^^^
Also, is there a way to apply ADSR etc to slices?
Not that I know of, unfortunately !
the envelope applies to the slices so yes, you can p-lock AHR settings to each step / slice
2 Likes
This is a bit hacky and awkward, but there is a trick for this. I thought this might be mentioned by now but it doesnāt look like it.
This is easiest to work with if your slices are evenly spaced, and the number of slices you have is a power of 2, especially 64, 32, or 16. If you have a different number of slices, it would work best to pad the end so that you have 64 evenly spaced slices but the last few are silence.
After that:
- Turn off slice playback.
- Create a custom āstraight lineā LFO with value 1 for all 16 steps.
- Set the LFO target to sample start.
Now, sample start controls slice number, as usual, and LFO depth controls position within the slice. If you used 32 slices instead of 64, slices begin every other step instead of every step.
1 Like
Iām wondering if anyone knows a work around for this.
Iām using chains for drum parts but every time a new trigger hits, the previous one cuts off. Is there any way to have the full slice play for each slice? ie. playing kick snare ride cymbol?
Sample Tracks are mono. No overlap.
You need to use 2 tracks or more if you want overlap.
2 Likes
You can also send the sound to the Reverb which will somewhat diminish the sound of the sample cutting off. Though ultimately āwhatā sezare56 said aboveā¦
Also, donāt forget you can resample. You could use 2 tracks to get the drum loop you want going and then sample it and just play it back on 1 track instead of 2.
2 Likes
Sorry for this massive necrobump, but one question remains for me regarding sample chains vs. sample change via p.lock (and please be gentle, I am a new OT user!).
Given that I do not understand exactly how the OT allocates memory, wouldnāt having a sample chain with, letās say, a 12 part drum kit, take more memory than switching between 12 sample slots in real time? Because sample chains are fun and all but I cannot stop thinking about the memory restrictions (sorry, I am an oldschool sampler guy, memory optimization is always top of my mind!)
That is, running under the assumption that the OT only takes as much sample memory in a track as the sample currently being played/selected, if it allocates RAM for all the samples in the pool you have allocated in a pattern, then itās the same thing.
1 Like
This sounds closer to being correct, in my understanding.
The memory use is about the length (in bytes / or seconds) of the samples that are available to be played instantly. So, a sample chain with 8 kicks concatenated takes the same RAM as those 8 kicks loaded separately. Of course, as mentioned above, the chain only takes one slot, while the 8 individual samples would take 8 slots.
There are tradeoffs, but it sounds like youāre on your way to understanding them.
Hope that helps.
Yeah but when the Octatrack changes samples mid-pattern in a flex track, did it preload all those samples in RAM previous or does it allocate them as you call them? Because you can change them as you are playing patterns so pre-allocating sounds insane, unless the OT reads your mind 
Just to be clear, I am trying to compare memory usage using one flex track changing samples with p.locks vs. loading a sample chain sample. It seems to me, if I understand correctly, that sample chains will use a lot more RAM.
Maybe I should do some testsā¦
1 Like
In order to assign a sample to play on a flex machine, it needs to be in the flex sample slot list, which means that it has been loaded into RAM.
Thatās not necessarily correct. All the sample data, whether itās in separate samples or one sample (that contains a chain) has to be in RAM if it is to be played by a flex machine. See āMachinesā in the Overview of the Octatrack structure section of the manual, or the Loading Samples to the Sample slots section.
There might be slightly more RAM used by a chain if there are gaps between events in order to enable easy slicing.
3 Likes
As I said, I am a new user, so I am trying to figure this out. This is something important
I didnāt understand it that way. So if I fill a slot on the flex sample pool, it automatically takes up RAM, whether I use it or not? The manual really does not explain this at all.
Also that kind of makes p.lock sample change a bit less useful. When I was explained this functionality by a friend, it was taught to me as a way to save memory.
It explains it many times. See Projects and RAM memory, for example:
The total amount of RAM memory available to a project is 85,5 MB. This memory is shared between the track recorders and the samples loaded to the Flex sample slots. By default the Flex sample slots can be filled with 64 MB of samples
As I mentioned, it might save a little bit of memory if you have unused material in a long sample (chain). It certainly enables you to have more than one sample used on a given track for a given pattern.
2 Likes
"This memory is shared between the track recorders and the samples loaded to the Flex sample slots. "
Thatās all it says, almost in passing. I didnāt understand it that way, thanks for clarifying it.
So yeah, a sample chain and the samples separately then would definitely take up the same space. Sample chains allow you to free up slots, and gives you a few other performance options. Got it.
Memory-wise, thereās no advantage, unless your individual samples are in differing sample rates, which I am not even sure the OT supports.
There are two slot lists.
One Flex (RAM)
One Static (stream from disk)
So if you load your samples into the static list. You have used ZERO RAM.
1 Like
The OT doesnāt support different sample rates at all (it only supports 44.1kHz).
I guess youāve meant different bit depths (16 or 24 bit), but these cannot be mixed at the same time. All samples get automatically converted to the globally set bit depth when they get loaded into RAM.
3 Likes