Audiobus: Use your music apps together.
What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.
Download on the App StoreAudiobus is the app that makes the rest of your setup better.
Comments
I've tried most of them and I would not call PUSH a 'hard clipper' as it's more of a 'compressor/limiter' with some control over what happens when the threshold is crossed.
K-Devices Shaper 2 is pretty nice for clipping and other 'trashing'
I really like this one but the UI takes some time to get used to and preset management is still NOT fixed!
https://k-devices.com/products/shaper2/
Most if the time I turn to BLEASS Saturator and I usually use my ears and do not look at spectrum graphs...
Cheers!
Toneboosters is the best if you want to avoid aliasing artifacts.
True, but at the cost of fluctuating latency as we never know how much the signal will get delayed
In AUM it works but in other hosts without PDC(Plug-In Delay Compensation) it can become a pita...
Thanks! I guess I was wondering about overshoot protection, as it had been recommended in the other thread. So when gusgranite mentioned Baricade being the only clean clipper, I thought perhaps they hadn’t tried PUSH or that it had failed the task. That’s of course assuming that I'm not getting mixed up with hard-clipping and what I’m after (The more likely scenario
)
TB Barricade is a 'must have' in my book...
...It's just nice to slap on the limiter and crank it up to 11 and know that it'll not get past the set point no matter what.
The compressor is also pretty nice in combination with the clippers, saturators etc.
So, if I'd have to select one 'clipper' TB Barricade would be it.
(That is when the other 'apps' don't have anything built-in).
Cheers!
Lots of nice soft clippers on iOS. Not many hard clipper options from my testing. Barricade will work but a dedicated K-Clip style clipper would be great on iOS.
Thanks, again.
Was kinda hoping not to have to go for Barricade in the short term, as I think any mastering I might do will either be done in Cubasis (Seems well provided for) or in Logic Pro. Was already planning on getting TB's EQ soon. And while these apps are not expensive in an absolute sense, those relatively-cheap amounts start to add up
PUSH looks like a really flexible compressor, and cheaper. But if it’s not going to work out for me, then…
That does indeed seem to be the case. I went through all the settings in the FabFilter apps and found nothing comparable to Toneboosters’ clarity and ease of use.
An iOS version of Free Clip would be so nice...
Would you mind explaining the relationship between clipping and limiting on Barricade?.
I see there’s clipping algorithms but I don’t really know how it works and/or if it’s independent of limiting.
When you select 'Hard clip' as the model (lower left) you set the 'level' at which the input is 'clipped'.
Then you can use the limiter section (Gain) to increase the overall volume.
If the volume is still too low you can use the compressor and its Makeup to boost the signal before it reaches the limiter.
(This might be needed if you set the Hard clip level very low <-35db as it 'cuts down' the input down to <-35db and the gain at the limiter section only provides +20db)
You could in theory use the 'gain' to boost the signal to clip but since Barricade is designed to be a 'clean limiter' it doesn't clip the signal even when you increase the gain a LOT this they added the 'Hard clip' mode that is applied to the input signal...
If you've got the app play around with it, experimentation increases experience
Good luck!
Thanks!. I hadn’t actually tried the clipping mode. I thought of hard clipping as a kind of a limiter in its own way. Seeing the waveform after, it sort of is but it does let some transients through.
I'm guessing that there is a bit of analog thought/concepts coming through here. A hard clip should mean that the output is cutoff at the set level. But ...
An MXR Distortion+ is pretty much the standard for a hard clipping distortion pedal. The Tube Screamer is pretty much the same for a "soft clipping" distortion. The Distortion+ has the clipping diodes going from the output of the op-amp gain stage directly to ground. So, (theoretically at least) it shaves off everything above the voltage drop of the diodes to ground. For the Tube Screamer, the diodes are in the feedback path of the op-amp amplifier so they change the response of the amplifier. It isn't a direct shave off because it also involves the response time of the op-amp and it feeds the clipped signal back through the amplifier further changing the response. The output of a Tube Screamer has a kinda rounded chopped off waveform. The Distortion+ is pretty much chopped flat. But, the Distortion+ uses a germanium diode. These have fairly slow response and recovery times. So, a bit of transient gets through and the shape isn't a direct cutoff. The transients still won't go too far above the voltage level of the diode and it will be altered from the input signal.
For me, a digital hard clip shouldn't let any transients or signal through that's above the clipping voltage. But, if someone is doing some modeling of what would come out of a Distortion+, it would be valid to call it a hard clip and still see some response above the clipping level.
Nice explanation. Thanks! 🙏
I just tried the har clipping again with 2 different sources and I’m certainly not fully getting how the clipping works.
Here’s 2 screenshots of the same Barricade settings on 2 files. First is a full mix and the second a bass line. You can see on the waveform where the threshold is lowered to engage the hard clipping.
On the full mix you can see how some transients are allowed through by the clipper. Whereas the bass file seems completely shaved off. Why?. Does it have to with some analog emulation so that it’s not responding equally to all frequencies?.
What happens if you push the clip level down even more? (like -20 or -25?).
The RMS level goes drastically down when applying a high-pass filter to the source.
On the bass track the input level is quite a bit higher compared to the main mix?
What I would like to see is a 'clipper' with 'auto gain' that keeps the output at a set level.
(ie. if you 'clip' at say -15 that -15 would be raised to the 'target level').
Purely 'clipping by numbers' will create some nastiness and what I think TB is doing here is to filter out unwanted aliasing distortion which can have an effect on the overall level and explain the difference between the clip level of the bass and full mix.
For experimental music a purely numerical clipper that just 'shaves' the numbers to cap out at a fixed level could be nice
(Depending on how one looks at it it could be either aliasing heaven or hell).
Next step is to look at wave-shapers with a transfer-curves like the Bleass Saturator...
Here’s pushing it to -25db. This time instead of lowering the threshold I changed the algorithm from “none” to hard clip to ensure that the levels were exactly the same. Same results, bass is shaved while full mix let’s some transients through.
And here’s a guitar and a different bass file , just to test some more…
These last 2 examples might be a good comparison. The waveforms don’t seem very different regarding transients, yet the bass, lower frequency, if a lot more shaved than the guitar.
Don’t know what it means but I’m finding interesting 😀
Hehe, another addition to the experiments would be to engage the compressor, max out the ratio and really push it with minimum attack and release times, with Makeup enabled you should be hitting the limiter quite hard to really shave everything off if not increase the gain at the end of the chain (you can safely max it since the limiter will makes sure it doesn't get past the set target level).
Good luck with the experiments
Looking at your screenshots, I had to go take a look at the manual for Barricade as I don't use it. The thing that stands out to me and probably explains what you are seeing is that the hard clip is part of the "input modeler" and is supposed to be modeled on analog circuits. Then the compression and limiting is applied. The differences between the full mix and the bass image make sense in that context. Even if it were a pure digital hard clip, content that is below the threshold could be lifted above in a frequency dependent way in the following stages.
I think of the usefulness of a hard clip as final protection on the output. They are using it as a distortion source on the input. I can't imagine ever running a recorded mix through a Distortion+ before hitting the compressor/limiter in an analog setting, but maybe someone does want to do this. I think maybe I need to go write a blown speaker modeler for when I get nostalgia for my teenage days of not having enough money to replace the woofer in my stereo.
A pure digital hard clip does sound really nasty/awful. That CircuitBreaker project I put up on GitHub is one and it sounds horrible. But, that's the point to me. It's supposed to be a big audible warning that something is wrong and protect your hearing at the same time.
Completeley agree. about 2 years before "covid" I had to shut down my studio and was offered a room at a local hip hop studio. I brought my essentials along with my console. Everyone was always amazed at how loud I was able to quickly get their "two tracks loud as hell" bussines went from booked all day long to litterally zero the day of covid. Honestly was happy I had to move on because I couldnt stand the mindset of non preperation and lack of care for the music portion. Obvioously this is only a portion of that community.
Anyways my point is. Well my trick was simple and went against what school and most would say. I had a fully loaded Burl Mothership running madi into my rig. Bringing the mix directly from the monitor in real time I would record back in to an empty stereo track hitting the input of the burl converter extrememly hard. Like plus 14 sometimes and to this day I listen back to how clean and punchy those masters came out.
The closest thing ive come across since then is Barricade. The first time I used the new version I remember smilling and thinking these guys have done it. I dont know how its possible.
Cheers
I hadn't heard of the Burl stuff before. I went and had a look. Looks fun and pricey! If I'm interpreting your setup and their description correctly, it sounds like you are getting transformer saturation and discrete op-amp saturation by overdriving the inputs. Kinda like in the old days of overdriving both the mixer and the tape machine inputs. Nice!
If you want to kind of reverse your experiments, AU3FX:Push has the ability to apply the hard clip at the end of the signal chain after the compression. Could be interesting to compare the outputs.
I’ve just realized that AUM has a built in hard clipper in the dynamics section that I never tested. Will try it out tonight.
Let’s remember we’re talking about two different requirements from clipping here.
As far as I can tell, the hard clip node in AUM is a gain control with a 0dB digital hard clip after. I could be wrong, but that's what it feels like to me.
I don't understand the term transparent hard clipping. Does that mean clean gain up until you back it off because you hear the harsh clipping? Because after you hit the clip point --- analog, analog modeled, or digital --- hard clipping is anything but transparent.
As far as I've understood it 'transparent' when it comes to clipping is the reduction of the generated overtones that can cause aliasing especially when the frequencies get higher.
There is an ever growing playlist from Baphometrix that goes DEEP into this (he calls it Clip to Zero). If you make club/festival music that needs a loud master then it is a really interesting area.
I tried to watch the video. I can't. The voiceover itself is so distorted I can't listen to it. That is very definitely not transparent.
This is a response I have to much of current music/audio production. The distortion from both the mixing techniques and inter-sample peaking drives me nuts. Add in all the phase distortion in the bass from all the multi-band processing and inserted white noise for "energy," and it really does make it hard for me to listen to.
OK, ignoring my grumpy old man take on things, I still don't understand what transparent means in this context. Transparent to me means it doesn't add color. I can maybe understand extending this to mean that you can add harmonics that sound good. Once you get to the point of adding intermodulation distortion, I can't call that transparent in any way.
Hmm, I can kinda get this. But, there are going to be all sorts of added overtones and harmonics below the aliasing frequency that are going to make for a very non-transparent output from any hard clip or aggressive saturation. E.g. my mentioning of intermodulation distortion in my comment above.
True...
As far as I know TB Barricade does some adaptive oversampling and filters out the stuff that would cause potentially aliasing.
There was a havoc a here at the forum a while back when doing a spectrum analysis on gain/distortion plug-ins and some of them were 'cleaner' than others.
I prefer to use my ears rather than look and spectrum graphs
If you don’t like his voice then he’s written up a summary here https://docs.google.com/document/u/0/d/1Ogxa5-X_QdbtfLLQ_2mDEgPgHxNRLebQ7pps3rXewPM/
Thank you for the link.
It wasn't his voice that was my issue. It was the distortion in the audio on the video. According to the doc, he uses his mixing technique on his videos as an example. It sounds really bad to me. It's not just him either. As he points out in the article, lots of people do something similar. I don't like the sound of any of those either.
I'm actually a bit more confused now too. First thing is that mixing near 0dB in a 32-bit float setting won't introduce any clipping. So, the part about discovering this by just pushing the mix up near 0dB doesn't make any sense to me. But, pushing the mix up to 0dB does make it sound "better" to everyones ears just because that's the way our brain works. Louder sounds better and it's something you have to work to get past because everything louder altogether doesn't sound better and is fatiguing to listen to as well.
So, just pushing to 0dB not clipping the individual stages means that he's putting clipping in at every stage (in what he's calling gainstaging). So, I'm still back to asking what is transparent clipping supposed to be? I went and looked up the distortion plugin he recommends KClip Pro 3 and this is what I find
I don't know what smooth and crisp are but none of that sounds transparent to me. It's also multi-band, so that introduces more coloring. And, it does seem like I should go write the broken woofer plugin I was thinking about. The description of the KClip Pro plugin does use the term transparent though.
I mentioned the MXR Distortion+ as an example of a Ge diode based hard clipper above. If you take the output of a Distortion+ and run it straight to a mixer without the EQ'ing from a tonestack and a speaker cab, it sounds awful. I'm pretty sure that there is going to be EQ applied in all of those clipping modes. So, I'm going to go with the idea that transparent isn't really an important adjective here.
Comparing what KClip 3 does to Barricade, it looks to me like Barricade (as well as a bunch of others) would work to do this mixing technique.