Offline
Devon UK

I'm thinking of putting together an EP of some of my chiptunes, and I'd like a bit of technical advice. I'm using WavePad sound editor and recoding audio from a DS into line in on my PC.

1) Do I need to "normalize" all tracks to the same level, and if so what level?
2) How much space should I leave at the beginning and end of each track?
3) This probably sounds dumb, but when saving my tracks as mp3s, what bitrate is most suitable for music made on a DS? (I'm using 128, but I'm thinking maybe I don't need such a high rate for such lo fi music?)

Any help much appreciated.

Offline
Indiana
Robin wrote:

This probably sounds dumb, but when saving my tracks as mp3s, what bitrate is most suitable for music made on a DS? (I'm using 128, but I'm thinking maybe I don't need such a high rate for such lo fi music?)

How are you planning on releasing these recordings?

Unless there is a particular codec you're set on, it shouldn't be necessary to convert your lossless to MP3. Most release platforms will handle the conversion themselves (bandcamp etc.)

Also, 128 definitely is not a "high rate" it's pretty antiquated and there will be noticeable distortion introduced to your music, easiest to hear in quiet sections that will sound unnaturally "warbly"

Normalize your mixes is a good idea as well. If you're looking for a mix straight from your DS without any buss processing then just normalize to somewhere around -.3 dBFS

Spaces left before or after a recording is to taste. Small bits of added silence can help differentiate your music from music played before or after it on an iPod or some playback device. CD's are typically set up to automatically insert 2 secs of silence between tracks, so you if you're planning making CD's it may be less necessary.

Last edited by Fudgers (Jan 3, 2016 12:09 am)

Offline
Chicago

If you're really serious about distributing an album, I would look into having it professionally mastered and let the audio engineer take care of those things. It makes a big difference, and I'm sure you can find some recommendations here for people who would do it for not too much money.

Offline
Devon UK

Thanks,

I'd probably use bandcamp.

So, even if I'm recording 8-bit chiptune (eg LSDJ), I still need a high bit rate mp3 for it to sound good?

No sure what "lossless" means. I've been saving my work as mp3 from WavePad. Should I save it as .wav?

Offline
IL, US

yes, wav would be lossless.. bandcamp allows upload of wavs, so its best to go with wav and let them handle any conversion

Offline
Nottingham, UK

MP3 Conversion affects the frequencies present in your music.  128 mp3's have a noticeable roll off on high frequencies, as well as a weird, quite noticeable distortion. Chip music has a lot going on in the high frequencies, so exporting as wav and then letting bandcamp convert to 320 or V0 is the way to go.

Offline
Unsubscribe
e.s.c. wrote:

yes, wav would be lossless.. bandcamp allows upload of wavs, so its best to go with wav and let them handle any conversion

bandcamp REQUIRES high quality wav.

Offline
Toronto, Ontario, Canada
Robin wrote:

So, even if I'm recording 8-bit chiptune (eg LSDJ), I still need a high bit rate mp3 for it to sound good?

Robin wrote:

No sure what "lossless" means. I've been saving my work as mp3 from WavePad. Should I save it as .wav?



The 8-bit term is a little misleading. It generally refers to the CPU architecture of the platform rather than a meaningful summary of the programmable sound generator's capabilities. Even if the PSG has a totally 8-bit architecture - the Game Boy does not, as much of it is 4-bit - the sound is eventually converted from purely digital information into an analog output.

When you're using your computer to record audio from your Game Boy you are taking the analog waves that the Game Boy outputs and converting it back into a digital format; with WAV files, it's stored in the form of PCM samples. This WAV file stores the amplitude of the incoming analog signal as measured at a regular interval, such as 48kHz. In this form, the data is uncompressed and "raw". This does not mean that it is a perfect reproduction of the input signal - the bit-depth and sample-rate, as well as the components of the physical hardware you're using to record the audio, will determine how close to the input signal you can get - but a 24-bit WAV file at 44.1kHz or 48kHz will sound very close to the original.

Unfortunately, stored in this format, files are quite large. Compressed formats exist to ease this issue. The mp3 format is a compressed audio format that allows for much, much smaller file sizes. The downside is that the mp3 format is "lossy", meaning that some data is thrown away from the original to make the file smaller. This will generally make an audible difference between the raw and mp3 version, but the mp3 format aims to minimize this issue by being a little clever about what it throws away. Higher bitrates here will sound closer to the original than lower ones, with 320 kbit/s being rather good and 128 kbit/s being rather bad.

A middle ground exists in formats like FLAC. Unlike raw formats, FLAC is compressed and is therefor a more manageable size, but the compression is done in such a way that the original data can be perfect reconstructed from the compressed data. The downside here being that the format can't achieve sizes as small as lossy formats and it's less widely known and supported than the ubiquitous mp3.

TL: DR;

The mp3 bitrate affects how closely the resulting audio will sound to the original. When internet speeds were slow and hard drive space was at a premium, low bitrates had their place, but there's no excuse nowadays to use anything but 320 kbit/s (the maximum for a standard mp3).

Offline
Melbourne

^ great explanation


bandcamp supports FLAC uploads, btw; much faster to upload than WAVs.

Offline
(ノ◕ヮ◕)ノ*:・゚✧ el ass dee j
pselodux wrote:

bandcamp supports FLAC uploads, btw; much faster to upload than WAVs.

Holy crap I didn't know that! I'm gonna do that from now on!

Offline
United States

make it as loud as possible