Observation: If you tweaked that slightly to C, D#, F#, A you would have regular minor 3rds all the way.
(EDIT: Which has benefits to my OCD brain, probably very few in reality, unless you wanted to automate things)
3 Likes
Yep, I would have preferred to do it that way, but chose to move the D# up to E so that I could type one less character and because E was also commonly used as a default root note by some samplers.
1 Like
makes sense, guess i was just curious how that sharp came to be. good over-thinking!
cheers
1 Like
(I also sing best in F#… 
)
1 Like
F#m is my fave scale for reasons beyond me!
sry didn’t mean to derail the thread. I enjoyed the tangent though
1 Like
F#m is my most used key… my guitarist teases me about it.
2 Likes
Ive got tonnes of samples of synths ive owned over the years. Today i found the OT only has a 2 octave pitch from one sample. But it does mean i only have to load a few samples.
1 Like
Found some old tips I wrote for customers:
Definitions & Tips for Creating Multi-Sampled Instruments:
- Key-Mapping: ‘Root-Key’ (aka key center or key index) is the sample’s root pitch.
- Key-Mapping: ‘Key-Range’ is the Lowest-Key and the Highest-Key that the sample should play.
- An economical & popular method for sampling an instrument is to sample 2, 3, or 4 notes in each octave: (2 notes: C, F#), (3 notes: C, E, G#) or (4 notes: C, D#4, F#4, A). Notice that there are identically numbered ‘note gaps’ between each sampled note for each of the 3 examples. When mapping the samples, set the key-range of each sample to fill in the gaps. A good practice is to set the key-range to go more to left than to the right of the root-key (i.e.-when necessary, go lower in pitch versus vs going higher since it usually sounds better when going lower. If the Highest-Key is too high, samples will play faster and may finish too quickly. The ‘Chipmunk effect’.)
- It’s a balancing act: recording each note/key/fret would be ideal, but the work involved in mapping all of those samples is very time-consuming, and more samples takes up more system resources, like RAM, disk space, & CPU.
- Sometimes a samplepack will have a key-definition file (i.e. - *.sfz file) which you may be able to open in a text editor. This can be helpful for finding the appropriate key-ranges, and setting the sampler and synthesis parameters. (Also *.nki files can be opened in Kontakt Player for checking in the Mapping Editor, plus *.exs files can be opened in Logic’s EXS24 sampler for checking parameters.)
- Sometimes samples with be ‘Key-Labeled’, meaning that they will have the root-key in the file name (i.e. - “Piano C3.wav”). Or possibly it may be in the MIDI Note Number format (i.e. - “Piano 060.wav”). Certain samplers can read this info and will auto-map the sample to the correct key. Middle C depends on the manufacturer (Usually C3, C4, or C5).
Below is a convertor for MIDI Note Number to Key, Hz to Key, and samples to kHz.
- Sustaining Sample Loops: This is probably the most time-consuming and frustrating process. This is used so that the sample sustains while using an ADSR envelope. If you hold a key down for a long time, the samples will sustain and continue to play inside the loop. MONO samples are significantly easier to loop than STEREO samples! That’s because with STEREO, there are two waveforms - a left & a right, and those waveforms are almost never identical! Clicks & pops where the loop starts and ends (aka ‘Loop-points’) are common. Use ‘Zero Snap’ to find where the waveform crosses 0 (basically silence) to help set loop-start & loop-end points.
- Crossfade looping: This helps immensely if you still get clicks & pops at the loop-points, but enabling this will use more system resources. Crossfade looping is sometimes the only option for stereo samples.
- Multiple Layers: This adds to the realism and complexity of the instrument. It’s also useful for sorting samples that are too loud or quiet, or have a different timbre. Layer string samples with piano samples for the classic Piano+Legato Strings patches. Or layer different drum samples for maximum impact. Or for sound effects, soundscapes, pads, etc.
- Modulations: These can help liven-up a sound, making it more complex and interesting. Put a slow tempo-synced LFO on the filter cutoff, for example. Try modulating snare and hats amp decay time. Use small amounts of Random or S&H to modulate the pitch, amp decay, filter, panning, etc. LFOs that can reset at note-start or are free-running are very helpful, as well as modulation sources that can be synced to the tempo of your session.
7 Likes
Excellent info, thanks!
Nice timing on this thread, only yesterday I decided to sit down and have a deeper dive into sampling in my EPS classic. Except instead of sampling synths I literally grabbed the most recent audio on my phone which was a mates voice message explaining the process he went through to clean his son’s projectile vomit off his front door.
2 Likes
MPC Beats as mentioned up thread is quite good for multisampling, and is free, the one thing that I would change about it is to have sampling determined by note length rather than time, other than that no complaints.
Sampling phrases is also something worth doing with analog synths, you can sample a few takes maybe with slightly different settings to add variation.
The other thing quite nice to do is sample a single sound and use the samplers synthesis functions to make a new sound, OT excels at this but plenty of other samplers like PT, Deluge, MPC etc can do interesting things with single note samples.
Sampling is probably my favourite technique outside of analog synthesis and I think they combine both ways very nicely - sample analog synths, and running samples through analog filters etc.
One of my fave techniques is an old staple, sample a fairly long analog synth sample where the filter is being swept by the envelope then play the sample back at different start points for some nice animation.
Drum synthesis is also something I am very fond of, every modular and non modular analog synth can often be used to make interesting percussion.
1 Like
I like to sample my tones while working with the notion that this will not replace my synthesizer. That’s why I don’t go for different filter positions or multisampling or velocity or any of that. Instead, the synth sound I sample is fodder for samplers. The complex sound from the synth is now the building block, the oscillator, for a sampler. The sampler is now the synthesizer. Bit rate, sample depth, filters, keytracking, layering, resampling, and all of the things that samplers do will now be the thing that will ensure I get new tones.
For me, the synth acts in service of the sampler rather than vice versa.
2 Likes
That’s a great way to look at it. One of the main things that has intimidated me about building sample banks with synth sounds is this fear of not having synthesis-level control of the sounds once they’re sampled, but your framing of it is the perfect way to treat it as its own workflow and “instrument”.
1 Like
This, super simple workflow.
I take it Sample Robot, Mainstage, and MPC Beats are still the top choices for software auto sampler?
Trying to resist GAS for one of the hardware solutions (MPC One/Live)