Renoise 3 XRNS what and how are things saved?

I probably overlooked it, but I couldn´t find any clear info on this in the manual. All my samples are saved (lossless) within the xrns file right?

yep,

in the preferences you can also set your sampled instruments to be saved as .flac instead of .wav.

All my samples are saved (lossless) within the xrns file right?

Correct.

The XRNS file itself is essentially just a standard ZIP archive which contains your song data in XML format, and samples in either FLAC or WAV format.

We use FLAC compression by default, but you may also choose WAV if you prefer, via Preferences > Files > Song/Instrument Export.

Using uncompressed WAV generally allows the song to be saved faster, which can be handy if you have automatic backups enabled, but it has the obvious downside of creating much larger XRNS files.

Okay cool, thanks! Very usefull info :walkman:

Correct.

The XRNS file itself is essentially just a standard ZIP archive which contains your song data in XML format, and samples in either FLAC or WAV format.

We use FLAC compression by default, but you may also choose WAV if you prefer, via Preferences > Files > Song/Instrument Export.

Using uncompressed WAV generally allows the song to be saved faster, which can be handy if you have automatic backups enabled, but it has the obvious downside of creating much larger XRNS files.

I did not know this… Oh god… what’s the easiest way to convert giant .xrni’s to be .wav’d .xrni’s? Drumdrops drumkits sound really awesome but every time I hit ctrl-s it takes forever for them to resave (hoping for external .xrni’s one day). Do you think it would make a significant enough difference to be worth it? The kits are around 250MB - 300MB.

what’s the easiest way to convert giant .xrni’s to be .wav’d .xrni’s?

If you change the setting to uncompressed WAV in Renoise, then it will apply to all samples saved within your song, regardless of the mode used when the XRNI instrument itself was created.

So there’s no need to convert your entire XRNI library. In fact, it’s probably better to leave them saved as the default FLAC-based instruments, just to save on disk space.

Do you think it would make a significant enough difference to be worth it?

This really depends on your system, and the balance (and potential bottlenecks) between your CPU power and disk speed.

On one hand, you may find that it’s faster to simply write a larger WAV to disk, than it is for your CPU to compress the WAV to FLAC.

On the other hand, your CPU may be totally badass and able to compress the WAV to FLAC super fast, and will then benefit even further from writing the smaller file to disk — doubly whammy!

If you have a decent hard disk that’s at least 7200 RPM and not fragmented to hell, or you have an SSD, then you’ll quite likely see faster save times when using uncompressed WAV.

Just create a test song with shit-loads of audio data and perform some timing tests of your own :slight_smile:

I do have an SSD but it’s tiny and right now pretty much only being used to hold Windows, and all my data is on other slow drives (that I can’t remember the rpm’s of). CPU is an i5.

I actually did recently buy another (small) SSD just to screw around with linux on, but I feel like that experiment may be coming to an end soon. Been considering formatting it and using it as my new Renoise drive. It’s not quite big enough for everything though, so maybe I’ll just wait until I can afford a larger drive (too many samples). I will go ahead and just do a test on SSD though, I seem to remember trying it before and it saving so fast (while flac) that it blew my mind.

Test results:

Loading XRNI’s:

Compressed on new SSD: loads in 5 seconds, and for an additional 14 seconds Renoise just hangs up with a blue circle. 19 seconds 291 MB XRNI

Uncompressed on new SSD: as far as my stopwatch is concerned, identical. 583 MB XRNI

Uncompressed on 7200: Actually pretty similar, but hang time was longer, bringing total to 23 seconds.

Conclusion: I didn’t spend enough money on my SSD?

Saving uncompressed XRNI test:

New SSD for storage: 5 seconds then 7 seconds of not responding, 12 seconds total

Old SSD where windows is: identical

7200 RPM HD: 29 seconds, appears to “hang” entire time, whereas on SSD it shows activity then hangs.

Conclusions:

Uncompressed .wav doesn’t seem to be any faster to work with than the default FLAC.

Load time does not seem to be significantly different comparing SSD to normal HD.

Save time for SSD is pretty bonkers faster. Almost 3 times faster.

Disclaimer: I am not a scientist

Scientific enough for me. :smiley:

I have question about ogg sample compression in song, just like in Hunz xrns-s.

https://forum.renoise.com/t/hunz-album-xrns-files/29733

How the song files where prepared in terms of sample file format? I am asking because when i open them and save without any changes the file size gets much larger.

How the song files where prepared in terms of sample file format?

The song was most likely processed by xrns_ogg

http://xrns-php.sourceforge.net/

The script is rather simple: unzip the song, ogg compress samples, then re-zip it.

So it works fine with 3.x songs as well (I used it recently for the Mutant Breaks compo)

But yeah, it’s not recommended to continue working on a song which have the samples OGG compressed. Upon loading, the samples will be uncompressed and then stored like that. I would use xrns_ogg exclusively as an ‘export’ function.

just convert samples?! I don’t need to change any xml song data and renoise will ‘catch’ that samples have different format/extension?

in that case should be fairly easy to write my own script to do that, witch could be a good exercise

THX

just convert samples?! I don’t need to change any xml song data and renoise will ‘catch’ that samples have different format/extension?

No, I believe the sample names are being swapped too.

But why re-invent the wheel? Especially with the windows front-end, xrns_php is a breeze to use.