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.
Synthesis types
I suspect there are several out there that are with me in not quite understanding the different synthesis technologies, so I hope this might be of some help...
https://en.wikipedia.org/wiki/Category:Sound_synthesis_types
Comments
Ooh nice link, thanks!
FM and Wavetable synthesis are my personal favorites.
I couldn't say what my favorite is really, as I am not clear on how to discern one from another. That site would have been even better with audio examples. I'm making it my mission to learn each a bit better though.
It would be cool if the Wiki articles gave examples of famous songs/works which use each synthesis type.
Folks should post videos demonstrating/explaining different synthesis types. This is a wonderful video about the history of FM synthesis and the DX7 specifically:
From earlyer accounters with synths on computers and later and still I always mused on how any instrument or synth type emulated or sampled sounds would reveal harmonies and wobbles and everything, when played low register and slow ... (and loud sometimes), equing, fx- and sequence - sculpturing. Old style nostalgic and futuristic utopian, detailed sophistication, sometimes by pure chance ...
I'll check out the video after work @CalCutta. In the meantime, I missed the link, on the link I posted above, for sound examples.....
https://commons.wikimedia.org/wiki/Category:Sound_synthesis_types
Thor is pretty cool, as an educational synth, with the expansive oscillator section- analog, fm, phase distortion (Casio CZ), wavetable, noise.
For that matter, so is iOS, with the plenitude of affordable synth apps, of different types.
I would guess the order of use that you actually need to know is is something like:
Some of the others only really matter if you're trying to simulate a real instrument, while the rest seem pretty esoteric (Scanned Synthesis).
Agreed 100% and I'd consider this the single most underrated part of iOS music. It's the cheapest way I know of to get experience with a variety of different synthesis types.
Thanks very useful share.
@tja
)
problem is with point 2/ “subtractive synthesis” - in my experience vast majority people are understanding this classificstion incorrectly (and so does GPT too cause it’s knowledge is build on people knowledge
Suntracrive is description of whole synth where rest poonts are more description of exclusively oscillator section.
You can use and of ocillator types 1,3,4,5,6,7,8 in substractive synthetizer by applyinh filter on sound created by oscillator.
So:
1,3,4,5,6,7,8 - synthesis types, eg. oscillator types
2 - synth type
Just a reminder ChatGPT’s expertise is how to say things not knowledge. It is a language AI not a truth AI.
It often generates well-articulated language with factual errors.
in my experience it's good, after it answers you, ask once again something like "are you really sure this is correct / right ?"
often it then apologises a provides correct answer in ase previous one was wrong
Not entirely true .. It works exactly same like people (brain). It answers you what he (she, it) thinks is true, simply based on object relations stored in neural network. Basically same thing brain is doing, same like people may remember thing wrong, AI can too.. Cause same as human brain, it doesn't store memory like data are stored in database but more like interconnected network of related objects and events ..
But yeah.. in some way it "just" understands language - but not just ordinary language we use to speak - language as wider abstract concept, set of rules describing some set of objects are relations between them.
It may be traditional language, programming language, understanding of images context, mathematics, ... those all are basically "languages" .. Internally it may even develop own special language for translating internally between different types of languages.
Developing complex language was actually main impulse why our brains evolved from more primitive into modern human stage.
yes but if you know properly how to ask you can in most cases mitigate wrong answers and navigate it to correct mistakes or even admit he/she/it doesn't know .. As i mentioned above, for fact-sensitive topic it's good to re-ask if given answer is really ok, or even paste back answer and ask for validating corectness of that answer - that's also working pretty well.
As every komplex tech, you need learn to use it properly, it's not just that easy like most people think...
You simply cannot rely on ChatGPT to be factually correct. If one does not have the expertise to evaluate what it produces, one cannot count on it being correct. Many people are swayed by the quality of writing and don’t realize that convincing writing can be factually and/or logically flawed.
ChatGPT’s expertise is writing.
If one doesn’t have sufficient knowledge and analytical skills, open-ended queries can lead to factually incorrect but very convincing writing.
While asking it to correct itself SOMETIMES improves the factual accuracy, it doesn’t always.
Working “pretty well” is a low mark if one is relying on the information. If you provide good information and context, it is a useful tool.
It is worth reading about the experiences of content experts analyzing the output to queries about which they have expertise.
question:
explain me bell’s inequalities
gpt 3.5 answer:
Bell's inequalities are a set of mathematical inequalities developed by physicist John Bell in the 1960s. These inequalities relate to the correlations that can arise between two particles that are entangled, or quantum mechanically connected, and how these correlations compare to classical physics.
In classical physics, the properties of two particles are considered to be independent of each other. However, in quantum mechanics, two particles that are entangled can exhibit correlations that cannot be explained by classical physics.
Bell's inequalities provide a way to test whether these quantum correlations are truly non-classical. The inequalities show that, if certain assumptions about the behavior of particles are true, the correlations between two entangled particles cannot exceed a certain limit. If the correlations between two entangled particles violate this limit, it suggests that the particles are exhibiting non-classical behavior.
The violation of Bell's inequalities has been observed experimentally, providing strong evidence for the non-classical behavior of entangled particles and supporting the principles of quantum mechanics. Bell's inequalities have had a significant impact on the development of quantum theory and our understanding of the fundamental nature of reality.
—————————
If you ask me, pretty accurate general description.
Yeah GPT is not wikipedia, that is not what i am trying to say. But it is pretty good in explaining general ideas and concepts, or providing overview of some topic (often with relatively good details) - so you get basc point and then you can use wiki and other sources to dig deeper if you want.
If you know how to use it properly (and you combine it with other sources, or you use things like AutoGPT) then it is very powerful tool.
My point is that it is unreliable for fact and analysis (it is not designed for that — so it is user error when people treat it as in anyway authoritative).
Sometimes it gives good information—sometimes incorrect information. So, unless one is capable and willing to research the accuracy of the information, one can easily spread misinformation.
Finding examples where it does well doesn’t make this less of an issue.
Quite a few savvy people have pointed out that the big downside of ChatGPT is that people often accept what it churns out as true because what it churns out is well-phrased.
A couple of scientists had fun having ChatGPT churn out essays about their research. The essays were compellingly written… but full of factual errors and even made up sources and quotes. People other than experts were easily convinced about wrong conclusions.
Throwing this oldie in here because it relates somewhat to the topic. Of all these synthesis types, which do one need? Totally subjective ofcourse but helped me out.
/DMfan🇸🇪
1 month ago, chatGPT was claiming that the Earth could sometimes be between Venus and the Sun.
I asked several times the question in different ways just to be sure it was not a misunderstanding.
Asking again tonight and this has been fixed.
Always be cautious 😅
You beat me to it. LOL. Did you know you can use ChatGPT to generate patches for specific synths? Crazy.
Try getting good results out of google. I'm not proposing a bake off. I'm just suggesting that when people scream "ChatGPT got a fact wrong!" …of course it did. I'm using ChatGPT and google side by side, and sometimes to provide a better result than the other service.
Yet again wrong. “Subtractive synthesis” in one context with FM, Wavetable or Additive synthesis.
Same mistake repeated in every article. No wonder that GPT then produces same mistake.
It’s like to say:
The types of cars are: Sedan, Hatchback, SUV, VAN, Crossover, Electric vehicle
Do you see the point ?
Any of these : FM, Wavetable, Additive, etc can be part of Subtractive synth (good examole is Butter Synth)
It's important to distinguish between types of synthesis (in relation to pure oscillator section) and types of synths.
Also most of dsp synths with just simple saw/sine/square/etc waves in oscillator section are technically by implementation wavetable synths because that’s simply way how in most of dsp synths is oscillator implemented - as wavetable with oscillator waves :-))) They do not “generate” wave in realtime with some math magic - they just have internally stored wavetable.
Example - Digitone
So if you ask me, Digitone is subtractive synth with combined FM/Wavetable synthesis in oscillator section
Example of pure FM synth is DX7.. example of pure Additive synth is VirSin Addictive ...
I would say much more sense makes splitting types of synths to 2 main categories - West Coast (developed by Buchla) and East Coast (developed by Moog) .. West coast synths use harmonically rather simple oscillators which is followed by wave folter which adds more harmonics, East Coast is classic subtractive synth - it uses oscillator producting harmonically rich signal which is then reduced by filter.
In both cases, as core oscillator section, can be in theory used wavetable, additive, or FM synthesis - of course in West Coast type, from practical reasons, it makes sense to keep complexity of sound produced by oscillator on simple level, cause complex signal sent through wavefolder ends mostly like noise
)
Example of pure West Coast synth on iOS is Ripplemaker.
TLDR: Does it have classic resonant filter section ? It is subtractive / East coast / synth. Doesn't ? It is West Coast synth
this one line .. "subtractive" is not synthesis type, it's synthesiser type
subtle but crucial difference