Don't resample too much! // Digitakt Destroyed my Samples 🤦‍♂️

[@tuftyindigo]

OK dude, time to get less confused about sample rates. As you say, the sample rate is the time resolution of the wave: a 48 kHz (48000 samples per second) wav file can contain frequencies up to 24 kHz. When you pitch up and resample, it plays back those samples faster, low-pass filters the result (to avoid aliasing), and takes 48000 new samples. The low-pass filter it uses is a clean, steep, digital filter, not like a ladder filter you'd find on a synth. So when you pitch up an octave, anything that was above 12 kHz before is now above 24 kHz and gets filtered out. Then when you pitch down an octave, you get back the original signal, low-pass filtered at 12 kHz, with a tiny amount of noise. What you're effectively doing is making the anti-aliasing filter built into the Digitakt work at a frequency you choose by how far you pitch up the signal before you come back down. You're also moving the noise around the frequency band: no filter is perfect, so if you normally get a little bit of aliasing around 24 kHz where you can't hear it, you've pitched down that frequency aliasing to 12 kHz where you can hear it. But the amount of aliasing and noise on any pro audio gear will be tiny - that's why you need to do it so many times to get audible "digital dust". Overall you can get the same effect faster by just using a clean 24 dB/octave filter at the frequency you want and adding some noise before the filter. A quick extra note on aliasing. This is what happens if there's any frequencies higher than 24 kHz when you sample the signal at 48 kHz. It's like a reflection: anything at 25 kHz becomes 23 kHz, 26 kHz becomes 22 kHz, and so on. This is why there's a low-pass filter I mentioned above, to avoid aliasing. When effects like the Ottobit or ZOIA do sample rate reduction, they omit the anti-aliasing filter so that they do give aliasing - the lower the sample rate, the more aliasing. This gives you a bigger source of artifacts than just pitching the sample up and down in a sampler.

[@ghfjfghjasdfasdf]

☝️

[@RickyTinez]

you sir.. are a scholar.. thank you for ALL this info!!

[@callum6224]

I learned about this inadvertently through the game God of War for PS4. When they recorded the world serpent, who has a very pitched down voice, the highs are very clear still. This is because they used a microphone that captures very high frequencies, well beyond 24 khz, and when pitched down it doesn’t eliminate the high frequencies.

[@Michael_Smith-Red_No.5]

*Nyquist has entered the chat*

[@Kebin-Blebin]

it can also add noise based on how it resamples. do two adjacent samples get averaged? does one get dropped? if you don’t pitch in octaves, it could take the nearest sample or do a weighted average. this also adds error, not just due to filtering.

[@davidmoneymaker5649]

Ricky talks a lot about his gear, but I'm curious about his hair! How does it hold so well and why has it been hidden for so long haha

[@christiangreven8991]

the cap exploded :-D

[@aaronpdoucette]

First thing I noticed as well. Then the fingers flying through that machine like he wrote the source code.

[@jonathanjacquet4628]

He came by motorbike 🏍

[@MajorOSC]

It's this new hair product (of circumstances) called covid-19 I think. Although he's looking alot more well kept that most of us that much is for sure.

[@blushingfrieza]

Have you seen conan O'Brien? It turns out more hair makes you look younger.

[@gelsemium15ch]

“The bits are gone bro. They left” 😂

[@RileyGein]

Had me dead

[@KristofferEikrem1]

haha lmao

[@thunderrunner666]

lol no! not the bits!

[@Kebin-Blebin]

sample rate is temporal resolution (left/right on the wave), bit depth is amplitude resolution (up/down on the wave)

[@Kebin-Blebin]

the bucket analogy is correct, but you could also do this for the amplitude resolution by resampling at lower volumes and bringing it back up by normalizing. you could do both at once too!

[@Kebin-Blebin]

@@oOFTJOo hmm, not really, that would add randomness which isn’t correlated with the input. sample rate and bit depth reduction adds predictable “errors” rather than randomness. multiple samples get averaged together or a sample’s value gets rounded up or down a bit. a quantizer running on an audio rate signal would be like bit depth reduction, but could shape the signal in more interesting ways. (idk if any quantizers run at audio rate though.)

[@Kebin-Blebin]

@@oOFTJOo i’m not sure i understand what you mean, but i don’t think it would work. maybe i’m wrong! if you can make it work, might be worth making a video about it

[@Kebin-Blebin]

@@oOFTJOo sorry man i just don’t think that’s right

[@obsessive_discipline]

@@oOFTJOo I also don't totally understand what you're getting at, but phase shifters shift the phase of different frequencies by varying degrees. S&H is something totally different and does not result in any phase shift, since it captures in theory the exact value at each subsampled time step.

[@lff5367]

I really like your Digitakt videos, great to see you enjoy creating grooves and messing/grooving around with this machine!

[@GabeMillerMusic]

I'm loving these experiment type videos. And the hair looks fly as hell

[@MidlifeSynthesist]

Love these experimental videos dude! Also, since it doesn’t get said enough, your videos are visually stunning! Amazing shots, lighting, color grading all aces! Pure inspiration for an aspiring fellow synth youtuber like myself. Thank you!

[@disparity-bit]

I love these technique deep dive videos. Watching the exploration in roughly real time is extremely valuable for understanding the techniques you're working with!

[@birdsiview6845]

In the beginning there was Jack, and Jack had a Groove... This is our house, and our house music...and this is Fresh!

[@DarkMetaOFFICIAL]

@11:18 when you said, " That's tight.. minus that.." it hurt my soul 😂 that little sudden note change was dope af lol.. is like a core component of my beat making style when i do sampling based stuff like this little lofi beat you had. i love clipping notes and samples like that to make patterns That was kinda funny. i see that a lot like NOOO omgg.. i literally painstakingly try sometimes to achieve a certain strange thing and other people are like ooo.. delete 😂 im just like nooooooo pls maek it staahp lol

[@marantz1]

"the most time consuming way to make a crappy bell sound" heheh

[@GrahndDorkus]

FM in a nutshell

[@goeiemorguh]

"The bits are gone bro, they left"

[@Vitaphone]

This is basically an old Ensoniq trick back when samplers had virtually no memory, I honestly haven't thought about this since back in the day.

[@XerosXIII]

my quarantine hair: _eh._ Ricky's Quarantine Hair: _"I AM SEX"_

[@lilmarktube4773]

it's why the sp1200 and mpc60 dudes used to sample records faster and and pitch down . Also to save memory.

[@alexchavosaurus]

I feel like your work flow on the Digitak is similar to how you work with the MPC 3000? "Am I trippin?". I also like how you brought out the note pad and only used it like twice throughout haha. Great content as always mate. This thing's sampler/resampling is wild sauce.

[@tristangruener9571]

Love the digitakt vids!!

[@midi_lizard]

"I am sitting in a room" but make it 2020 synth youtuber content lol. Hair looking great btw

[@zooblestyx]

"The bits are gone, bro. They left." needs to go on a tshirt. Or "Ladies and gentlemen... the bits have left the sample."?

[@exxstacey]

legendary!

[@Michael_Smith-Red_No.5]

Stop staring at my bits. Works for anyone.

[@maxdiamond55]

Thats how the T2 soundtrack sounded so good. They slowed down samples to destruction with the fairlight cmi to add the ambiance.

[@user-ld8ep5qr7m]

Which ones?

[@maxdiamond55]

@@user-ld8ep5qr7m violin sample pitched 3 octaves lower and stretched out. Alex Ball has a video on it.

[@probune]

Always love seeing that digitakt. great stuff

[@newnewnew-q4c]

i saw that the digitakt playlist had been updated yesterday, with a private video added to it. i was so excited i literally emailed ricky asking when it would drop lmao

[@fvmurphy3071]

You can think of resampling 2 octaves up as a pitch shift (which is what you hear). But it very quickly gets out of the audible range. If your sample originally had its main tone at 1kHz, then 2 octaves up (x4) is 4kHz, 2 more ocates up (x4 again) is 16kHz - which oldsters can’t hear - and 2 more octaves up is 64 kHz, which is beyond your sample rate and thus can’t be captures.

[@solojo5000]

"The bits are gone bro". Love this channel.

[@Zeal808]

Always a pleasure, keeping the creativity free 🌞

[@luke3055]

Like a musical mad professor in the lab. Ricky you’ve well and truly graduated from level 9 geek to to full blown level 10, Which we love. Great note taking!

[@pk3776]

Ricky, a genius experimenting ....... always a pleasure to watch and listen 🙏🖖🇬🇧

[@edwardprue]

"The bits have left the building...". If working ITB is your thing, Audacity + Live is a great way to leverage this kind of "manual BRR+SRR" sonic treatment (endless hours of fun, you'll never get *any* music-making done). I guess with Digitakt or Octatrack, you could process and use the results in real-time, if you perform live. Fun stuff! For the technical side, you're replacing a lot of actual audio information with interpolation from the algorithm of the tool you use. Thanks for making videos, I really enjoy watching along with your audio and music-making experiments.

[@MrOuija-rr8kq]

You've etched out a weird lane for yourself. Just mangling samples with samplers. I live for that

[@OscillatorSink]

"Is there a Pre-Delay on here" - I see you're a man of culture.

[@RickyTinez]

ghahaha

[@basgoossen]

Basicly you are reducing the sample frequency and timbre of the sound. By going up an octave keeping the same samplerate (as the digitact does) you just throw away half of the samples (points of signal in time). By going down again it "makes up" samples in between to make the sound longer (this is called integration). By repeating it one octave up then down and up and so on, in a perfect world the sample would remain roughly the same after the first cycle (it doesn't in real life because of how the equipment works, even further reducing definition in each step.) Going up twice and then down twice is a completely different story. By going up twice you do not throw away half of the samples but 3 quarters. Effectively reducing the sample frequency to 11kHz. Going 3 up and 3 down to half of that and so on. This is reducing the definition much faster. You just throw away more samples before integrating the signal again (in other words, the digitact has to make up more of the signal by itself).

[@GeorgeLocke]

This. It's effectively sample rate reduction.

[@GeorgeLocke]

Note that going down then up (in principle) doesn't lose information because you store the slow result in twice as many samples.

[@birdsiview6845]

Reminds me of some records on TCR , Thursday club, from early 2000’s NEW SCHOOL BREAKS king of the beat gonna rock the place!

[@oulivier]

4:04 In the beginning, there was House ! ( and also there was Jack, and Jack had a groove, And from this groove came the groove of all grooves And while one day viciously throwing down on his box, Jack boldly declared "Let there be House!" and house music was born !!!! ) Thanks for your satisfying scientistic music videos

[@JJ-ik1sf]

It would be cool to see a video about resampling reverb/delay tails and pitching them on the digitakt 👌👌

[@HeathHolme]

Cool experiment & dope result ✌️

[@DARIODOESSTUFF]

if the sampling rate is 44100hz, the nyquist frequency is 22050 hz. when you speed up too much you exceed that frequency and you get aliasing, which sound like a sub harmonic of the original. and that's what you are working with in the video. Google nyquist frequency for more info 😉👍

[@unkwn_io3350]

That’s correct. If it was at 192 you would have more room for higher freq less aliases but correct with nyguist.

[@florianoberleitner3933]

That is only half right. Aliasing around the Nyquist Freq is only an issue if the input signal before digital sampling extends beyond it - i.e. a sine with 32050hz would show up as 12050hz. The first step of A/D conversion is always a low-pass shaving of everything above the Nyquist freq - to avoid these aliases in the sampled data. Also the last step of D/A conversion is always a low-pass shaving everything of above the Nyquist freq to smooth the hard steps of the digital signal. What this speed-up-resample-speed-down-resample process basically does: removing a part of the higher spectrum of the original sample. Ricky is correct when he states that the final outcome has a lower temporal resolution than the original. Lower temporal resolution is just a synonym for lower sampling rate. Nearly the same would happen if you simply use a lower sampling rate - two resamples with an octave up and then 2 with an octave down basically is equal to reducing the sampling rate to 1/4 of whatever Digitakt uses.

[@DARIODOESSTUFF]

@@florianoberleitner3933 hmm no it's not just resampling, because A/D D/A conversions are not perfect and depending on the chip you get more or less aliasing: I have many digital synths and everyone shows aliasing at some point. anyway, it's a cool effect and many old school musicians used to do that. I believe junkie xl has a demonstration of those tricks on KZbin.

[@rebours]

"subharmonic of the original" is the most invalid description of aliasing I have ever read...

[@DARIODOESSTUFF]

@@rebours come on it's just to super simplify in a short message

[@enchanterthetim]

The best analogy is like those kids books "connect the dots". If you have 1000 dots you're clearly represent an image, like a persons face. If you reduce it to 500, or even 50 the dots will be randomly removed, leaving strange dots in the place, you may have a spike for an ear, or just remove the ear entirely. The final picture will be a distorted representation. Think of those dots as 44.1 thousand dots, and it can reproduce frequencies up to 22.05hz because it has twice the "dots" as i'm calling it, to represent that frequency as a wave. It's not representing those 20k waves well though, that's one of the reasons 48k is better.

[@IhorMedia]

Oh man! I need to try this on some deep house drum loops!

[@GloPhase]

that beat is gorgeous

[@jasonmelstad]

those higher pitched kicks at the end sounded like bubbles popping. if you detuned them slightly and gave them a subtle reverb tail it would sound awesome i bet

[@KristofferEikrem1]

Ricky, could you please post more jams? They are super cool and inspiring!

[@mafrodriguez7327]

You added a new keyboard rack! Make another studiooo tour. I love how you explain your hybrid studio. I feel related always pluggin and unpligging thing finding new ways to work trying different workflows

[@linkster9]

Using your analogy in reverse - turn that one bucket into many and then back to one. It works much saucier.

[@julesmadrid6546]

Yo, your hair's craaaaaazy

[@IhorMedia]

Hey Ricky! Got some good laugh while watching it :) I've heard people prepare samples for Digitakt. Like cutting some lowend, or even doing some processing before sampling into Digitakt. Would be interesting to know if you do such preparations?

[@RickyTinez]

sooommmeeeeetimes. very rare. Since this i do mainly drums i don't find that issue too often. and my other samplers have filters that i can use to resample with filters on it when needed

[@IhorMedia]

@@RickyTinez yes, makes sense. Since I got the OT I also use DT for drums mainly. Thanks for replying!

[@Manusmusic]

I believe what happens with the audio mangling is that it actually removes overtones one by one. Think about this this way: Original sample has for example main tone on 440 Hz, and then overtones on every doubling point. When you move it by octave, you move them all by the doubling distance removing the overtones going past 20kHz. Then when you move it back, it comes back without the information (Plus whatever the gear does while resampling(gain, eq, dist))

[@charliefoley4794]

I'd love to know, where do you get those TRS cables? I'd love to colour code all my shtuff

[@heavysystemsinc.]

Basically the sample rate is how many times the analog waveform is...sampled. The bit depth is how many steps there are in amplitude for each of the samples taken in the duration of the sound. So 16bit (bit being on/off or binary value of 1 or 0) means there's 16 different values each sample can have. Now when something disappears due to pitch, that can be a combination of going outside the range of human hearing, outside the range of what the machine can playback (remember, the output of a machine is ALSO digital and ALSO sampling the internal sounds before it poops them out to the digital to analog convertor for your speakers/cans...so there's a thing where if the original sound doesn't 'line up', depending on how the machine handles this, the important bits can get 'lost in the cracks'. It's also the cause for digital synthesis to alias...the machine it's running on has a DAC that's not compensating for sample rate changes that end up 'chopping off' or truncating waveforms, so you get weird resoanance and harmonics that are tied directly to your machine's always-on sample rate. I hope I explained it. None of that really matters if you're just making music and intuitively understand how resampling affects sound. None of us need to really know this technical mumbo jumbo to make a fresh beat. And even still, knowing this doesn't make it easier to predict exactly how the sound will turn out, we still gotta resample and see what happened.

[@JimmyA459]

Reminiscent of Korg Poly 61 ravey stabs. Love it

[@Tokuushi]

Thanks for educating us! “Knowledge is power” ✊✊✊

[@scottsweatshirt9753]

No one comments about the aesthetically pleasing and unorthodox Hosa cable color scheme used by the inimitable Ricky Tinez. The yellow and cornflower blue (from the Ottobit demo) versions are ace.

[@dziwic]

'The bits are gone, they left' has to go on a t-shirt

[@FormatWarsBand]

That is some serious Digitakt muscle memory! Clack clack clack boom

[@wickeddubz]

Ricky: clack,clack,clack - makes cool beat. Me: clack,clack,clack - 4.50 total, do you need package?

[@BigFknRobots]

Have a read about the “nyquist frequency” It’s the highest theoretical frequency reproduceable from your sampling rate For 44.1khz, it’d be 22.5khz Which is outside human hearing (Which is said to be 20khz, but is often lower for adults) In practice, if you want to accurately recreate a frequency, you want a bit higher than double... otherwise the sinewave will drastically drop in volume (or disappear entirely) if the phase is a tiny bit out from the sampling. Eg. Say you sample 10khz and record a sine wave at 20khz Loudest part would be if it samples the level at the highest point and lowest point of its cycles. If it was quarter a cycle out of phase, it’ll just record the zero crossings... So I think that’s why they go with 44.1khz or 48khz for CD’s and DVD’s when they say humans hear up to 20khz That buffer ensues that the 20khz is actually a well produced 20khz If you start with a 440hz sound You pitch it up an octave, it doubles the frequency to 880hz, up again and it’s 1760hz... Etc You pitch that original sound up 6 octaves, it’s around 28khz and it’s gone But complex sounds can have frequencies all the way up the frequency spectrum, it’s not just a singular frequency. Say your sound has frequencies up until 10khz, you double it, it’s fine, frequencies are now up to 20khz, double it again, you’ve just pushed a chunk of the sound off the edge and it’s not coming back. Even if your pitched up 3 octave sound has a fundamental frequency of 1760hz (well below 22.5khz) The effect you get of losing time resolution and higher frequencies is essentially the same as rate reduction in a bit crusher There’s another thing of aliasing, which can be another topic, but that high frequency content that is too fast to be sampled, can get picked up as a lower speed waveform that acts as a kind of distortion That contributes to the sh*tty lofi destruction going on. In the days of early samplers (with low bits and low sample rates) People would sample lower notes of their synths, as it would sound better Pitching up, throws away information Pitching down, brings more noise or lack of information from above the nyquist frequency and brings it down into the audible range. It’s inventing information from nothing. A related phenomenon, is time-stretching. Speed a beat up, you throw away information, still sounds good. Slow a beat down, you get the stutters and other artefacts. Conversely Pitch a beat up, it’s inventing samples to fill the gap, pitch it down, it’s taking more time to play what it has.

[@BigFknRobots]

If you’re doing extreme sample transposing That’s an area where working in higher sample rates can really be a benefit Eg. 192khz has a nyquist rate of 96khz Which is like 5 times higher than you can probably hear 96khz has a nyquist frequency of 48khz Etc It also lowers the problem of that aliasing distortion, by pushing it so far from the frequencies we listen to (this is also why some digital audio people talk about “anti-aliasing filters” in their products... it gets rid of that undesirable distortion)

[@cheeyse9857]

This is insane - Would love to see what you could do with an analog rytm!

[@k-g-p]

Man! I just love you man... 😊

[@Max-zc4de]

Love the digitakt content, your vids are half the reason why I own one! Any chance we could get some more hydrasynth stuff?

[@scholzdigital]

Where did you get those beautiful orange connector cables from?

[@beepboop4846]

ngl, im digging this pseudo Bruno Furlan hair styling

[@ernestomercado9181]

Hey man, I love your vids. They’ve helped me immensely to learn the digitakt, and the potential for the octatrack. Unrelated question though, Where do you get your 1/4” color coded cables?and are they balanced?

[@DarkMetaOFFICIAL]

bru have u ever tried using a granulizer? (probably) but if not, take a sample, pipe it thru a grain fx unit then export your sound- The process would be: Send sample thru grain fx--> Adjust grain lengths, number, + sizes, --> Resample (export sound) --> Then do the same up/down pitch/speed process you did here, then re- granulate, re-pitch or speed back up or down to original.. and hear so many crazy fragments and textures in the new sample. Then these remnants and artifacts can also be fractally repeated, either stretching or compressing.. it's quite ridiculous. 😂🎶 endless fun This creates so many combinations you can experiment with. do: length low, then go up, low length, back down.. length high, than all same: length high, then go down and up, length high, down twice, length low.. etc. then change 2nd parameter, etc. going through all options (which is probably impossible) exponentially unlimited options and all unique samples : ) epic af

[@t7H2si0vß2]

He should've pitched up an octave then pitched down an octave multiple times to pull the artifacts out. Pitching it up over and over again is just going to push it above nyquist, basically deleting ALL information, hence the silence at 3:10.

[@UncleC1025]

isn't that what he did in the other video?

[@t7H2si0vß2]

@@UncleC1025 which other video?

[@t7H2si0vß2]

@@UncleC1025 what are you talking about

[@andres.n.1706]

Awesome bro

[@squoblat]

This video was brought to you by an acid trip

[@physicsandpoetry]

Isnt this basically roundabout bitcrushing?

[@MartinRuss]

No. Bitcrushing reduces the number of bits that are used to represent the audio. This is more like 'bandwidth crushing'...

[@danielgosling9900]

Can you do a review of your new headphones? Audeze right?

[@GeorgeLocke]

Try doing this in audacity with the spectrogram. When you go up an octave, you chop off half the information, and then it doesn't come back when you come back down.

[@arjanpetersen]

How is generally your work flow with ableton and digitakt? I recently got the Overbridge stuff running on ableton. Somehow I think I’m better without overbridge.

[@Xealexx]

Hey Ricky, have you tried messing with the Wavestate?

[@julesmadrid6546]

You got any reason for using mac instead of windows? Love your videos, muack

[@Claidheambmor]

Built for audio and video editing.

[@panskibinski]

The case with big up and down pitch sampling is that you will loose high freq timbre part of the sample due to sampler limitation and fact that selected (re)sample freq range moves from 20Hz-20kHz to e.g. 30Hz-30kHz. Also the as you described with the water buckets you will loose samples per second ratio which will be then re-calculated by the sampling engine when going pitch down You can check how much easily by resampling white noise 3 scales up and then 3 scales down

[@isaac.anthony]

Jack boldly declared: Let There Be House!

[@isaac.anthony]

and then you make hiphop, LoL

[@repasiv]

"I mean this is the longest way to create a crappy bell sound, but it worked" :D

[@Amlanting]

So sample rate in this case means the same as sample frequency. In a way this is comparable to FPS in video. For example: In video 24 frames per second is needed for a nice moving flow. Anything below that will result in choppy video. For sound this is a minimum of 8000hz. Below this rate the playback will not be distinquishable, there is just to less information to make out. 44.1Khz is for example the standard for a music CD. Personally I like to compare the sample rate to a container of sorts. Let's say a bucket. The bigger the bucket, the more samples I can put in it. Then there is also something called Sample Size. (Sample depth) Most common are 16bit and 32bit. The sample size influences the quality of each individual sample by range of amplitude. The audio sample can represent a higher range of amplitudes with a higher bitrate. The more bitrate, the closer the recording will be to the original in terms of acoustics. So.. let's fill the bucket with tiny marbles. Each marble represents a sample. A bigger marble can hold more data for one sample. But it also means I can put less marbles in the same bucket. And this is where bitrate comes in. Which is basically the samplerate and samplesize mashed together for your convenience.

[@beepboop4846]

1) i feel like what happens is that when you resample and stretch/squash the sample, i think that theres not necessarily loss of bitrate or anything, but more that theres too much data being thrown at the recording, that it looses some frequency 2) i would love to see a comparison of the two now established methods and see how differently the processing happens

[@rixouney]

"Yes, yes, yes!"

[@aloharay]

Using a visual example, i think of your original sound as a picture ten inches wide, sliced into 100 vertical slices. Resample it by removing every other slice, leaving you 50 slices and a 5 inch wide image. Resample again by removing every other slice, now you have a 2.5 inch image made up of only 25 slices of the original picture. Now stretch those remaining 25 slices out to the original 10 inch wide picture size. How much of the original image is there? Can you see the original image intent with only one quarter the original's information (aka slices)?

[@BeefyTime1]

Sounds like it needs a message like Akais s900 when you resample at half the bandwidth, “ENOUGH ALREADY!!!!”

[@MrOuija-rr8kq]

You turned your Digitakt into a Fairlight

[@zbsfm]

how is this different from downsampling? doesn't the dt have a sample rate knob?

[@DamnHeadHumpers]

super funky beat

[@valkyrie9042]

What are the headphones he’s using?

[@alchemist2816]

Can you you try the same thing with some vocal samples????? soul music , rock etc..

[@jtreezy255]

If you had to choose between Digitakt and Blacbox, which one?

[@itsdrawkcab5899]

Is this the same thing as lowering the samplerate, just like what you would do in a plugin like decimort?

[@mspath1001]

Hi Ricky i would really love to see a video about the synthstrom delige of you.. i think you would love this thing..:)

[@cold_fashioned]

There needs to be a second channel for Ricky's hair! The jealousy is strong.

[@TheTechnobear]

indeed you loose information each time - also if you take to very high frequency you'll hit nyquist, so the wave will fold... so when you pitch down again, still a folded form - so different timbre - no?

[@VincentsVideoVisions]

Why does Ricky's hair look like Whis from Dragon Ball Super?

[@zzirjukez2899]

I really want to use a digitakt

[@Soundwrecker]

Nyquist theorem says what you are doing is applying a lowpass filter at 1/2 the lowest effective sample rate, with the speed up considered. 11KHz? The whole low end should be unaffected, unless it is resampling analog w/ adc.

[@argiletonne]

I did a lot of this using energyXT, one of the few fun things,. makes great tiny sounds if working on computer noises; did all the windows package noises a long time ago with the windows vista premium edition that comes with tinkering sound pack. The tinkering sound pack got me thinking I could make my own. Or sounds that will sound good for cellphones. Tiny sounds can be fun. I made a sound pack for myself back then I called 'fczkfuru'.

[@lundsweden]

Sounds like a steel pipe!

[@drumatic]

More digitakt vids please! When u getting an SP1200? Ha

[@tiouip]

That’s a big pair of, headphones.

[@panskibinski]

Try adding noise + overdrive with the highest pitched sample , resample and go pitch down ;)

[@RichardJNeo]

In the beginning there was JACK!

[@jerrypizzini]

I was wondering about that video.

[@PandaPotPies]

You should try the love hertz lossy vst I think it’s called or good hertz

[@Alphonso_026]

Down the rabbit hole again are you?! 😂🙈