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 Store

Loopy Pro is your all-in-one musical toolkit. Try it for free today.

Video: What is waveshaping? How does it affect guitar amp tone?

2»

Comments

  • espiegel123espiegel123

    @Blue_Mangoo said:

    @McD said:

    @ohwell said:
    Also can't wait to play around with your saturation plugin!

    There's probably a good discussion that can come from having and reading a good
    Frequency Spectrum display app. What's the best tool for analysis of existing sound
    streams visually. I think there's one that came with ApeMatrix that is an AUv3 loadable app.

    When I really need high resolution spectrum analysis, I use Izotope RX, which is not available on iOS, unfortunately.

    For most other stuff the spectrum in our parametric EQ is enough to adjust the EQ. If you need an analyser that allows you to adjust the frequency resolution, get the one from 4pockets.com

    It would be nice if the frequency analyzer were "freezeable" and allowed the frozen analysis to be visible along with the live analysis.

  • wimwim

    I’m getting some pleasing results by chaining additional AUM saturators, each with less clipping than I would with just one.

  • Blue_MangooBlue_Mangoo

    @wim said:
    So, when there are multiple tubes in a guitar amp, are they distorting in serial, or parallel?

    Nearly every valve guitar amplifier is a class AB amplifier. In a class AB amplifier the input signal is rectified to split the negative voltages into one signal path and the positive side voltages to another. Typically if you see two 6L6 tubes in the power amp, one is for the positive signal and the other is for negative. So they are in parallel.

    But an all tube amp has both preamp and power amp tubes. The smaller tubes, usually 12AX7 type, are preamp tubes, and usually these are also working in parallel on a rectified pair of negative and positive side signals.

    Therefore the complete answer is that if you have four tubes in the amp, 2 preamp and 2 power amp then it’s is both series and parallel. The negative side goes in series through one 12AX7 preamp tube and one 6L6 power amp tube and the positive side also the same.

    Technically if you are modeling a tube preamp and power amp in software then you should have two gainstages in series. However if your saturator does waveshaping only without modeling sag then there isn’t much point to going through a second stage because you could get exactly the same result with a single stage.

    Unless you have a variac to starve the power supply of the power amp stage like Van Halen did, the power amp is usually designed to give clean amplification of whatever the preamp gives it. This is not the case in some high gain amplifiers but I can’t comment on that with confidence because I haven’t researched it.

    Anyway, in many cases the power amp doesn’t contribute much to the distortion tone and you can model the amp well using just a single distortion stage. Especially if your software model of a tube is just a saturation plugin, I would usually recommend trying to model the whole amp as a single gainstage. Of course, if you find it sounds great with two of them then go for it.

    In our saturator I tried using up to four gain stages in series and I ended up going back to just one. The problem with stacking them is that the first stage compresses the output down low, which means the stages that come after it haven’t got high gain at the input, and therefore they don’t distort much.

    I tried to fix that problem by putting a gain boost between stages so the input of all stages is overloading the tubes, but that also doesn’t work very nicely. The problem with that approach is that the first stage compresses everything down to the same volume range and then the distortion from the later stages can’t respond differently to soft and loud dynamics in your playing.

    In summary, if you want the sound to clean up when you play softly then all of your distortion tone needs to come from just one gain stage. If you add a second stage in series behind that, it should be only a subtle saturation to add warmth, not a second distortion. Of course, if you want 120db of face melting gain at all playing volumes then by all means, ignore everything I said and use as many distortion gain stages as you want.

  • rs2000rs2000

    @Blue_Mangoo Have you tried multiple different parallel distortion lines that are mixed in the end, including adjustable tiny phase shifts or delays in each stage and fine-tune them by ear?

  • Blue_MangooBlue_Mangoo

    @rs2000 said:
    @Blue_Mangoo Have you tried multiple different parallel distortion lines that are mixed in the end, including adjustable tiny phase shifts or delays in each stage and fine-tune them by ear?

    I know that guitarists often use amps in parallel but as an engineer I am hesitant to use EQ in parallel because one way to explain how EQ works is to say that it splits the signal into several parallel streams, delaying each one by a small number of samples, adjusting the gains on each channel, and finally summing back together. If you try this you will see the problem:

    1. In aum, make two tracks with the same input.
    2. On track 1, boost 15db at 100hz, then cut -15 dB at 100 hz, so that the boost and cut cancel out and you are back to flat EQ
    3. In track 2, boost and cut at 2000 Hz, also cancelling the two out so you get back to flat EQ.

    Expected result:
    When you mix tracks 1 and 2 together, the output is flat EQ.

    Actual result:
    Because the filters cause heavy phase shifting, when you mix tracks one and two the output is not flat at all, in fact it’s wild and crazy and totally unpredictable (unless you actually do the math to calculate the phase shifts.)

    Of course, the ultimate rule is “do whatever sounds good” but I have never ran different EQ on two separate signals in parallel inside any of our apps because I would have no way of guessing what the output would be. The only time I have done this is when there is a crossover splitting the signal into bass,mid, treble frequency bands, because there is less concern about phase in that situation since the bands don’t share the same frequencies.

    If you do want to amp in parallel, you can build your own crossover by using a lowpass filter on one cb annuel and a highpass on the other. Ideally these two filters should be set at EXACTLY the same cutoff frequency and they should have a Q setting of 0.5 to ensure that the phases line up and there is no boost or cut at the transition band when you recombine them.

  • wimwim
    edited November 2019

    Thanks for the great explanation @Blue_Mangoo! I really appreciate it.

    I’m finding that putting two saturators in series, with an eq and gain between helps me get closer to tone I like than just one. There’s a lot of variety to be found by experimenting with different levels of saturation on the two.

    That said ... after a couple hours messing around, and thinking I liked what I found, I opened up Tonestack ... and decided maybe I’m not as good at crafting my own sound as I thought I was. :D

  • Blue_MangooBlue_Mangoo

    @wim said:
    Thanks for the great explanation @Blue_Mangoo! I really appreciate it.

    I’m finding that putting two saturators in series, with an eq and gain between helps me get closer to tone I like than just one. There’s a lot of variety to be found by experimenting with different levels of saturation on the two.

    That said ... after a couple hours messing around, and thinking I liked what I found, I opened up Tonestack ... and decided maybe I’m not as good at crafting my own sound as I thought I was. :D

    One thing that helps me is to be very cautious about the level of complexity in your setup. In the digital domain we can build very complex filters for free but in analog amps and pedlas, every filter costs money and has to be built by hand when the engineer builds the prototype design.

    The more filters you have, the more ways you can get it wrong. If you look at the electrosmash website where they plot frequency responses of classic amps and pedals, you’ll notice that the lowpass and highpass filters in classic analog gear are always first order (6db/octave). I realised that in my videos I am often using second order highpass filters because that’s the default in blue Mangoo EQ. But a second order filter is much more expensive to build and harder to design in analog.

    If you use first order filters (get them in our EQ by pulling the q slider on low cut and high cut all the way down to the bottom) you’ll see that they don’t have adjustable Q at all. Only the frequency can be adjusted. At first I thought that was a crippling limitation but when I realised that ALL the HP and LP filters in the gear I am modeling are first order, then I saw that the extra adjustment options of the more complex filter were only working to get me further away from the sound I want.

    I think of this in terms of degrees of freedom. If you have just one slider to move then you have one degree of freedom. You slide it till it sounds good and you are done. But when you add the second slider or knob into the mix, you are now searching for the perfect combination of two sliders. It’s a lot harder and takes a lot longer to get it right.

    Imagine you have an app with a switchable mid boost that has ten settings from 1 to 10. Then you have 10 settings to try. Now add another bass boost with 10 settings. Your total number of combinations to try goes from 10 to 10x10=100 just by adding one more switch. And if it’s three switches then you have 1000 different combinations.

    This is one reason why I recommend using just one gain stage. EQ>saturation>EQ is already enormously complicated. Unless you are 100% sure that you have tried everything possible with the si ng le stage setup, you’ll be glad you didn’t complicate it further.

  • ohwellohwell

    @Blue_Mangoo said:

    @ohwell said:
    @McD Oh, I didn't mean I'll wait to start using simple saturators with guitar, instead of black box, all-in-one amp modellers. ;)

    Agreed, it's extremely instructive to spend time learning how to achieve target guitar tones from simple, flexible building blocks!

    One thing: with typical saturation algos (like the one in AUM), adding EQ's is usually not enough IMO to get in the ballpark of an amp-like feel. As @Blue_Mangoo points out, another crucial part of what guitar amps do is reshape the dynamics (e.g. power amp sag) of the signal. With the typical simple tools, it's usually crucial to include some kind of compression. This is actually one of the many things that makes me especially curious about @Blue_Mangoo 's saturator: I'm looking forward to seeing how it tackles the dynamics side of tube behavior, along with the waveshaping behavior the above video focuses on.

    If you do use a compressor, I recommend putting it after the waveshaper with very short release and attack time (less than one ms). Unfortunately most compressors don’t have a release time setting that is short enough to imitate power supply sag. Our own compressor release time doesnt go lower than 50 milliseconds, so that won’t get you close. Compressors reduce distortion by releasing more slowly than they attack so if you do see a compressor with a release time that low, it will be extremely noisy and distorted.

    Yup!! Typical compressors don’t behave anything like sag. I had a few different things in mind, all of which make your approach extremely compelling to me.

    1. On their own, a plain wave shaping stage that responds to picking dynamics + pre/post eq usually ends up with a much broader dynamic range at the output than guitar players want. Even though it won’t sound like a tube amp, following that up with a traditional compressor often helps get the final signal within the right loudness range. (A helpful trick for light/medium drive pedals too: eg following up something like a Fulltone OCD with a touch of compression.)

    2. There are unusual compression/dynamics algos that behave more like tube amps. Airwindows PowerSag is literally designed for that. In practice, something like AW Recurve is likely just as, if not more valuable. But you help me see how the lesson there is the same: to get the dynamic response to be more like that of a tube amp than a typical compressor allows, some out of the box approach is required, and there are not enough good tools exploring that territory. Which makes what you’re working on super exciting!!!

    3. As soon as we go down the path of adding anything beyond pre/post eq and a single distortion stage, things very easily get out of hand. From the perspective of giving users flexible but manageable tools to achieve the results they want, you are helping me realize just how much of a boon it would be to have a tool that can cover as much of the saturation/dynamics territory guitar players need as possible as a single stage between two EQs!! (Ergo: hurray blue mango saturator!!)

    Your insights are truly fascinating!!

  • Blue_MangooBlue_Mangoo
    edited November 2019

    @ohwell said:

    @Blue_Mangoo said:

    @ohwell said:
    @McD Oh, I didn't mean I'll wait to start using simple saturators with guitar, instead of black box, all-in-one amp modellers. ;)

    Agreed, it's extremely instructive to spend time learning how to achieve target guitar tones from simple, flexible building blocks!

    One thing: with typical saturation algos (like the one in AUM), adding EQ's is usually not enough IMO to get in the ballpark of an amp-like feel. As @Blue_Mangoo points out, another crucial part of what guitar amps do is reshape the dynamics (e.g. power amp sag) of the signal. With the typical simple tools, it's usually crucial to include some kind of compression. This is actually one of the many things that makes me especially curious about @Blue_Mangoo 's saturator: I'm looking forward to seeing how it tackles the dynamics side of tube behavior, along with the waveshaping behavior the above video focuses on.

    If you do use a compressor, I recommend putting it after the waveshaper with very short release and attack time (less than one ms). Unfortunately most compressors don’t have a release time setting that is short enough to imitate power supply sag. Our own compressor release time doesnt go lower than 50 milliseconds, so that won’t get you close. Compressors reduce distortion by releasing more slowly than they attack so if you do see a compressor with a release time that low, it will be extremely noisy and distorted.

    Yup!! Typical compressors don’t behave anything like sag. I had a few different things in mind, all of which make your approach extremely compelling to me.

    1. On their own, a plain wave shaping stage that responds to picking dynamics + pre/post eq usually ends up with a much broader dynamic range at the output than guitar players want. Even though it won’t sound like a tube amp, following that up with a traditional compressor often helps get the final signal within the right loudness range. (A helpful trick for light/medium drive pedals too: eg following up something like a Fulltone OCD with a touch of compression.)

    2. There are unusual compression/dynamics algos that behave more like tube amps. Airwindows PowerSag is literally designed for that. In practice, something like AW Recurve is likely just as, if not more valuable. But you help me see how the lesson there is the same: to get the dynamic response to be more like that of a tube amp than a typical compressor allows, some out of the box approach is required, and there are not enough good tools exploring that territory. Which makes what you’re working on super exciting!!!

    3. As soon as we go down the path of adding anything beyond pre/post eq and a single distortion stage, things very easily get out of hand. From the perspective of giving users flexible but manageable tools to achieve the results they want, you are helping me realize just how much of a boon it would be to have a tool that can cover as much of the saturation/dynamics territory guitar players need as possible as a single stage between two EQs!! (Ergo: hurray blue mango saturator!!)

    Your insights are truly fascinating!!

    Power supply sag is really noisy; usually we don’t want that in a compressor, unless it's the Magic Death Eye compressor:

    https://apps.apple.com/us/app/magicdeatheye-ddmf-compressor/id1466360561

    I said in my compressor shootout video that this compressor doesn't seem to be oversampled, therefore it aliases heavily for input frequencies above ~3kHz, but if you are lowpass filtering the input, that might not bother you at all. I haven't tried it myself but it's possible that this compressor would sag the way you want an amp model to do.

  • McDMcD

    @flo26 did a YouTube video that featured the MagicDeathEye and it made me put down my $25 ASAP. I add it to my guitar rig's and it always sounds better (according to may tastes for what a good amp would add). It seems to add 2nd order harmonics because it adds a sweet distortion that's not edge-y but warm.

    The video is here... subscribe to his channel if you like Guitar and IOS. He only has 43 subscribers and the tips and inspiration to practice are worth the time.

  • rs2000rs2000

    @Blue_Mangoo said:

    @rs2000 said:
    @Blue_Mangoo Have you tried multiple different parallel distortion lines that are mixed in the end, including adjustable tiny phase shifts or delays in each stage and fine-tune them by ear?

    I know that guitarists often use amps in parallel but as an engineer I am hesitant to use EQ in parallel because one way to explain how EQ works is to say that it splits the signal into several parallel streams, delaying each one by a small number of samples, adjusting the gains on each channel, and finally summing back together. If you try this you will see the problem:

    1. In aum, make two tracks with the same input.
    2. On track 1, boost 15db at 100hz, then cut -15 dB at 100 hz, so that the boost and cut cancel out and you are back to flat EQ
    3. In track 2, boost and cut at 2000 Hz, also cancelling the two out so you get back to flat EQ.

    Expected result:
    When you mix tracks 1 and 2 together, the output is flat EQ.

    Actual result:
    Because the filters cause heavy phase shifting, when you mix tracks one and two the output is not flat at all, in fact it’s wild and crazy and totally unpredictable (unless you actually do the math to calculate the phase shifts.)

    Of course, the ultimate rule is “do whatever sounds good” but I have never ran different EQ on two separate signals in parallel inside any of our apps because I would have no way of guessing what the output would be. The only time I have done this is when there is a crossover splitting the signal into bass,mid, treble frequency bands, because there is less concern about phase in that situation since the bands don’t share the same frequencies.

    If you do want to amp in parallel, you can build your own crossover by using a lowpass filter on one cb annuel and a highpass on the other. Ideally these two filters should be set at EXACTLY the same cutoff frequency and they should have a Q setting of 0.5 to ensure that the phases line up and there is no boost or cut at the transition band when you recombine them.

    Actually I was talking about distortion, not EQ. EQ might be part of one distortion channel but the point is rather to mix different distortion paths in order to get a sound that is more pleasant to the ear. It's more like the opposite of the multiband distortion you described: Rather let the channels overlap. But even a mix of both might sound good, who knows.

    Also, the way software EQs are usually implemented today is rather by doing convolution with a filter kernel that represents the filter impulse response, there's no "splitting" involved.
    The filter kernel will not only affect filter latency but also determine the frequency-dependent group delay (or phase response if you will), so you cannot say that the side effects you mentioned are a given ;)

  • Blue_MangooBlue_Mangoo

    @rs2000 said:

    @Blue_Mangoo said:

    @rs2000 said:
    @Blue_Mangoo Have you tried multiple different parallel distortion lines that are mixed in the end, including adjustable tiny phase shifts or delays in each stage and fine-tune them by ear?

    I know that guitarists often use amps in parallel but as an engineer I am hesitant to use EQ in parallel because one way to explain how EQ works is to say that it splits the signal into several parallel streams, delaying each one by a small number of samples, adjusting the gains on each channel, and finally summing back together. If you try this you will see the problem:

    1. In aum, make two tracks with the same input.
    2. On track 1, boost 15db at 100hz, then cut -15 dB at 100 hz, so that the boost and cut cancel out and you are back to flat EQ
    3. In track 2, boost and cut at 2000 Hz, also cancelling the two out so you get back to flat EQ.

    Expected result:
    When you mix tracks 1 and 2 together, the output is flat EQ.

    Actual result:
    Because the filters cause heavy phase shifting, when you mix tracks one and two the output is not flat at all, in fact it’s wild and crazy and totally unpredictable (unless you actually do the math to calculate the phase shifts.)

    Of course, the ultimate rule is “do whatever sounds good” but I have never ran different EQ on two separate signals in parallel inside any of our apps because I would have no way of guessing what the output would be. The only time I have done this is when there is a crossover splitting the signal into bass,mid, treble frequency bands, because there is less concern about phase in that situation since the bands don’t share the same frequencies.

    If you do want to amp in parallel, you can build your own crossover by using a lowpass filter on one cb annuel and a highpass on the other. Ideally these two filters should be set at EXACTLY the same cutoff frequency and they should have a Q setting of 0.5 to ensure that the phases line up and there is no boost or cut at the transition band when you recombine them.

    Actually I was talking about distortion, not EQ. EQ might be part of one distortion channel but the point is rather to mix different distortion paths in order to get a sound that is more pleasant to the ear. It's more like the opposite of the multiband distortion you described: Rather let the channels overlap. But even a mix of both might sound good, who knows.

    Also, the way software EQs are usually implemented today is rather by doing convolution with a filter kernel that represents the filter impulse response, there's no "splitting" involved.
    The filter kernel will not only affect filter latency but also determine the frequency-dependent group delay (or phase response if you will), so you cannot say that the side effects you mentioned are a given ;)

    If you can find a distortion plugin that doesn’t have any filters in it then you should be able to run in parallel with no phasing issues at all. Aside from the saturator in aum I’m not aware of others that are filter-free.

    I was being very creative with the mathematics when I said that filters are splitting the signal, delaying, and merging back together at different gain settings. Convolution does exactly that but you have to turn the formula on it’s side to see it. If you are curious, I can do a video about it. ;)

  • [Deleted User][Deleted User]
    edited November 2019
    The user and all related content has been deleted.
  • rs2000rs2000

    @Blue_Mangoo said:
    If you can find a distortion plugin that doesn’t have any filters in it then you should be able to run in parallel with no phasing issues at all. Aside from the saturator in aum I’m not aware of others that are filter-free.

    You know, I took the more adventurous trip to discover different ways to do distortion for myself :smiley:

    I was being very creative with the mathematics when I said that filters are splitting the signal, delaying, and merging back together at different gain settings. Convolution does exactly that but you have to turn the formula on it’s side to see it. If you are curious, I can do a video about it. ;)

    Thanks for the offer, I know how convolution works and what the different kernels do but maybe someone else here would be interested?
    Not sure though. This is not a DSP forum...

  • rs2000rs2000
    edited November 2019

    @Blue_Mangoo Here's an example of what I mean:

    The dual-channel version sounds more consistent and "creamy".

  • Blue_MangooBlue_Mangoo

    @rs2000 said:
    @Blue_Mangoo Here's an example of what I mean:

    The dual-channel version sounds more consistent and "creamy".

    I agree. It does sound nice with two in parallel.

    I didn’t mean to say it would sound bad due to the phase issues; only that it would be difficult to predict. There is a time for calculating and planning a time for just trying stuff.

Sign In or Register to comment.