About Extremely Long Samples

Hi,

I use certain long samples quite alot (vocal clips from movies, random FX, etc). However, when I play the next pattern without listening the pattern where I used the sample first they won’t be audioble because they are not triggerd. This seems logical but it is rather anoying infact. It increases the time to compose by being forced of using one or more pattern(s).

What if there would be a system that checks the length of a sample and check if it is triggered thus far behind that it can still playback. Perhaps an automation of sample offsets in each pattern?

Well, I wonder if it is possible and worth the effort developing it :)

I am hoping for this one day aswell. In the meantime you can use energyXT to play back your samples in an XT sequencer audio track. The sample will then be triggered from the correct position wherever you start playback in renoise.

The method is outlined for freezing tracks in this thread but the steps are similar for long sample playback also:

http://www.renoise.com/board/index.php?act…7e5eb88dcd248e1

I have a similar problem. I often record long sections in renoise, with the built in recorder for external stuff and with tapeit for live recording of vst’s. But after recording i have to cut it up into slices (of pattern length or 4 bars length), assign these to keys, and write the notes into a track. Altough the slicing is easier with 1.9 and the snap to beat option (before i had to use soundforge), its still bothersome. I think the “chase samples backwards” suggested here would be the best solution to this problem. It could be implemented as a checkbox in the sample properties.

If this is hard to implement another way would be to have an “auto split” function in the sample editor. When you click this you get a dialog box where you can choose the resolution of the split in bars. Then renoise automatically creates the slices, maps these to keys, and insert notes, from the start of the currently selected pattern. A similar idea was suggested here but i think it would be better to implement it outside of the recorder, because this way you can also use it with samples recorded outside of renoise.

I think only hard drive streaming can solve this. as far as I understand, this will take quite a while, but probably possible when an arrange window is possible.

You can use the rendering function as workaround meanwhile. Just solo the track with the long sample, deactivate the track and master fx and while rendering tick the box “Save each pattern into a separate file”. You can then drag the resulting samples into an instrument and use the generate drumkit button to quickly map them.

Yep, and with 1.9 you can ‘snap to beat’, making it a lot easier to cut up your samples to fit the current beat of the track.

I also prefer to work with smaller samples anyway, so with snapping I’m all good until some serious hard drive streaming can be done.

I also prefer using shorter samples. When i record a vocalist for example i always cut the sample into smaller parts assembling it back together. Although, often my vocal samples start very late on the previous pattern, maybe just a few words there then playing on to the pattern i’m working on. So when this happens i often use 09XX code on the vocals, somewhere at the start of each pattern making sure the vocal track plays when i hit play on that pattern.

It’s usually not that painful getting it to sync and merge well with the previous pattern. But that’s tracking (Using Codes) :P The Effect commands are there for a purpose.

That’s brilliant… I wonder why I never thought of doing that :P

I also sometimes use long vocal samples, and I think this is a problem skimmed over in previous posts about visual sample lengths. I would like to see a fix for this, but it can be worked around by the afore mentioned methods…

but I ask, would it be too hard/ worth it to make it so your sample is heard where it is is in relation to where it was triggred?

[

Hmmmm reread your post and got an idea. Maybe it’s not doable but there should be some kind of effect column code for playing a sample by tick. Or maybe you could have a checkbox in the Sample editor telling this sample to perform as you want to. Wich means play when you’re pressing play at the right position even if it’s not triggered by a note. eeehhhh Hmmm don’t really know how to explain but I’ll try.

So for instance… Pressing play as it is now samples are just triggered when you’re passing them in the pattern editor. Maybe there is a way to add code te renoise software (I’m not a programmer) that automaticly keep track of these samples with that checkbox marked.

What it does??. Well… that function would have to know the where in the song this sample is played from start to finish. Pattern 0 to pattern End. and then kind of automaticly slizing the sample in Line or Pattern sized chunks triggering every line/Pattern they are played. So. If you put a 3 pattern long sample at pattern 0, and pressing play at pattern it would retrigger these “chunks” based on line or pattern or whatever. So if you move to pattern 2 this function would know to play your sample at “chunk part 2” (if pattern oriented) or at Line # XYZ position (Line oriented) when you press play on that pattern.

So if you nudge your note in pattern 0 this function would atomaticly recalculate the start and stop values of each “chunk” so that when you press play on any pattern where this sample should be played It will certanly do. maybe it will be heavy performance wise I don’t know but maybe this is a possibility. And that’s why this option should be enabled by a Checkbox so that you only use it on stuff you really need to play when it should.
Another scenario is when you change the pattern lenght of pattern 1 for example. This function would know that it has to redistribute time or sample data to chunk 1 and taking away some from chunk 2. I don’t know how the memory stores samples and how it works but this is just my idea of how long samples could be handled as “chunks” or pices by the memory when enabled. Any devs or coders able to ched some light on this? is it doable or not?

But my common sense tells me that the playback speed or note has to be fixed when using this function. No pitch effects and basically all effect column effects could have no effect on samples with this option clicked. I don’t know if it would make this feature less usable, but for REALLY long samples that you don’t plan on doing anithng with it could be quite useful.