Clipper Shootout 2020

Do all clippers sound the same?

Do clippers change the tonal balance of a mix?

Let's find out!

I tested the AU (Audio Units) version of each plug-in below in Logic Pro X:

• Boz Digital Labs Big Clipper 1.1.0
• IK Multimedia Classic T-RackS Clipper 5.4.0
• Joey Sturgis Tones JST Clip 1.0.4
• Kazrog KClip 3.2.3
• LVC-Audio Clipped-MAX 2.1.7
• Newfangled Audio Saturate 1.7.0
• SIR Audio Tools StandardCLIP 1.5.034
• Stillwell Audio Event Horizon 3.0.2
• SubMission Audio Flatline 1.0.0
• Tokyo Dawn Labs TDR Limiter 6 GE 1.2.2
• Venn Audio V-Clip 1.0.3
• Voxengo OVC-128 1.7

To mitigate aliasing all testing was conducted at 96 kHz (sample rate).

To keep the comparison fair, plug-ins with oversampling options were limited to 4x oversampling.

The sample consists of a drum track from @ kaide mixed with guitars from @ CaptainHook .

The sample was processed through each plug-in using the settings below.

Global settings: Gain/Drive +6.0 dB, Ceiling 0.0 dB, Hard clipping, 100% wet:
• Boz Digital Labs Big Clipper: Crossover Mode - Clipper | Limiter​ / 20 kHz crossover, Shape 100% (hard)
• IK Multimedia Classic T-RackS Clipper: Slope 0.0 (hard)
• Joey Sturgis Tones JST Clip: Trim 0.0 dB
• Kazrog KClip: Single band, Crisp mode, 4x Oversampling
• LVC-Audio Clipped-MAX: Master module, Hard algorithm, Harmonic 100%, Linear-HQ 4x Oversampling
• Newfangled Audio Saturate: Clipper shape 100%, Detail preservation 100%
• SIR Audio Tools StandardCLIP: Mode: Hard Clip, 4x Oversampling
• Stillwell Audio Event Horizon: Clip mode, Soft Clip 0.0 dB
• SubMission Audio Flatline: Shape 100%, 4x Oversampling
• Tokyo Dawn Labs TDR Limiter 6 GE: Quality - Precise, Clipper only, Mode - Brick wall, Threshold 0.0 dB, Knee 6.0 dB
• Venn Audio V-Clip: Hard Clip, 4x Oversampling - Linear Phase
• Voxengo OVC-128: Hardness 100%

After rendering the processed clips were volume matched to the original.

The Stillwell Audio Event Horizon clip contains a volume anomaly at the beginning - this was created by the plug-in.

The posted clips are 24-bit / 96 kHz WAV format files.

Nice work! I will listen tomorrow but I have a few initial comments:

1) Why cripple oversampling on certain plugins when oversampling is an integral component of what makes a clipper sound good or not?

For example OVC-128 is named after its prized 128x oversampling. Turning that off (if you could) would defeat the reason why the clipper was made in the first place right?

So I suggest you allow each plugin to rock their max oversampling settings. Best foot forward approach.

2) You mention the gain/drive at +6dB, but don't tell us the max clip level. Or did you?

We dont know if the original WAV had 2 or 5dB of headroom, resulting in either 4 or 1dB of clipping respectively. Knowing the approximate amount of clipping is important (to me) as various clippers perform differently according to how deep the clipping is.

Thanks again for doing this! Looking forward to listening soon.

Robbie,

Thank you for the feedback!

I have re-run the tests for the clippers with oversampling options, selecting the maximum OS setting:
• Kazrog KClip: Single band, Crisp mode, 32x Oversampling
• LVC-Audio Clipped-MAX: Master module, Hard algorithm, Harmonic 100%, Linear-HQ 128x Oversampling
• SIR Audio Tools StandardCLIP: Mode: Hard Clip, 256x Oversampling
• Tokyo Dawn Labs TDR Limiter 6 GE: Quality - Insane, Clipper only, Mode - Brick wall, Threshold 0.0 dB, Knee 6.0 dB, HQ Rendering
• Venn Audio V-Clip: Hard Clip, 256x Oversampling - Linear Phase

The following were already at their maximum OS setting:
• SubMission Audio Flatline: Shape 100%, 4x Oversampling
• Voxengo OVC-128: Hardness 100%, 128x Oversampling

Attached please find the waveform statistics for the original sample.

Headroom is a little under 2 dB.

The maximum indicated clip level (GR) is -4.2 dB.

-Simon

Awesome Simon. Thanks for appeasing me.

Still haven't listened yet, but 4dB of clipping should be quite revealing across all these plugs.

Cheers
RR

What would we do without dedicated audio geeks like you who take the time to do these sorts of invaluable comparisons?

Thanks so much.

I have five of the ones you tested and they all held their own quite well. Good to know. It keeps the GAS under control, especially during big sales seasons!

Amazing, thanks for this (and for the updated versions per Robb's comments). Off out to lunch but will have a listen when back in the studio.

May I ask a small favour too? Using the fully oversampled versions, would you be able to upload a randomised order set that are labelled A, B, C etc? This would make a great blind test.

I'd do it myself but I'd know the results

@ bmanic - would you be up for creating a Stereo Tool clipped version? :¬)

will take out time later today to check it all ...

thanky for the work! IT looks already very useful!

If I remember right, oversampling doesn't always make for a better result when it comes to clipping. Or maybe 2x has the best result in the context but 128x would be audible. Etc.

I personally try to not use oversampling when clipping at 96khz, but I'm also just shaving some infrequent peaks, not "treating" the audio. But, still, sometimes I use 2x or even 4x in StandardClip.

I agree that real time oversampling in most plugins has been detrimental in my testing, and working at 96khz as well means I try to avoid it, but I must say that whatever Voxegno is doing in OVC is impressive. I know he recently updated the oversampling algo in OVC and I swear it sounds smoother. : /

Id love to hear Stereo Tool clipping as well. And the only other one im missing from this comprehensive list is the Massey clipper called Golden Master.

Listened to all the samples but I cut them down to 5-second clips starting 10 seconds in. I ordered the clips by transient impact.

Very little difference overall, I think. The original came in at #1 (surprise), followed by Clipped-MAX, KClip, Flatline, JST Clip, Big Clipper, Saturate, Limiter No6, OVC-128, StandardClip, Event Horizon, T-Racks Clipper, V-Clip.

The last two or three I felt killed the transient impact a little bit. There is really no order before that, I think.

Thanks @ Simon Morrison !

Clippers should be relegated to channel duties. On the bus they generally make things sound worse.

Interesting: I went through the clippers and there's basically two groups: One that more or less flat lines the clipped signal (V-Clip on the screenshot, top right) and one that keeps some information. Kazrog KClip seems to keep most, followed by Limiter No6 and Newfangled Audio Saturate.

Screenshot shows what I mean.

Sorry, missed this somehow. Yes sure, I'll make a few Stereo Tool versions. The problem will be how to settle on the various settings! There are so many variations to choose from and I don't think Stereo Tool clipper satisfies the "hard clipping" category as it's so highly variable in it's settings. I think there might be tons of processing it does under the hood.. some of it being limiting (for instance the separate bass limiting part of the advanced clipper that does some clever dynamic EQing or something in the background).

But yes, I'll try to post a few renders with varying options.



Cheers to the OP for creating this!

EDIT: Aww crap.. the audio posted is 96kHz. I don't think Stereo Tool does 96kHz correctly. Will investigate!

Here is an example of Stereo Tool. It sounds ok at 96kHz so I guess it's working.

.. but like I said, I don't think this is a strictly hard clipping thing so probably not really fair to include it here.

Attached is my initial evaluation and comparison.

Please note that my hearing and perception may differ from yours!

StandardCLIP the only participant with 5/5 Transient Definition and Neutral Tonal Balance

I love reading that!

Any thoughts on the Stereo Tool example I posted? Still unsure if it actually works correctly at 96kHz. It sure sounds _very_ different to all the other clippers.

@ bmanic - thank you for the Stereo Tool example!

I compared the (volume matched) Stereo Tool clip to the other participants.

I have now included Thimeo Stereo Tool in my evaluation and comparison chart (attached).

Awesome!

I think Stereo Tool is a bit weird, and I'm pretty sure it's not just pure clipping. You might want to add that in your note with an asterisk.. or include it in a separate section under "strange DSP process". Also, not exactly sure what the default settings of Stereo Tool are. I just bounced with my personal current default starting point. I'll try to do another one with the actual factory defaults.

Cheers!
bM

Thought this was fitting for the shootout since a lot of those plugins are in this video.

Turns out in hard clipping mode, boz little clipper, free clip, gclip, and flatline are EXACTLY THE SAME. Proved with a null test with both a sine sweep and song.

Crazy how clever marketing and a fancy GUI might influence what we hear when mixing music! Hope you find this interesting.

only while hard clipping though. dont know whats so exciting about that finding

Attached please find PluginDoctor harmonic analysis charts for the tested clippers.

PluginDoctor was unable to analyze Kazrog KClip and LVC-Audio Clipped-MAX.

The plug-ins were configured as follows:
Global settings: Gain/Drive 0.0 dB, Ceiling 0.0 dB, Hard clipping, 100% wet:
• Boz Digital Labs Big Clipper: Crossover Mode - Clipper | Limiter​ / 20 kHz crossover, Shape 100% (hard)
• IK Multimedia Classic T-RackS Clipper: Slope 0.0 (hard)
• Joey Sturgis Tones JST Clip: Trim 0.0 dB
• Newfangled Audio Saturate: Clipper shape 100%, Detail preservation 100%
• SIR Audio Tools StandardCLIP: Mode: Hard Clip, 256x Oversampling
• Stillwell Audio Event Horizon: Clip mode, Soft Clip 0.0 dB
• SubMission Audio Flatline: Shape 100%, 4x Oversampling
• Tokyo Dawn Labs TDR Limiter 6 GE: Quality - Insane, Clipper only, Mode - Brick wall, Threshold 0.0 dB, Knee 6.0 dB, HQ Rendering
• Venn Audio V-Clip: Hard Clip, 256x Oversampling - Linear Phase
• Voxengo OVC-128: Hardness 100%, 128x Oversampling

PluginDoctor was configured as follows:
• Sample rate: 96 kHz
• Quality: Highest
• Speed: Realtime
• Test frequency: 999.0 Hz
• Level: 0.0 dB

Wow at Event Horizon! Oldi but goldi, as it seems. Super clean!

thanks for this great work! didnt expected this result.

Yeah some surprises in there (Event Horizon especially). Thanks!

If I read your settings correctly, none of the plugins should do any clipping? In other words, any plugin that shows harmonics does something beyond clipping?

Yes, several clippers are introducing harmonic distortion, including aliasing, during normal processing.

Is it possible to analyze Kazrog KClip and LVC-Audio Clipped-Max in Plugindoctor by using them in Metaplugin?

Attached please find the harmonic analysis chart for LVC-Audio Clipped-MAX 2.1.7

Clipped-MAX was loaded in PluginDoctor via DDMF's Metaplugin.

PluginDoctor was unable to load/analyze Kazrog KClip directly, or via DDMF's Metaplugin.

The plug-in was configured as follows:
Global settings: Gain/Drive 0.0 dB, Ceiling 0.0 dB, Hard clipping, 100% wet:
• LVC-Audio Clipped-MAX: Master module, Hard algorithm, Harmonic 100%, Linear-HQ 128x Oversampling

PluginDoctor was configured as follows:
• Sample rate: 96 kHz
• Quality: Highest
• Speed: Realtime
• Test frequency: 999.0 Hz
• Level: 0.0 dB