Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Lowering The Sample Rate
![[Deleted User]](https://secure.gravatar.com/avatar/b784dd32cce208260321a2c2b1cbd583/?default=https%3A%2F%2Fvanillicon.com%2F1e0e61fd9eb6ef13a1b0720066db0070_200.png&rating=g&size=200)
If I try to match the sound of a vintage sampler using a “bitcrushing” plug-in like TB Bitjuggler, reducing the sample rate is sometimes not enough: I often have to reduce the bitrate to lower than the bitrate the original vintage sampler claims to have. This is especially evident when trying to match vintage machines that had a variable sample rate, like the boss SP202. Why is this? Does the bitrate reduce on a vintage sampler when you reduce the sample rate, or does sending a “hot” signal into a 16 bit sampler create a bit-reduction-type sound effect?
Comments
https://apps.apple.com/nl/app/rx950-classic-ad-da-converter/id1406136418
It's not always as simple as just slapping a decimator/bit-reducer at the end of the signal chain and expecting it to sound 'vintage'. The old samplers used various methods to transpose the sounds up and down which played a huge part on the overall sound.
Modern samplers and interpolation methods are almost 'too clean' and new apps do as much as they can to reduce arising noises which again was part of the character of old samplers.
It took a long time for me to realize what gave my old Amiga samples their 'classic sound', and surprise surprise it was not the '8-bits' but rather a combination of sampling-rate and overdriving/clipping the AD converters when sampling.
Not may sampler-apps allow keyboard tracking of the decimation frequency which can partially be used to re-create the aliasing noises when transposing samples up/down so...
...one tip to create 'vintage' samples is to sample thru an effect chain with a some filters and decimators and shape the sound before it's actually sampled, not the other way around starting with 'clean' samples.
Cheers!
If the „hot“ signal exceeds the analog input range of the converter, it will distort - but still output the same (maximum) sample value.
You‘d have to lower the signal fed to the converter to achieve a bit reduction effect, but then noise will increase dramatically.
A simple bit reduction has few in common with a converter for a lower bit rate (according to standards at the time of design).
Many of these chips were considered state of the art back then and had a specific analog performance.
The RX950 plugin/app is an example that emulates such details to a fairly high degree.
Most digital units from the early days had tremendous effort of analog signal processing before and after conversion stage.
See also NE570 compander
https://httprutgerverberkmoes.com/?page_id=2869
Just to be clear, I’m not looking for bitcrusher recommendations- I’m asking if samplers like the Sp202 also lower their bit rate when you lower the sample rate.
I really don't get the 'Bit Rate'. Do you mean 'Bit Depth'?
Most samplers even the old ones worked with a fixed 'Bit Depth' (8,12,16 etc.) and variable 'Sample Rate' was used mainly because memory was expensive and CPU's were slow.
In order to minimize 'aliasing' when sampling with lower sampling rates some kind of filtering was done to cut out frequencies that would cause aliasing with the lower sample-rates. (ie. if one samples at 8k nothing above 4k would be reliably captured).
As CPU's were too slow to do proper re-sampling/interpolation the playback was done by adjusting the clock of the of the converter (or do a very rough repeat or drop a sample) to be able to play the sample back at different speeds and then passed thru a low-pass filter to minimize aliasing.
Nowadays all hardware works with 'fixed' sample-rate (44.1, 48, 96 etc.) so in order to have something at 'lower sample-rate' it has to be 're-sampled' using various interpolation methods.
So to answer your question of samplers lowering the sample rate, it's a NO as the current hardware is sample-rate and most often bit-depth locked.
What can be done is to emulate the effect of sampling and playing back at lower frequencies using different methods like sampling thru a decimator with filters.
Currently the best SP like sampler we have for the iPhone & iPad is Koala Sampler.
If there is the need to process the audio that is to be sampled Koala can be hosted in AUM's effect slot after the effect processing.
The SP202 always uses 16bit (depth) for AD/DA, regardless of the sample rate.
And the sample-rates are quite low as well I noticed...
Hi-FI - 31.25Khz
Standard - 15.63Khz
LO-FI 1 - 7.81Khz
LO-FI 2 - 3.91Khz
So it's very, very likely that they apply quite heavy high-pass and low-pass filtering to the input signal before it's 'sampled' to avoid extreme aliasing. So for the 31.25Khz sample rate It's quite likely that there's a pretty steep low-pass filter starting around 12-13k and likely a high-pass filter at around 100-200hz range.
For Standard they'd need to cut out everything above 7.5k
LO-FI 1 everything above 3.5k
LO-FI 2 everything above 2k
So with proper settings the mentioned TB Bittjugler could most likely be used to process the sound before it's being sampled.
(Heck I even got them to implement the pre decimation low-pass filter for the sole purpose of emulating old samplers!).
On the other hand if the 'mp3 like' bit-rate reduction is needed LO-FI-AF might do the trick.
Old samplers are fun for sure. Still got my old Korg ES-1 which has a very 'rough' 32Khz 16-bit AD/DA converter and the fun stuff with that one begins when you start to over-drive/clip the AD converter
(Need to find some Smart Media cards so I can pull out the samples though).
Good to know it’s fixed. (The manual says 20 bit, rather than 16) I just don’t understand why there is quantisation noise on the sp202 when you lower the sample rate, cause that usually comes from reducing the bit depth. The quantisation noise isn’t there at higher Sample rates on the sp202.
I’m not sure just HOW heavy the filtering is on the sp202, because you can hear quite a bit of aliasing and ringing at the low bit rates, and leaving Bitjugglers filters open gets the right sound. But you can also hear a lot of quantisation noise using the SP202 at low sampling rates, which usually comes from bit depth reduction. Why would that be? The only way to emulate it in Bitjuggler is to lower the bit depth, but apparently bit rate is fixed in the sp202.
There’s quite a lot of options in BitJuggler when it comes to the quantization and method of rate reduction and how the gaps between the samples are filled. The the ‘ringing’ is aliasing noise when the highest frequency of the source is too high for the selected sample-rate, it something that can be adjusted by ear and not strictly looking at numbers using both a low pass filter and sample-rate.
In order to do faithfully replication I’d need to find more technical details of the SP202 than what the user manual provides. And preferably sample the same source and then dial in the sound.
But that will have to wait for now…
One thing I like to do is send whatever source from iPad/Android Phone > SP-404 > Boss MicroBR > Zoom Sampletrak
And then send the samples back to iOS.
Sometimes I could use some an Electribe (EA, EMX) or Monotron to play with extra filters before the MicroBR too
Most of the gear I mentioned is already too 'clean'. The dirty one is the Zoom (18bit DAC, Sample rates: Hi‑fi (32kHz); Standard (16kHz); Lo‑fi (8kHz).)
Miss my ES-1
The motherboard got burnt
Yes, the ringing caused by aliasing is very easy to replicate in Bitjuggler. You can get it to sound almost exactly the same. It’s the quantisation noise that’s mysterious, cause you can match that too, but it requires reducing the bit depth.
This video is a good example of all the sample rates. You hear the ringing on the lowest sp303 rate quite clear, and you hear the quantisation noise on the low sp202 sample rates. YouTube compression will be adding something too, but it gives you the idea. It’s clearest after 2:40, when there are pianos etc, rather than just drums.
The different re-sampling options (sine, linear, inter-leave, s&h) together with the AD and DA knobs are quite handy for tuning in the sound. The extra ‘shark tooth’ icon can add additional ‘crunsh’ as it enhances the frequencies above the sample-rate which is handy.
Mostly I use Linear or S&H, Inter-Leave adds too much ringing for my taste at lower sample rates.
For bit-reduction I mostly use LPC and the amount knob to ‘smooth the gaps’ between the bit-jumps and the drive knob to avoid too much ‘noise’ when the bit-depth is low.
I don’t use any presets and tweak on a use-case basis when needed.
Personally I would like to see Samplers like Koala implementing alternate interpolation modes for pitch-trasposition etc. as it would allow the end-user to shape the sound even fiurther.
Great stuff! Thanks. I’d love it if Koala could implement all the old SP sampling rates. Keep it all in the box.
SOLVED: Regarding the original question of why it sounds like the bit depth is also reducing on SP samplers when you lower the sample rate:
These samplers use RDAC
As a waveform coding algorithm, it is similar in behavior to ADPCM, but is significantly more complicated. It exhibits quantization errors which increase in proportion to the frequency of the signal being encoded. The result of RDAC encoding should be a noise spectrum that is heavily weighted to the >10kHz band.
The basic premises of the RDAC coding appears to be that 1) low-level quantization errors (read that "noise") are psychoacousticaly masked by loud signals, and 2) most program content (read that "music") has all or most of it's pure tone signals below 10kHz.
that's very low fi xD