New Release of the Endgame: Singularity Music Pack with Six New Tracks
286 views
Skip to first unread message
Phil Bordelon
Dec 15, 2012, 2:35:02 PM12/15/12
to endgame-sin...@googlegroups.com, endgame-s...@googlegroups.com
Folks,
I am pleased as punch to announce a new release of the Endgame: Singularity
music pack. Version 007 (heh) sports six new tracks by our resident musical
genius Maxstack. He used a new bit of software this time around (Renoise),
but both the rendered audio and the source files are under the same BY-SA
3.0 license as before. (We'll be adding the source files to the E:S repo
soon to go along with the Psycle sources for the other music that are
already there.)
I've also taken the time to rerender all of the older tracks with a modern
Ogg Vorbis encoder, which should give some minor quality improvements; in
addition, all of the tracks have had ReplayGain-based level adjustments
applied to them, so the sound volume should be roughly equal no matter
which track comes up.
The one change which may surprise you folks (or at least your ears) is that
one of the old tracks ("By-Product") is a 44.1kHz track, but Pygame played
it at 48kHz, pitching it up a bit. It's been resampled to 48kHz in the new
render, so it'll sound a bit different when played in-game.
Enjoy, and thanks as always to Maxstack for his fantastic music. Thanks too
to MestreLion for getting the ball rolling on this process.
P
I am pleased as punch to announce a new release of the Endgame: Singularity
music pack. Version 007 (heh) sports six new tracks by our resident musical
genius Maxstack. He used a new bit of software this time around (Renoise),
but both the rendered audio and the source files are under the same BY-SA
3.0 license as before. (We'll be adding the source files to the E:S repo
soon to go along with the Psycle sources for the other music that are
already there.)
I've also taken the time to rerender all of the older tracks with a modern
Ogg Vorbis encoder, which should give some minor quality improvements; in
addition, all of the tracks have had ReplayGain-based level adjustments
applied to them, so the sound volume should be roughly equal no matter
which track comes up.
The one change which may surprise you folks (or at least your ears) is that
one of the old tracks ("By-Product") is a 44.1kHz track, but Pygame played
it at 48kHz, pitching it up a bit. It's been resampled to 48kHz in the new
render, so it'll sound a bit different when played in-game.
Enjoy, and thanks as always to Maxstack for his fantastic music. Thanks too
to MestreLion for getting the ball rolling on this process.
P
MestreLion
Dec 15, 2012, 3:05:49 PM12/15/12
to endgame-s...@googlegroups.com, endgame-sin...@googlegroups.com
That is great news!
About the git repo with the new source songs and updated readme file, that was done 12 days ago:
https://github.com/MestreLion/singularity/commits/feature/music
Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
About the "album name", I loved your "Endgame: Singularity (Extended Knowledge)" suggestion. The also proposed "Further (or Advanced) Research" were also great.
As for ReplayGain... did you make sure to apply the gain to the sound *data* itself? Usually RG is only applied as tags, but since pygame is unaware of them, they should modify the data. The FLACs should only have tags tho.
Speaking of FLACs... can you also provide the original ones somewhere? We could offer 2 OST packs: the FLAC and the Vorbis. Or 6, if you consider "Original", "Extended" and "Full". I would gladly upload them to github.
Last but not least... you sure about "fixing" By-Product? I admit I was quite used to the faster version. We could feed the FLAC to any music editor to speed up the tempo by 48/44.1 ratio and then encode the faster version at 48khz.
Also, what quality are you using for the Vobis songs? Bumped to 160kpbs or still keeping it at 112? IMHO, 128 would be the best compromise between a still small package and a small bump in quality. Vorbis is VBR by default, so 128 is quite good for a game.
ML
>--
>
>
>
About the git repo with the new source songs and updated readme file, that was done 12 days ago:
https://github.com/MestreLion/singularity/commits/feature/music
Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
About the "album name", I loved your "Endgame: Singularity (Extended Knowledge)" suggestion. The also proposed "Further (or Advanced) Research" were also great.
As for ReplayGain... did you make sure to apply the gain to the sound *data* itself? Usually RG is only applied as tags, but since pygame is unaware of them, they should modify the data. The FLACs should only have tags tho.
Speaking of FLACs... can you also provide the original ones somewhere? We could offer 2 OST packs: the FLAC and the Vorbis. Or 6, if you consider "Original", "Extended" and "Full". I would gladly upload them to github.
Last but not least... you sure about "fixing" By-Product? I admit I was quite used to the faster version. We could feed the FLAC to any music editor to speed up the tempo by 48/44.1 ratio and then encode the faster version at 48khz.
Also, what quality are you using for the Vobis songs? Bumped to 160kpbs or still keeping it at 112? IMHO, 128 would be the best compromise between a still small package and a small bump in quality. Vorbis is VBR by default, so 128 is quite good for a game.
ML
>
>
>
Phil Bordelon
Dec 15, 2012, 3:23:57 PM12/15/12
to endgame-s...@googlegroups.com
On 12/15/2012 02:05 PM, MestreLion wrote:
> Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
As far as I'm concerned, you can go ahead and do it. The keys are yours.
> Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
> About the "album name", I loved your "Endgame: Singularity (Extended Knowledge)" suggestion. The also proposed "Further (or Advanced) Research" were also great.
> As for ReplayGain... did you make sure to apply the gain to the sound *data* itself? Usually RG is only applied as tags, but since pygame is unaware of them, they should modify the data. The FLACs should only have tags tho.
first digital audio rodeo. :) (This required some shenaniganery, but I
got it done.)
> Speaking of FLACs... can you also provide the original ones somewhere? We could offer 2 OST packs: the FLAC and the Vorbis. Or 6, if you consider "Original", "Extended" and "Full". I would gladly upload them to github.
at least a temporary basis. Not today, though.
> Last but not least... you sure about "fixing" By-Product? I admit I was quite used to the faster version. We could feed the FLAC to any music editor to speed up the tempo by 48/44.1 ratio and then encode the faster version at 48khz.
at 44.1kHz, then that's how it should sound.
> Also, what quality are you using for the Vobis songs? Bumped to 160kpbs or still keeping it at 112? IMHO, 128 would be the best compromise between a still small package and a small bump in quality. Vorbis is VBR by default, so 128 is quite good for a game.
for my music collection for over a decade. There's no need to go higher,
particularly with electronic tracks.
P
MestreLion
Dec 15, 2012, 5:21:45 PM12/15/12
to endgame-s...@googlegroups.com
At 18:23 15/12/2012, Phil Bordelon wrote:
>On 12/15/2012 02:05 PM, MestreLion wrote:
>>Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
>
>As far as I'm concerned, you can go ahead and do it. The keys are yours.
Done at https://github.com/singularity/singularity/tree/feature/music
>On 12/15/2012 02:05 PM, MestreLion wrote:
>>Just don't merge it right away, I still want to to add the info on the commercial instruments on original and extended tracks.
>
>As far as I'm concerned, you can go ahead and do it. The keys are yours.
I'll still way for a "Great, go for it" to push to google code. There are were many commits to master too, on the issues we discussed at IRC. So a bit of testing (or at least "proof-reading") os advised.
>>As for ReplayGain... did you make sure to apply the gain to the sound *data* itself? Usually RG is only applied as tags, but since pygame is unaware of them, they should modify the data. The FLACs should only have tags tho.
>
>Yes, I applied it directly to the data; it's not in tags. This is not my
>first digital audio rodeo. :) (This required some shenaniganery, but I
>got it done.)
>>Speaking of FLACs... can you also provide the original ones somewhere? We could offer 2 OST packs: the FLAC and the Vorbis. Or 6, if you consider "Original", "Extended" and "Full". I would gladly upload them to github.
>
>It's a 200MiB collection. I can definitely make it available somewhere, on
>at least a temporary basis. Not today, though.
When it happens, send me the URL so I can upload the FLAC pack to github as well.
>>Last but not least... you sure about "fixing" By-Product? I admit I was quite used to the faster version. We could feed the FLAC to any music editor to speed up the tempo by 48/44.1 ratio and then encode the faster version at 48khz.
>
>I believe that the artist knows best; if it's supposed to sound like it does
>at 44.1kHz, then that's how it should sound.
And, quoting Anakin Skywalker: "Faster By-Product... I will miss you... *so* much..."
(lame attempt to soften you heart and keep the faster tempo one :)
>>Also, what quality are you using for the Vobis songs? Bumped to 160kpbs or still keeping it at 112? IMHO, 128 would be the best compromise between a still small package and a small bump in quality. Vorbis is VBR by default, so 128 is quite good for a game.
>
>I use -q 3, which is the same quality I've used for my FLAC->OV downsamples
>for my music collection for over a decade. There's no need to go higher,
>particularly with electronic tracks.
ML
Phil Bordelon
Dec 16, 2012, 12:28:32 AM12/16/12
to endgame-s...@googlegroups.com
On 12/15/2012 04:21 PM, MestreLion wrote:
> I'll still way for a "Great, go for it" to push to google code. There are were many commits to master too, on the issues we discussed at IRC. So a bit of testing (or at least "proof-reading") os advised.
Yeah, the text on the music side needs some work. I'll poke at it
> I'll still way for a "Great, go for it" to push to google code. There are were many commits to master too, on the issues we discussed at IRC. So a bit of testing (or at least "proof-reading") os advised.
soon.
Possibly. I got it done, though :) Thanks for the pointer, though.
> I don't think he would mind, and he is already aware of the issue.
>
> And, quoting Anakin Skywalker: "Faster By-Product... I will miss you... *so* much..."
>
> (lame attempt to soften you heart and keep the faster tempo one :)
believe the author is completely dead. Authorial intent still means
something to me.
P
Evil Mr Henry
Dec 18, 2012, 8:36:43 PM12/18/12
to endgame-s...@googlegroups.com
Updated
On 12/18/2012 4:51 PM, Wingman4l7 wrote:
> Don't forget -- the music pack link at
> http://www.emhsoft.com/singularity/ needs to be updated. It's still
> pointing to version 6 =)
> --
>
>
>
On 12/18/2012 4:51 PM, Wingman4l7 wrote:
> Don't forget -- the music pack link at
> http://www.emhsoft.com/singularity/ needs to be updated. It's still
> pointing to version 6 =)
>
>
>
MestreLion
Dec 19, 2012, 4:40:16 AM12/19/12
to endgame-s...@googlegroups.com
Folks,
The source files for the new songs, in Renoise .xrns format, is now in the git
repository, along with a small directory re-estructure and updated README, at
./data/music-src
This latest push to master also includes almost 20 new commits and a lot of new
functionality (testers welcome!). But this is something for another post, we're
here now to talk about music :)
Phil, I had also prepared a music pack, and comparing to 007 it turns out they
are quite similar (as expected). The differences are:
- Applied ReplayGain value: calculating RG on 007 oggs should lead to ~0dB
gains, or close to, but they range from +0.95dB to -2.03dB, a 2.93dB diff,
averaging on -0.55dB. The ones I generated go from +0.15dB to +0.01dB, a
0.14dB diff, averaging on +0.07dB. Foobar2000 uses the quite recent (and
supposely improved) EBU R128 for its RG analysis, maybe this explain the
differences?
- Added the "Album Cover" image, converted from .gif to .png
- The URL for E:S website, http://emhsoft.com/singularity, moved from CONTACT
tag to Comment. Many players don't even bother to show custom tags, while all
show the Comment (including file browser right-click->properties).
- Dropped the "()" in new songs' Album name tag: "Endgame: Singularity Advanced Research"
for consistency with the original ones.
- Date tag uses just the release year, not a full date. And original songs were kept
as 2007, since it was when that "album" was "released". New ones are 2012.
- Added Genre tag, as "Game"
- Kept Tracknumber tag (but not in filename)
- Re-calculated RG info on generated oggs and applied its tags. Useless for pygame,
but shows us (and users) that files were properly normalized
Other than that, files have the same major properties: size (52.3MB), quality (-q 3),
bitrate (112kbps nominal, although actual VBR average for each song differ 1~3kbps
from 007), sample rate (48000Hz).
And no, I didn't cheat with By-Product. I used the slower-but-correct version :)
But I did make a "faster-and-familiar" version, not included in the zip, just in
case you or anyone is interested.
So... I was going to upload this pack, along with both FLAC packs (renamed and fully
tagged and RG'ed) and the alternate By-Product.ogg, to github, when I found out this:
https://github.com/blog/1302-goodbye-uploads
Yes, I missed it by 1 week :( So, is there any interest in any of these packs?
ML
I used foobar2000 (yes, shame on me!) for all tagging, conversion and RG processing
>--
>
>
>
The source files for the new songs, in Renoise .xrns format, is now in the git
repository, along with a small directory re-estructure and updated README, at
./data/music-src
This latest push to master also includes almost 20 new commits and a lot of new
functionality (testers welcome!). But this is something for another post, we're
here now to talk about music :)
Phil, I had also prepared a music pack, and comparing to 007 it turns out they
are quite similar (as expected). The differences are:
- Applied ReplayGain value: calculating RG on 007 oggs should lead to ~0dB
gains, or close to, but they range from +0.95dB to -2.03dB, a 2.93dB diff,
averaging on -0.55dB. The ones I generated go from +0.15dB to +0.01dB, a
0.14dB diff, averaging on +0.07dB. Foobar2000 uses the quite recent (and
supposely improved) EBU R128 for its RG analysis, maybe this explain the
differences?
- Added the "Album Cover" image, converted from .gif to .png
- The URL for E:S website, http://emhsoft.com/singularity, moved from CONTACT
tag to Comment. Many players don't even bother to show custom tags, while all
show the Comment (including file browser right-click->properties).
- Dropped the "()" in new songs' Album name tag: "Endgame: Singularity Advanced Research"
for consistency with the original ones.
- Date tag uses just the release year, not a full date. And original songs were kept
as 2007, since it was when that "album" was "released". New ones are 2012.
- Added Genre tag, as "Game"
- Kept Tracknumber tag (but not in filename)
- Re-calculated RG info on generated oggs and applied its tags. Useless for pygame,
but shows us (and users) that files were properly normalized
Other than that, files have the same major properties: size (52.3MB), quality (-q 3),
bitrate (112kbps nominal, although actual VBR average for each song differ 1~3kbps
from 007), sample rate (48000Hz).
And no, I didn't cheat with By-Product. I used the slower-but-correct version :)
But I did make a "faster-and-familiar" version, not included in the zip, just in
case you or anyone is interested.
So... I was going to upload this pack, along with both FLAC packs (renamed and fully
tagged and RG'ed) and the alternate By-Product.ogg, to github, when I found out this:
https://github.com/blog/1302-goodbye-uploads
Yes, I missed it by 1 week :( So, is there any interest in any of these packs?
ML
I used foobar2000 (yes, shame on me!) for all tagging, conversion and RG processing
>
>
>
Reply all
Reply to author
Forward
0 new messages