Blog

RSS
  • More recognised samples: Darren Korb's Bastion Soundtrack / Enigma's "Seven Lives,…

    Nov 20 2011, 4h31

    I'm not the first to notice this, but a few of the samples used in the Bastion soundtrack (2011) by Darren Korb are strongly recognisable as being also used in the earlier album Seven Lives, Many Faces (2008) by Enigma.

    As the Bastion soundtrack is primarily a digital download (with limited edition physical CDs), I've had no luck finding any sort of liner notes which would give credit to samples used.

    It's very possible that these are royalty-free or public domain samples, in which case I'd be curious to know which sample kit/compilation they might've come from.

    Distorted guitar loop:
    0:15 in Downtown Silence
    3:41 in Bynn the Breaker

    Solo strings refrain:
    0:00 in Touchness
    1:39 in Pale Watchers

    Strings and drum loop:
    0:24 in Seven Lives
    0:28 in From Wharf to Wilds
  • Duplicate artists in different languages

    Mar 12 2009, 10h49

    I've noticed for years that there exist multiple ways of writing some artists' names and some track titles; one has only to look at anime soundtracks and any names/titles which are written differently depending on the preferred/active language of the writer.

    Entire anime soundtracks can exist here in duplicate (or more) because some people tag their files with romaji, while others use the original Japanese text (hiragana, katakana and kanji). A random example would be Chiisana Te no Hira. This could be written in hiragana as "ちいさなてのひら" (although that isn't the proper way to write it), or it could be "小さなてのひら" (which seems to be the most correct form that I can find here, and is probably the right one; but, not being a native speaker of Japanese, I can't determine this by myself). And yet it seems unfair to expect users who cannot view Japanese text to tag all of their Japanese tracks correctly. On the other hand, it would be an insane undertaking for anyone to try and set up autocorrection of tracks from romaji to kana/kanji...

    Speaking more generally, those who use entirely different writing systems end up contributing to duplicates of artists. As just one example (among countless others), there exists the well-established Antonio Vivaldi, and then there also exists Вивалди, which is just "Vivaldi" transliterated in Cyrillic. Or, to use another example, there's the difference between English and German transliterations of names that are neither English nor German (a common example being Russian or other Slavic names). The sound that is transcribed as "y" in English is instead written as "j" in German; this can lead to duplicate artists (in addition to the original name written in its language of origin!). A similar correspondence exists between English "z" and German "s"... and so on.

    Can all users be expected to understand these linguistic differences when tagging their music, never mind appreciate them? No, of course not. I just wonder whether there can ever be a way to at least semi-automate the discovery of duplicates arising from the cases described above.
  • Finally found the Samurai Shodown (Samurai Spirits) movie ending theme!

    Jun 10 2008, 10h30

    I stumbled across this song a very, very long time ago -- years, in fact. (I think it was around 2001 or 2002.) Someone had ripped it from their video of the Samurai Shodown movie, so it was of very low audio quality, and abruptly faded out at the end.

    After some searching a couple of years ago, I found out that the artist is Gwinko. I also eventually found the song title: ”大好き”と言う嘘をついた

    Well, I spent yesterday on a renewed hunt for it, and actually found it at IMEEM: Daisuki to iu uso wo tsuita

    The lyrics (in romaji and translated) are also available at Anime Extravaganza.

    (This should be interesting to fans of Samurai Shodown/Samurai Spirits.)
  • PC DOS game music revival / how to encode to MP3 in Linux using GStreamer

    Jun 3 2008, 11h43

    Revisiting some DOS games of old...
    I've been playing old PC DOS shareware games lately in DOSBox, and one of the most nostalgic aspects of this is always the music (of course ;).

    So, I've been playing Alien Carnage, also known as Halloween Harry. I found that I still knew the melodies (if not the exact sound of the instruments) despite the fact that I had last heard them over 10 years ago. The music was composed by George Stamatiadis, with the title theme done by Steven Baker; the latter also composed the soundtrack for Mystic Towers, another Apogee game with great atmospheric music (and a killer main menu theme).

    I also happened upon Brudal Baddle, a not-so-great fighting game which I nonetheless remembered from the same mid-90s era, and first tracked down the soundtrack files, then, unexpectedly, the composer's website (with downloads and original track titles). The artist goes by The Finn for these tracks.

    Encoding to MP3 with GStreamer
    Eventually, I wanted to have this music extracted and playable in mpd (my music player of choice). But how? After some googling, I found a starter's guide to GStreamer, from which I picked up the following command:

    gst-launch filesrc location="INPUT_FILE_HERE" ! decodebin2 ! lame quality=3 vbr=4 ! filesink location="OUTPUT_FILE_HERE"

    I tried this on a .MOD file, and it worked! I'm not sure whether the ability to decode MODs is provided by a specific plugin, though; I have the Good, Bad and Ugly plugins installed for GStreamer.

    Anyway, typing/pasting a long command like that for every single file would be tedious, so I adapted it as a shell script, which I saved as tomp3.sh:

    gst-launch filesrc location="$1" ! decodebin2 ! lame quality=3 vbr=4 ! filesink location="$2"

    The $1 and $2 variables are replaced at execution time with the first and second arguments passed to the script, respectively. So, I can convert FOO.MOD to BAR.MP3 by running:

    ./tomp3.sh FOO.MOD BAR.MP3

    However, I still wasn't satisfied, as I was too lazy to even run this for every single file. So I made a simple loop to run the script for all the .MOD files in the current directory, like this:

    for a in `dir *.MOD`;do;sh tomp3.sh $a $a.mp3;done

    It's not perfect; for FOO.MOD, it'll produce a file named FOO.MOD.mp3. But that can be solved by using sed to substitute ".mp3" for ".MOD" in the second argument, like so:

    for a in `dir *.MOD`;do;sh tomp3.sh $a `echo $a|sed s/\\.MOD/.mp3/`;done

    This'll take FOO.MOD and produce an output file named FOO.mp3, which is about as far as I'd go in terms of nitpicking about these filenames.

    After that, I opt to grab a medium- or large-sized image of the game's box art, crop out a 300x300 region (or, at least, the largest square region possible from the source image), and use that as the "album cover" image. Just for completeness.

    Note: In these example commands, I wrote gst-launch. However, on my particular system, this app was installed as gst-launch-0.10. Your setup may vary.
  • Recognising a sample (again).

    Mai 25 2008, 15h53

    I was listening to Dimanche by Origa, and I immediately noticed that the sample used in the beginning (as well as sprinkled throughout the track) was also used in Infected Mushroom's Groove. Seeing as it sounds the same on both tracks, I'm assuming that this is the unaltered, original sample.
  • Same sample, vastly different artists...

    Mar 30 2006, 7h18

    I thought it pretty nifty when I listened to Semantic Spaces and recognised several samples borrowed from early Enigma. It made sense; Enigma and Delerium are both electronic/New Age/worldbeat artists, and they were supposed to have influenced each other.

    On the other hand, I was listening to Rosenrot last night, and it puzzled me to hear the female voice sample* from Innocente at the start of Zerstoeren. They were processed in different ways (Delerium used it more or less as-is, whereas Rammstein added an echo effect and used it as the backdrop for a thumping drum beat).

    I don't know, it's just funny to find the same material being used in quite different artists of quite different genres. Or maybe I just have too much time on my hands.

    ---
    * My search for information/credits for this vocal sample were fruitless. The specific sample isn't named anywhere - not even on Delerium.com, which offers the full liner credits for individual tracks. Odd.