Clarification: on the point of the number of samples per cycle - 2048 is what the Midi Control Centre expects as an audio import format - not what appears to be ultimately stored on the MicroFreak (which is 256 samples per cycle). From the perspective of generating your own wavetables from audio files, this isn't particularly relevant, but it's interesting to note that your audio is internally resampled before it goes to the MicroFreak.
I was puzzled about your math in the video because I read about the 256 samples in the manual. How do you know about the 2048 samples for import? Why have they done it like that? It doesn’t make sense to me…
The 2048 samples is mentioned on page 102 of the updated manual, regarding importing wav files in the MIDI Control Centre. The software must be resampling, because I spent the afternoon investigating the format that you can export and confirmed that it's 256 samples. I might to a follow up video when I've learnt more.
For me importing Serum wavetables works nicely. AFAIK their cycle length is 2048. That’s probably why Arturia chose it for the input format. The pitch also seems to be right.
Im convinced that the microfreak has the highest ratio of price to features of any synth i have ever seen. Its ridiculous what this thing can do, all for $350. And they are STILL putting updates out and adding more features. Absolutely my favorite synth ever.
@@paulb1953 , technically you are not wrong, but it is one thing to use an open source algorithm and it is another one completely to integrate it in a pre-existing user interface and adapt it to whatever controls configuration your system allows. So one should not believe that this was just a matter of uploading bits of open source code: it did require serious integrative development by their in-house programmers. And whatever time they spent doing that will only benefit their brand image in the long run, and not provide any immediate return on investment, since we are talking about machines that were mostly already paid for by customers. The way I see it, the guys at Arturia’s are using the MicroFreak as a test-bed for new potential sound generators or user interface configs that may end up in their higher-end synths, all while collecting users’ feedback from their existing MF customer base. They certainly have a lot to gain from that in the long run, but it is commendable to see a company bet on long-term image and customer feedback to improve on their products rather than relying on pure marketing strategies. I wish most technological companies would have the same work ethics.
It's really good to see consistent and ongoing support on an inexpensive instrument like this. It's pretty crazy to compare it to how it was on release day.
@@inevitablecraftslab, it is one thing to use an open source algorithm, it is another completely to integrate it in a pre-existing user interface and adapt it to whatever controls configuration your system allows. So you should not believe that this was just a matter of uploading a bit of open source code: it did require serious integrative development by their in-house programmers. And whatever time they spent doing that will only benefit their brand image in the long run, and not provide any immediate return on investment, since we are talking about machines that were mostly already paid for by customers. The way I see it, the guys at Arturia’s are using the MicroFreak as a test-bed for new potential sound generators or user interface configs that may end up in their higher-end synths, all while collecting users’ feedback from their existing MF customer base. They certainly have a lot to gain from that in the long run, but it is commendable to see a company bet on long-term image and customer feedback to improve on their products rather than relying on pure marketing strategies. I wish most technological companies would have the same work ethics.
This is extremely exciting, and even more so because you took the time to blaze the how-to trail and deliver it so quickly, thoroughly, and clearly!! 👍🏻 🥇
ZERO clue why I wasn't subbed til now, but we all owe you AT LEAST that for this one! THANK YOU for diving right in there and knocking out the long division for us!! 🙏👍
Honestly the Oscillator Knob Speed is a major update for me. Many times I wanted to use the MicroFreak but I turned to another synth in my studio because I knew it would be a nightmare to turn those little knobs indefinitely to sculpt the sound. Now it's much better 🤓
It’s Arturia’s continued support and videos like yours that lead to me buying the microfreak around 2 years after release. A good way to keep sales going and improve on an already epic synth.
Creating custom wavetables is something that I do for some of my other synths - mostly Sequential Pro 3 and its digital oscillator. There is a great thread on Sequential forum about how to create wavetables in details, lots of very complex ideas and technical stuff, but for the basic, make sure you waves are long enough, meaning they fill entire cycle, make sure you align them on zero crossings, and also make sure these do not contain lots of noise, except if that is what you intentionally are going for. Make sure scw level is boosted enough, and try to have SCW in 1024 bit format. There are zillion waveforms across the internet, some of you Octatrack/Digitakt users might already have huge banks of single cycle stuff, here you just need to choose those you like the most and compose the wavetable - however, also think about morphing, meaning, the order of forms. This can actually mean a lot when you import file for scanning WT.
Including the noise modulation in your sample will have contributed to it sounding thin. Noise only sounds like noise if it's actually changing constantly. Each of the cycles in the wavetable is looping the same bit of noise which will manifest as high harmonics. It's a cool effect as long as you're happy to roll with it like you did.
It was the over-long brass sample that was particularly thin - the sample I took from the Microfreak itself came out pretty well (albeit with those extra high harmonics). I like how the high frequencies dance around as you scan through the wave table - like you say, it's a cool effect.
I love it when cheap synths get amazing updates. I hope the user Wavetable gets me using it, they put it on the Peak ages ago and I used it for one day and then never again. Everything just sounded thin and nasty, no matter what I tried to do with the Wavetable editor. Creation of Wavetables is much easier in the Peak, using Components you can just drag and drop wave files into the editor, 6 per Wavetable, and it will allow you to audition it in the computer before sending to the synth. Excellent demonstration.
I jumped into the microfreak-game at v3 and thought, thats perfect with enough sound-possibilities. And now this update comes up and blows my mind! Thx arturia and thx for the wonderful tutorial ja ❤
looking forward to playing with this - just updated this morning. I hope in future updates on top of wavetable they do a granular oscillator that allows import as well
@@OscillatorSink I think you could probably limit sample size and make it effective. the thing about granular synthesis is that you don't need a really long sample, even something like a 1 second sample can give really interesting results since you are usually shrinking rather than expanding the length of the grains
I am late to the game, here, but this has to be the single most useful technical video I ever saw on the MicroFreak Wavetable sound generator. I left that one a little unused for lack of interest in the wavetable presets, but your « trick » of 1.5 s signals detuned by -21 cents really opens up a whole new world of exploration. They definitely SHOULD integrate something like your explanation here in the user manual, back at Arturia’s. Thanks a lot for sharing your mojo !
Your detailed explanation of Wavetables is excellent, even better is your tutorial on how to create our own user Wavetables. Thank you for putting this video together
Thanks for working out the numbers on this. Sounds awesome with your 'mouth noise' 😄 Definitely going to try something like that on the soundtrack I'm working on 🤜🤛
You're very and thanks! It's probably some deep seated synth prejudice, but that one sounds particularly "wave-table-y". Just be warned - I had to pull out my best death growl to get low enough!
You've done a more than awesome job here, ty for you time and effort(ps I really appreciated the maths part). Oceans 11 is a good pedal for monosynths and it's not too bloody expensive.
great videotutorial with outstanding tips and informations. hope to handle my arturia microfreak devices from now on better. Thank you for your time! I will say that I LOVE that current Version 4 Update/Upgrade of the Microfreak - it brings me to buy an 3rd Microfreak for my studio. it is always good to have enough microfreak on board if you going to be synth creative ;) ..with "some more of them" I can arrange so much more different at the same time - especially with that wavetables-Feature of the Firmware 4.0 now !! Microfreak is currently really outstanding as algorhytmic synthesizer on the market. of course the product looks and feels not as an AAA-Product - but keep the price of it in mind before you judge! it is an really cheap/low price for so much innovation and features. thank you Arturia!! I think that the Microfreak will having an following Microfreak II in the future - and of course will having an good place in our heart of the most innovated synths.
That’ll be me tonight after work then… Arturia nail it again with this ever increasingly fabulous wee beastie! Cheers for an excellent overview man and particularly for the maths. Peace ✌️
I wonder if the wavetable editor in Vital could be used to create wavetables for this Edit: Yes it can! You can export them as .wav files. They run for about 5 seconds as opposed to the recommended 1.5ish, but they work decently from my limited testing Edit 2: The length of the .wav file is exactly 4x1.486 so halving the tempo twice will get you to the exact right length, although converting it at different lengths produces different timbres so it's an avenue to explore by itself. So, yeah, Vital doubles up as a wavetable creator for the Microfreak. Gotta give it to Matt Tytel, that synth is amazing in more ways than one haha
I guess you could also just edit the wav down to the right length (probably need to use something more surgical than a DAW for that - could probably script it easily enough in a few lines of python...) and treat a single Vital wave table as multiple MF tables?
Thanks Kryten, scrolled through the comments to see if anyone had tried Vital. And here you have. I grabbed Vital in march, but my laptop died, and the final piece in the puzzle should be arriving soon for my desktop build.
@@OscillatorSink From further experimenting, it seems like throwing the wav directly from Vital into the Microfreak works best, speeding it up by changing the tempo introduces distortion, although complex waveforms seem to have trouble no matter what I do
@oscillator sink Your excellent tutorial has inspired me!!! I got to doing some experimenting of my own and discovered that your audio length is much too long for more traditional sounding wavetabless. If you take the 1.486 audio rate sample and look at a .046 selection there is a lot going on in one cycle that will get lost in the resampling of the Microfreak. I happen to love the thin sound of that for cold digital sounding pads, but I'm sure not everyone does. If you want to get more traditional beefier sounding wavetables you should only take .186 seconds slow it down 8 times (x.125) then trim to the exact 1.486 seconds. I am sure this could be honed in to be a bit more accurate. I hope this tip helps for people looking for more traditional sounding tables.
I really enjoyed watching your well informed video, it's taught me a lot about this device. I have been putting off buying Micro-Freak for ages, but now this has inspired me to go ahead and treat myself to one. THANKS
Jolly good stuff Mr. Sink. You and Mr. Blake seem to be killing it with these MicroFreak V4.0 update videos. I've got so much gear on my list to purchase now and this is not making it any easier!
@@OscillatorSink Yes, I watched this directly after his video and there's so much good info to be had. I am definitely interested in all the MIDI CC# modulation of parameters that's possible. Hmmm will have to prioritise what to get next - the 'Freak or a Digitakt (plus some other modular goodies)
Very useful - I've recently got to play with a MicroFreak and this saved me a lot of experimentation. I found one error in your calcs: the frequency at 15:35 should be 21.5332Hz, not 21.5533Hz. If you use your figure, the final cycle ends on a non-zero value and causes a glitch when looped. As an experiment, I created a wavetable that was just pure sine waves at 21.5332Hz to see what happened as I changed the position setting. The loaded table on the desktop app showed just the sine waves but, stepping through the table position, more and more distortion and higher frequencies were introduced, so the synth is using some algorithm to change the wave shape based on position.
I don’t have any regrets happening upon a used Microfreak at a Music GoRound store in Duluth MN six weeks ago…and I snagged it! Now if anyone gets a used one, KEEP THE RECEIPT as you’ll need it to register that Freaky little thing to your name ( involve the Arturia Support Team right away and within half a day, it’s done!👍🏼). Oh, no: it’s got capacitive touch keys! I can’t adapt! Well, learn to like it! Gives you finger modulation capabilities that you’ll soon…learn to like. 😆😆 And you know already that you’ll need an effect pedal or two. But if you don’t have the money to plunk on a Line 6 or Strymon (for instance), I’ve found that a couple of good beginner, budget-friendly multi-effects pedals are the Mooer PE100 and the Zoom MS70-CDR. So, with those, you can “die a happy man”, but try to NOT die too fast! 😄😄
Thanks for checking in - the last few months have been especially "2021"-y, which drained a lot of my time and energy, but I'm doing better now - hence more videos!
thank you, that was helpful. I couldn't figure out why everything I loaded into the wavetable sounded exactly the same. Because i was playing in the mid range and everything was pitched up like crazy. Once I put the freak on the lowest octave, i was hearing all the weird stuff I loaded up
very high quality and depth demo, from my point the microfreak is true warrior from arturia, must buy the limited white vocoder version though it high priced and hard to found. the only thing need upgrade is limited voices, hope arturia can do something with it in the future.
i wonder how much free space there is on the MF. I wouldn't mind shelling out for expansions if necessary, particularly stuff like built in effects in the menu even if they're pretty rudimentary to save on CPU.
There’s no doubt that the Freak will be considered a classic, you don’t mention the bug fix that keeps the oled from flashing random parameters!! Huge improvement there!
Looks cool, but I feel like it would be nice if chord mode recorded the notes in the sequencer, so you could turn chord mode off and edit the notes of the chord in the sequencer
This is pretty cool, lots of potential. Have to say though, I'm surprised that with all the big upgrades there haven't been improvements in some smaller things...like for example I don't know why we couldn't have decay and release controlled separately with a shift function. Not going to complain about them doing big things but I'd like to see other changes too
Wow-thanks-well done! I love this little synth-endlessly usable with amazing capabilities! Once I turn it on, I’m gone for hours. Bravo Arturia!!! 😄 BTW: have you tried inputting with the mic attachment?
Really cool in depth demo of OS4 - kudos! We were really surprised that Arturia is developing this little beast for so long already, what will be next?
Nah, its a good synth all around. Was going to do the same before the last update. And kept it.. Arturia has its issues but good updates arent one of them.
Great review. Do you think the export of wavetables from something like Fors MaxForLive Device Chiral might work for this, or maybe the ModBap Osiris Wavetable editor work?
@@OscillatorSink yeah i did but was directed to a company in London and they wanted to know the warranty. I bought it a couple of years ago second hand on eBay
Waldorf NW-1 does user wavetables from audio unput, tape recorder style. Could MicroFreak software be doing something like creating a formant of what you put in, not blindly snipping a certain number of samples, regardless of crossing points? There is a definite metalic timbre to all wavetable synths, though in my opinion, less difference between various wavetable sounds, which lie in the same sound world.
ha i just chopped up (audacity) a random piano sample i found on a fourtet track and threw it into the mf & it sounds good. a lark singing recorded on my phone and then pitchshifted wasn't quite so successful.
As expected a clear presentation with wonderfully enthusiastic step-by-steps. I was a bit underwhelmed by the thinness and relatively narrow-range timbres with the audio sampled wave tables. I was wondering if there are lush, fat, wave table options available in the wild? Or is this not really in the wavetable ballpark? I also raised an eyebrow when in Midi Control Center, you added your samples and all the default wavetables disappeared. I assume with a less cavalier add and file management you can also retain the defaults to your liking. Lastly, it seemed like an in-joke on your more than one plea for reverb/delay. Surely there was something within arms reach?? Ha!
Less-thin = simpler waveforms. So, yes, absolutely you can do that, and the default waveform is an example of that. But to get something that simple when importing audio, you will need to be very precise, so it's probably "easier" to achieve this by directly designing wave tables, rather than importing audio. Personally, I err towards the thin, complex and weird side of things with wavetables - things that are hard to do by other means on the MF. On the file management - you can export each wavetable individually and import them as required - but a new bank is always empty (the factory bank can always be re imported and you can store as many banks on your computer as you like - only one at a time on the MF). Yeah, the delay and reverb thing was a bit of a joke - I hadn't set one up, but wasn't going to stop recording to fetch one - I was working on a slim deadline with this video!
