heard of him and he’s much better than lots of people you have heard of. Which is not to say he’s an undiscovered musical genius, but I like his 2008 album Substance a lot. I think it was self-released. I know it should have done much better. It’s catchy like Compulsion, but quieter and Tyrrell does something unusual in popular music. He sings clearly, so you can understand the lyrics.

David Tyrrell song at Youtube


Morbid Saint

Here’s a heretical thought. I don’t think Slayer are the real Slayer. I think the real Slayer – the real kings of crushing, red-in-tooth-and-claw ’80s metal – are Morbid Saint. They sound more brutal and more evil than Slayer. They play thrash metal and make it rage like death metal. So why didn’t they get the success they deserved? The delayed release of Spectrum of Death (1989) can’t have helped. Nor can the ludicrous cover. And yes, they’re obviously and heavily influenced by Kreator. But still: they deserved a lot more than they got.

Morbid Saint


Beach Riot

“Fuzz pop” they called their music. It was loud and bouncy, with alternating male-female vocals, and was a lot of fun. But after releasing a few singles and an EP, they disappeared. Shame. Also a shame is that some of their songs come in two versions: a version with energy and a version without.

Stop-Press: No, Beach Riot haven’t disappeared and have released their first album. Or something.

Beach Riot


Obiat

One way of translating the Polish word Obiat is “funeral feast.” And one way of describing Obiat’s music is “stoner-doom.” But translation and description fail to capture the full meaning and the full music. Obiat can be very heavy, but they can also be very quirky. In short, expect the unexpected. Trying to define Obiat’s music is like trying to herd cats. So it’s appropriate that one of their songs has guest vocals from a cat. And look at the cover for Accidentally Making Enemies (2002). What does it mean? Why choose a sunken speed-boat? I don’t know, but I like the cover and I like Obiat.

Obiat


Feline

Female-fronted rock from the 1990s with a good name, because there’s mystery and elegance in the music on their first album, Save Your Face (1997). Melancholy too. And menace. Velvet paws + razor claws. But they were never very successful. Grog, the female fronter of Feline, has soldiered on with Die So Fluid, whose music I also like. But it’s more metal and doesn’t have everything that Feline’s had, particularly not the mystery and the melancholy.

Feline / Die So Fluid


Split Enz

The nucleus of Crowded House. Split Enz were big in New Zealand, moderately successful overseas. I prefer them to Crowded House because their music is simultaneously more varied and, in a good way, more insular. New Zealand is an island nation, after all. The catchiness and melodies were there from the start, though.

Split Enz


The Chills

Another New Zealand band. They were like Split Enz, but more so: fairly big at home, moderately successful overseas. They had melodies and catchiness too, but they were more musically unusual than Split Enz. The late Martin Phillips was the mainstay and the motor of that. He was self-taught and his music had an alien, outsider edge to it, as though he’d taught himself by listening to fuzzy, fifth-generation pirate tapes of the Byrds, Velvet Underground and XTC whilst living in a hut deep in the rain-forests of the South Island. Or even in an oxygen-tent on Mars.

The Chills


The Heartbreaks

English indie-rockers who rose like a rocket with their debut, Funtimes (2012), and fell like the stick with the follow-up, We May Yet Stand a Chance (2014). Some invoke the curse of Morrissey, which dooms bands that Morrissey praises or takes on tour, but in fact no supernatural explanations are needed. Funtimes had some very good songs and We May Yet Stand a Chance had no good songs at all.

Afterword: Or so I thought when I first heard the two albums. I’m coming round to We May Yet Stand a Chance much more now, but a slow-burning second album would explain their fall too. Funtimes is immediately catchy indie rock. I thought: The Smiths. We May Yet Stand a Chance is trying to be sophisticated. I thought: Sinatra. Which wasn’t good. And the cover was a hostage to fortune too.

The Heartbreaks at Youtube


Anna Pingina

A Russian singer singing in Russian, which explains some of why I don’t think she’s been as successful as I think she should have been. She isn’t experimental or unusual in any way, but she can write attractive melodies and she sounds folky without sounding fey or feeble.

Anna Pingina


Necros Christos

I thoughtlessly assumed from their name that Necros Christos were Greek when I first heard them. So I rated their music higher than I did when I subsequently learned they were in fact German. That’s because it seemed competent, power-packed and intelligent in a way I don’t associate with Greek bands but do associate with German bands (which is naughty of me). Perhaps other people think the same way and N.C. would have been more successful if they’d been Greek. It’s hard to explain their relative unsuccess otherwise, because they had a distinctive sound, apparently sincere occult obsessions, and were, as I said, competent, power-packed and intelligent.

Necros Christos


Nubes en mi Casa

Years ago I downloaded a lot of free MP3s, listened to them, deleted the ones I didn’t like, then listened on-and-off to the rest. “Mareo” by Nubes en mi Casa was one of the ones I liked and kept. But I didn’t notice the sweetly surreal name of the band (“Clouds in my House”) or the true quality of the music until I was listening to a load of MP3s on random play one day. Then the power of contrast came to its rescue. After a lot of stuff I recognized at once and more or less enjoyed, “Mareo” started playing. I thought: “Hold on, what’s this? It’s good!” You could describe it as wistful indie. You could also describe it as wet indie. But I like it a lot and I hunted down more by Nubes en mi Casa, who were a female-fronted Argentinian band with Spanish lyrics. That explains at least part of their unsuccess.

Nubes en mi Casa


Chant of the Goddess

Brazilian stoner-doom metallers whose first album is an excellent illustration (audistration?) of a simple fact of auditory psychology: loud is louder when it’s mixed with soft. Chant of the Goddess go from quiet to cacophonic in a compelling way. Or they do that on their first album, at least. Their second album doesn’t grab me in the same way.

Chant of the Goddess


Red Eye

Spanish stoner-doomers who quote Lovecraft, use Old English, and play music that’s both powerful and intelligent. So why hasn’t that music had all the success I think it deserves? I see one obvious reason: “Red Eye” is a bad name. To 21st-century Anglophones it goes most naturally with jet-travel, not gigantic sounds. Were they translating Ojo Rojo? That means the same thing in Spanish and would have been better. In fact, they could have gone with rOjO as a logo. I don’t like their album covers either. But I do like their music.

Red Eye


16Volt

Kind of a cross between industrial metal, emo and indie. Nine Inch Nails territory. But I don’t like NiN and I do like 16Volt. I don’t like everything they’ve done or even most of what they’ve done, but what I like, I like. My first listen made me wish I were a teenager in sunny California in the 1980s or ’90s, which is not something that’s ever happened to me before. Onomastic psychology explains some or all of their unsuccess, I’d say. “16Volt” just sounds feeble. 16 is not just too small a number but too easily divisible into even smaller numbers: 16 → 8 → 4 → 2 → 1. Using a prime would have been better: “23Volt” or “37Volt”.

16Volt


Owlcrusher

A three-piece from Northern Ireland who really whip up a storm with their take on blackened doom. That’s black metal + doom metal. So they crush genres together in the way that their name crushes concepts together.

Owlcrusher


Akelei

Dutch doomsters centered on the ever-present Misha Nuis. They play meandering melancholy music that’s often very loud and sometimes very gentle. Perhaps the gentleness explains some of their unsuccess, but two obvious things come before that: their name and their lyrics. They sing exclusively in Dutch and their Dutch name means nothing to Anglophones. It’s actually the name of a flower, columbine or aquilegium, which is a quirky choice. And I like it. Singing in Dutch is a quixotic choice. And I also like it:

De reis gaat door met lenig hart
En zonder verwachtingen
Wij raakten allengs ver van huis
Alles is anders nu
Oud licht helpt ons aan nieuw inzicht
Onthult al wat komt hierna

Akelei’s “Dwaaluur” (Wandering-Hour)

The journey goes on with a shifting heart
And without expectations
We slowly drifted far from home
Everything is other now
Old light helps us to new insights
Reveals all that comes next

Akelei want to go their own way, not chase popularity. And their meandering melancholy reminds me of more depressive art from the Low Countries. It’s a book of 1892 by the Belgian writer Georges Rodenbach (1855-98). It’s called Bruges-la-Morte or Bruges the Dead City, it’s illustrated in melancholy monochrome, and it too wanders and westers and woes:

Le jour déclinait, assombrissant les corridors de la grande demeure silencieuse, mettant des écrans de crêpe aux vitres. Hugues Viane se disposa à sortir, comme il en avait l’habitude quotidienne à la fin des après-midi. Inoccupé, solitaire, il passait toute la journée dans sa chambre, une vaste pièce au premier étage, dont les fenêtres donnaient sur le quai du Rosaire, au long duquel s’alignait sa maison, mirée dans l’eau. Il lisait un peu : des revues, de vieux livres; fumait beaucoup; rêvassait à la croisée ouverte par les temps gris, perdu dans ses souvenirs. Voilà cinq ans qu’il vivait ainsi, depuis qu’il était venu se fixer à Bruges, au lendemain de la mort de sa femme. Cinq ans déjà ! Et il se répétait à lui-même : « Veuf! Être veuf! Je suis le veuf! » Mot irrémédiable et bref! d’une seule syllabe, sans écho. Mot impair et qui désigne bien l’être dépareillé.

Some melancholy monochrome from Bruges-la-Morte (1892)

The day was fading, darkening the corridors of the large, silent house, laying screens of crepe on the windows. Hugues Viane readied to go out, as was his daily habit as the afternoon faded. Idle, solitary, he spent all day in his room, a vast room on the first floor whose windows overlooked the Quai du Rosaire, along which his house lay, reflected in the water. He read a little: magazines, old books; smoked a lot; daydreamed at the window open on to gray weather, lost in his memories. He had been living like this for five years, ever since he came to settle in Bruges, the day after his wife’s death. Five years already! And he repeated to himself: “Veuf! Widower! To be a widower! Je suis le veuf!” An irremediable word, so brief! A single syllable, without echo. An odd word, and one that well captures this mismatched creature.

Akelei


Gull-Om, Gull-Un

Cover of Variations on a Theme (2005) by Om


Cover of Yr Wylan Ddu (1996) by Slow Exploding Gulls


Elsewhere Other-Accessible…

Om Vibratory — Om’s official site
Mental Marine Music — an introduction to Slow Exploding Gulls

Sanctisonic Symmetry

Holy Fawn, Realms EP (2015)


I like the symmetry and simplicity of this cover, though I think the style and color of the text could be improved on. Here are two variants on the cover:

Cover without text


Cover with mirrored text


Elsewhere Other-Accessible…

Holy Fawn at Bandcamp