Tags: posts polarity-music Plugins

DSEQ3 vs Spectral Compressor vs Spectraft - Spectral Plugin Comparison

Tutorial | Aug 22, 2025

Today I compared three spectral compressors, DSEQ3, Spectral Compressor, and Spectraft, explaining my preferences, typical use cases, and cautioning against overuse due to their potential to strip music of its character. I prefer DSEQ3 for its smart AI features, custom threshold curves, and affordability, but use Spectral Compressor mostly for spectral sidechaining and find Spectraft to be the most versatile, albeit pricey. Ultimately, I advise subtlety with these tools, emphasizing that a well-balanced mix done at the track level always sounds better than heavily processing the master.

You can watch the Video on Youtube

Short Overview

Today I shared my experiences using different spectral compressors, including DSEQ3, Specraft, and Spectral Compressor. I explained why I prefer DSEQ3 over Soothe and how I use these tools carefully to avoid over-processing my music. I also highlighted the dangers of removing too many resonances and stressed the importance of subtlety when working with these plugins. Overall, I use spectral compressors mostly for resonance control and sidechaining, but always aim to preserve the character and balance of my mix.

Introduction

Today I want to share my thoughts and workflow with three spectral compressors: Spectraft, DSEQ3, and Spectral Compressor. I've received a lot of questions about which one I use, why I choose it over others, and what I would recommend. I will break down how I use each plugin, the reasoning behind my choices, and provide some insights that may help you decide which tool fits your needs.

DSEQ3: My Go-To Spectral Compressor

Why I Prefer DSEQ3 Over Soothe

DSEQ3 is my main spectral compressor of choice. While a lot of producers swear by Soothe, I prefer DSEQ3 primarily because it is not iLok-based, which I find inconvenient. Besides, DSEQ3 is packed with additional features and comes at a more accessible price point. Essentially, you can think of DSEQ3 as a beefed-up, less restrictive, and cheaper version of Soothe.

Intelligent Features That Matter

DSEQ3 includes smart AI functionality to help set the threshold optimally. One highlight is its custom threshold curve feature. I can analyze a track’s spectrum shape and apply that exact curve as a threshold for another track, which is not possible in Soothe. This allows for powerful referencing across projects.

Using Threshold Curves Effectively

Unlike an EQ, the curve in DSEQ3 represents the compression threshold across the spectrum. For example, I often set a threshold that excludes bass frequencies, because I want my sub-bass (often just a stable sine wave) to remain untouched and consistent. Compressing the sub can result in unwanted volume fluctuations, so I filter it out from spectral compression.

Compression Controls Explained

DSEQ3 includes controls that are similar to a traditional compressor, such as attack and release, but the values are given in percentages, not milliseconds. I favor slow attack and release settings: this way, only sustained tonal resonances are reduced rather than short transients like drum hits, keeping the mix musical.

The Danger of Over-Processing

One critical note: spectral compressors are incredibly powerful, and overuse can suck the life out of your music. Resonances are a natural and essential part of sound and music. Removing all of them leaves a flat, sterile result. My advice is always to apply spectral compression sparingly and in context.

Selectivity and How I Use It

DSEQ3 offers a "selectivity" control. On "sharp," it targets single frequency bins; on "soft," it affects broader bands. High selectivity is good for surgical fixes (like removing rogue frequencies), while lower selectivity produces more natural, musical reduction. I usually set this somewhere in the middle for routine tasks.

Ratio and Gain Reduction Strength

The "strength" parameter functions similarly to a ratio on a compressor, but instead acts as a multiplier on the gain reduction. I rarely push this above 1x, preferring subtlety.

Custom Curves and Reference Matching

DSEQ3 allows you to generate a custom threshold based on an analyzed spectrum. For instance, if I like the frequency balance of a certain track, I can analyze it, set that as my threshold curve, and apply it to my material to quickly match tone and energy distribution. This is very useful, though not something I use every time, as most mixes benefit from a more general approach (like a 3dB-per-octave pink noise slope).

Using DSEQ3 as a Visual Analyzer

Often I use DSEQ3 just as an advanced visualizer. I’ll place it on my master, see which areas it wants to compress, and use that information to make manual EQ or level adjustments within my mix. The plugin is not always left active, but it guides my workflow.

Spectral Compressor: The Free, Barebones Option

Pros and Cons

The Spectral Compressor is completely free and open source, which is a huge plus, especially since it's Linux-compatible. However, it lacks a lot of the convenience and advanced features of paid options like DSEQ3 and Spectraft. Controls are limited mostly to global threshold, slope, and window size. There is no selectivity control, so all frequency bins are processed individually, which quickly leads to a harsh, “digital” sound if overused.

Limited EQ Curve Flexibility

While you can bias the threshold curve to be more aggressive in certain frequency ranges, you cannot draw or save custom curve shapes, and there’s no easy way to exclude bass or selectively target certain frequency groups outside of broad slopes.

Upwards Compression

One unique feature of the Spectral Compressor is upwards compression, which can “bring up” quiet details in specific frequency areas. In practice, I’ve found it sounds unnatural and often results in distortion if pushed hard, especially on drums. I rarely use this, and if I do, it’s only in dry/wet parallel setups for subtle enhancement.

Best Use Case: Sidechaining

The main thing I rely on the Spectral Compressor for is spectral sidechaining, such as ducking the bass bus only when the kick or snare hits. It’s powerful for frequency-selective sidechain relationships, letting the relevant frequencies dip mid-mix rather than chopping out entire instruments.

Spectraft: The Sophisticated and Pricey Solution

Features and Workflow

Spectraft is a more recent addition to my floating toolset. With it, you get a more traditional compressor workflow, with ratio, attack, and release controls all specified in milliseconds. Spectraft supports “profiling,” where you record the spectral shape of any track and use it as your compression curve for other material, similar to DSEQ3’s custom threshold but integrated into the plugin’s ecosystem.

Selectivity and “Q” Parameter

Spectraft’s "Q" control dictates how surgical vs. broad the processing is: at 0 percent you get very wide (broadband) changes, while at 100 percent it works per-frequency-bin, similarly to “sharp” mode in DSEQ3. Again, middle values are a good compromise for musical applications.

Adaptive Dynamics and Upwards Compression

A unique innovation in Spectraft is its adaptive dynamics section. When you reduce frequencies heavily, overall gain is also reduced. The plugin can intelligently “add back” global EQ moves (upwards compression) to compensate, restoring perceived loudness or tonal balance lost during correction. You can further tune the dynamics and tilt the spectral balance with additional controls.

Practical Applications

The profiling feature is an excellent quality-of-life advancement, letting me steal frequency curves from my reference library and apply them selectively. Like the others, I often use Spectraft as an analyzer to guide mix decisions rather than fix everything with blanket compression.

Key Concepts Explained

What is Spectral Compression

Spectral compression splits the incoming audio into hundreds or thousands of narrow frequency bands (using FFT or similar techniques), applying compression independently to each band. This allows for dynamic control over specific frequencies, useful for taming harsh resonances, smoothing vocals, or creating cleaner, more dynamically even mixes. However, aggressive use can result in lifeless, unnatural audio because music’s character is found in the complex interplay of frequencies, including their resonances.

Selectivity or "Q"

This feature controls how many frequency bins are grouped for processing. Higher selectivity means more surgical, bin-by-bin adjustments (good for precise fixes, risky for musicality). Lower selectivity means broader bands, preserving more of the music's natural character.

Custom Threshold Curve

Instead of using a straight-line threshold, you can set the compressor to follow a target frequency balance (matching pink noise, white noise, or an analyzed track). This lets you tailor dynamic correction to specific spectral goals.

Upwards vs. Downwards Compression

Downwards compression reduces the loudest elements above threshold. Upwards compression lifts quiet elements below threshold. Used with care, upwards compression can reveal detail, but too much can sound very artificial, especially when working per-frequency-bin.

Sidechaining in the Spectral Domain

Unlike traditional sidechain compression that ducks an entire track when another triggers, spectral sidechaining ducks only those frequencies which overlap, preserving clarity and punch in dense mixes.

My Workflow Strategy and Philosophy

In my experience, subtlety is critical. I use spectral compressors to visualize, diagnose, and apply the lightest touch needed to control resonances or fix frequency imbalances. Overuse will strip away what makes music engaging , the interplay of texture, dirt, and resonance. Automated, one-size-fits-all processes rarely yield the best sound. Manual, intentional decisions produce mixes that are more vibrant and unique.

Conclusion

Let me know about your experiences with spectral compressors, and if you have tips or disagree with anything here, feel free to share your perspective.

Full Video Transcription

This is what im talking about in this video. The text is transcribed by Whisper, so it might not be perfect. If you find any mistakes, please let me know.
You can also click on the timestamps to jump to the right part of the video, which should be helpful.

Click to expand Transcription

[00:00:00] Hey folks, welcome back.
[00:00:01] So today I want to talk about the Spectral Compressors here,
[00:00:04] Spectraft, DSEQ3, and Spectral Compressor.
[00:00:08] And some people ask me what do I use, why I use it,
[00:00:11] and what do I recommend.
[00:00:13] So I want to give you my personal opinion here
[00:00:15] what I use and how I use it and why,
[00:00:18] and also give you some, maybe some ideas, I don't know.
[00:00:22] So the Spectral Compressor I showed you in the last video,
[00:00:25] so I want to start here with DSEQ3,
[00:00:28] because this is something here that I would say
[00:00:30] is the same as Sooth.
[00:00:32] So I don't use Sooth, and I know a lot of people use it,
[00:00:36] but I don't use it, I use this one here,
[00:00:39] because it's not iLog based, I hate iLog for some reason.
[00:00:44] It also has some neat features on top of that.
[00:00:48] For instance, here, there's some smart AI stuff
[00:00:53] to figure out the threshold,
[00:00:55] there's here a custom threshold curve,
[00:00:57] so you can analyze a track and then take the spectrum curve
[00:01:02] or the spectrum shape and put it on another track.
[00:01:06] That's not possible with Sooth here, you can do it,
[00:01:09] and it's also way cheaper.
[00:01:11] So that's why I use DSEQ3, so if you use Sooth,
[00:01:14] you can just say this is more or less the same thing,
[00:01:17] just with more features.
[00:01:18] You can also see I take out the low end,
[00:01:21] and this is not an EQ, this curve here,
[00:01:23] this is the threshold curve,
[00:01:25] which means there is no threshold for the bass here,
[00:01:29] and this is here currently on the track on the master.
[00:01:32] And for the bass sound or for the sub,
[00:01:39] I usually just use a sine wave,
[00:01:40] and the sine wave is stable as stable AF, okay?
[00:01:45] So why do I should put here a compressor on it
[00:01:48] and mess with the volume of the bass, right?
[00:01:52] So this is here the low cut, you can disable it,
[00:01:55] and you can see it wants to instantly pull down here
[00:01:58] the volume of the sub,
[00:01:59] and you can see there's a fluctuation,
[00:02:01] it goes up and down.
[00:02:02] I don't like it, I want to have a stable bass,
[00:02:04] so I usually just take it out here,
[00:02:07] and just leave the compressor on the top end
[00:02:12] or on the mid range here.
[00:02:14] And yeah, this kind of works like a compressor almost,
[00:02:18] it looks like it because we have here something
[00:02:22] like chain link or channel link, not chain link,
[00:02:26] how it links basically the left to the right channel
[00:02:28] or if it processes this individually.
[00:02:31] There's here attack time, release time,
[00:02:35] but these two are not in milliseconds,
[00:02:37] it's in percentage for some reason,
[00:02:40] so that's why I mean that it's almost a compressor.
[00:02:44] So I have this here always on very slow attack times
[00:02:49] and release times,
[00:02:50] because if you pull down the attack time here,
[00:02:52] it reacts to every incoming little transient,
[00:02:57] and it sounds not good.
[00:02:58] So when I have here a slow attack time,
[00:03:12] means it only reacts to tonal sounds that are a bit longer.
[00:03:16] So it only detects or gets in sounds that are longer,
[00:03:20] not that the transients,
[00:03:22] not the attack times of the kick and so on.
[00:03:24] So you don't mess with that,
[00:03:26] you just want to reduce some resonances here
[00:03:29] in certain frequencies.
[00:03:31] When we talk about resonances,
[00:03:33] I have to make basically this kind of disclaimer.
[00:03:36] These tools here are very dangerous in my opinion.
[00:03:40] If you don't know what you do,
[00:03:42] and you just put it on and pull this down, right?
[00:03:45] I see this all the time,
[00:03:46] people do it on YouTube videos with Soothe,
[00:03:49] they just pull this down.
[00:03:51] You remove basically the resonances,
[00:03:53] which is kind of what you want to do,
[00:03:55] but if you do it too much,
[00:03:56] you more or less remove music from the music,
[00:04:00] because music is resonances,
[00:04:02] and if you remove all the resonances,
[00:04:05] there is nothing left to listen to, okay?
[00:04:08] So this is a bit of a drastic thing to say,
[00:04:11] but you know what I mean,
[00:04:13] you have to really use this in kind of subtle ways,
[00:04:18] because you can destroy a lot of things
[00:04:20] pretty easily with this.
[00:04:22] You just pull this down,
[00:04:24] and your music is basically gone.
[00:04:25] And now it sounds very spectral, spectral-ish.
[00:04:38] Also, we have just something like selectivity,
[00:04:40] which is interesting,
[00:04:43] because not every spectral compressor has this.
[00:04:46] For instance, here, the spectral compressor
[00:04:48] doesn't have this.
[00:04:49] So the selectivity means,
[00:04:52] when we pull down here the threshold,
[00:04:53] you can see individual frequencies are getting reduced.
[00:04:58] This is because we have this here on sharp.
[00:05:02] If we pull this down to, let's say, soft,
[00:05:06] then you can see we have more like a broad band
[00:05:10] gain reduction.
[00:05:11] So the selectivity groups certain frequencies together.
[00:05:15] So if you have this here on sharp,
[00:05:18] it takes basically every small little frequency,
[00:05:21] every frequency bin individually and pulls it down,
[00:05:25] which can make it sound very harsh or very digital fast.
[00:05:30] And if you have this here on soft,
[00:05:32] it's more like one band.
[00:05:34] You have one band, and this is one band gain reduction,
[00:05:37] so it doesn't make any sense, mostly.
[00:05:40] So you want to have this here on the middle ground,
[00:05:42] so something around 50, I don't know,
[00:05:45] where you have individual frequencies,
[00:05:47] but they're kind of grouped together
[00:05:48] into small little bands.
[00:05:50] So this is at least how I see it.
[00:05:56] So if you have, let's say, you have a recording,
[00:05:59] and there's some kind of small little single sound,
[00:06:04] single frequency in there that you want to reduce,
[00:06:06] then you can pull this up to sharp
[00:06:09] and just see if you can take or bring
[00:06:13] this one frequency down.
[00:06:14] So it's more like for restoration or damage control,
[00:06:21] but just for musical tasks,
[00:06:25] and you want to reduce some resonances,
[00:06:27] so you want to have this here on the middle ground,
[00:06:29] more or less.
[00:06:30] Then we have something here like gain reduction strength,
[00:06:34] which is something like ratio on a compressor,
[00:06:36] but it's not like the same here.
[00:06:39] You have a multiplier,
[00:06:40] so everything that's above the threshold
[00:06:42] gets pulled down one time.
[00:06:44] It's the strength, which is okay.
[00:06:47] So what goes above the threshold gets pulled down,
[00:06:50] which is okay, but you can also go up here to 10 times.
[00:06:53] So everything that's 1 dB above the threshold
[00:06:56] gets pulled down probably by 10 dB.
[00:06:59] So it's more or less like a ratio thing.
[00:07:03] So if you have like harsh frequencies
[00:07:05] that are just small in size
[00:07:07] and go barely above the threshold, but you can catch it,
[00:07:11] and then you want to pull it down heavily,
[00:07:13] then you can go up to 10 here,
[00:07:15] but usually I have this around one or maybe even less,
[00:07:19] just to have subtle changes to tame some of the frequencies
[00:07:23] here in a very subtle way.
[00:07:26] So that's usually how I use the DSEQ3.
[00:07:30] There's also something in here like a custom threshold.
[00:07:34] I also explained this, I think, in a video.
[00:07:36] It's very easy.
[00:07:38] At the moment here, we have a very straight threshold line,
[00:07:42] and you can change the line here with the slope,
[00:07:44] and at the moment here, it's 3 dB bar octave,
[00:07:47] which is pink noise.
[00:07:48] You can go to 0 dB, which is white noise.
[00:07:51] So you can see you catch more low end frequencies.
[00:07:55] You basically tilt or skew here the visualizer
[00:07:59] or the analyzer kind of.
[00:08:01] I usually have this at 3 dB per octave,
[00:08:04] which means the threshold is exactly in a straight line
[00:08:07] for 3 dB per octave or pink noise,
[00:08:09] and if I pull down the threshold,
[00:08:11] I kind of get a nice pink noise balance for the frequencies.
[00:08:16] You can also go up to 4.5 dB per octave,
[00:08:20] or I think it goes up to nine.
[00:08:22] If you only want to tame the upper frequencies
[00:08:25] and the base frequencies are never actually touched
[00:08:28] because it's so tilted.
[00:08:30] So I have this usually at 3 dB here, which is okay,
[00:08:35] but you can also go and use a custom curve.
[00:08:38] So here in the background, you can hear
[00:08:40] there's a track playing, right?
[00:08:42] And I can say, oh, I really like this distribution
[00:08:46] of the frequencies.
[00:08:47] I really like how it sounds.
[00:08:49] So you can analyze this here with a custom threshold.
[00:08:53] It just recorded for a moment.
[00:08:55] And when you're done, after a while, I don't know,
[00:09:06] you just have a new custom threshold here.
[00:09:09] You can also save this as a preset.
[00:09:11] But now we have to use here a slope of zero dB
[00:09:15] because we have a custom curve
[00:09:17] and we enable here the custom curve.
[00:09:18] You can see it drastically or heavily reduces here
[00:09:23] the volume or the frequencies.
[00:09:26] So to use this custom curve,
[00:09:28] we have to go up to 30 dB per octave.
[00:09:30] You have to use a slope here
[00:09:32] and we have to enable the custom curve.
[00:09:34] And now we have here, instead of the straight line
[00:09:37] of 3 dB per octave, we have basically
[00:09:39] the frequency distribution of the track
[00:09:42] we are currently rocking on.
[00:09:44] And we just pull this down slightly here
[00:09:47] and we make sure that every frequency
[00:09:50] is basically aligned with this curve.
[00:09:52] We just analyzed.
[00:09:53] We can also just analyze a different track,
[00:09:56] a different master track, for instance,
[00:09:59] and then see what's going on.
[00:10:01] And this is also something I do with DS EQ 3 a lot.
[00:10:05] I don't use it for resonance reduction all the time.
[00:10:10] I sometimes just see or look
[00:10:14] what the spectral compressor here is doing, right?
[00:10:17] I pull this down and I see there's a lot of activity here.
[00:10:22] Or here and then I know, oh, there's maybe too much content
[00:10:26] in this range or in this frequency area.
[00:10:30] And instead of just having this on the master
[00:10:32] and letting this work very hardly
[00:10:35] and very in a very hard way,
[00:10:39] I just go to the bus or to the individual instrument
[00:10:42] and just pull it down until it looks nice to my eyes here.
[00:10:46] And it sounds nice to me.
[00:10:47] Of course, that's the most important thing,
[00:10:49] how it sounds, right?
[00:10:50] And then I pull down the individual instruments
[00:10:52] or buses or whatever, and then I just remove this.
[00:10:56] I don't use it at all.
[00:10:57] I just use this basically as a visualizer
[00:11:01] or some type of analyzer to see what's going on
[00:11:05] actually in my music.
[00:11:07] So this is how I use the DS EQ 3 here.
[00:11:10] The custom curve is not something I use all the time.
[00:11:13] Mostly I just leave it here on 3 dB per octave,
[00:11:17] 3 dB or at pink noise, have a very slow attack,
[00:11:22] very slow release and then pull this slowly down
[00:11:26] and see what's going on.
[00:11:28] But here you can see my mix down is pretty flat already.
[00:11:33] So it tries to take down here every frequency
[00:11:38] in the kind of in the same way.
[00:11:39] So for me, this is pretty fine.
[00:11:42] I probably don't need to use a spectral compressor
[00:11:44] here at all.
[00:11:45] So yeah, that's what I can see at least here for this thing.
[00:11:51] So my mix down is pretty flat already.
[00:11:53] So this one had a lot of features already
[00:11:58] that you don't have in this one here.
[00:12:00] So this one is free, but it's also pretty raw
[00:12:03] and it doesn't lack some features.
[00:12:07] So I use this most of the times only to compress
[00:12:11] or to side chain my bass bass with the drums.
[00:12:16] Like I showed you in my last video,
[00:12:20] I don't use the spectral compressor on the master at all
[00:12:24] because it lacks this kind of selectivity feature here.
[00:12:27] So when I pull down here the global threshold,
[00:12:30] it tries to reduce all the little frequency bins
[00:12:36] individually, which makes it sound very spectral
[00:12:40] on digital very fast, in my opinion.
[00:12:43] So we can't group certain frequencies together.
[00:12:47] It always works on a per bin base, this.
[00:12:52] Which is not something I really like.
[00:12:56] But here you can also kind of low cut here the analyzer
[00:13:02] by using here where is it high frequency roll off.
[00:13:07] Or you can also make, you only can make a roll off here
[00:13:10] at the top end, but the low end is missing.
[00:13:13] So you can't take out here the bass.
[00:13:15] You can play around here with the threshold slope
[00:13:18] and with the curve, right?
[00:13:20] We can do something like this,
[00:13:23] but you can't really do here some individual things
[00:13:27] where you enable your certain bands and say,
[00:13:29] I want to reduce here some frequencies more than here
[00:13:32] and so on.
[00:13:33] This is not possible here.
[00:13:35] We have just a straight line and you can bend the straight line
[00:13:39] and that's more or less it can also change it this.
[00:13:42] So we can say, oh, only reduce the top end,
[00:13:45] take the low end out.
[00:13:47] This is possible,
[00:13:48] but you have to walk around certain problems here.
[00:13:50] Let's go back here to 3DB,
[00:13:54] which is also kind of pink noise.
[00:13:56] So I like to have the 3DB thing here
[00:13:58] just to see what's going on.
[00:14:00] Here it's more like a compressor.
[00:14:04] We have also upwards compression,
[00:14:06] which is kind of missing here.
[00:14:07] There is no upwards compression.
[00:14:09] So this is, I tried to use this on drums
[00:14:13] with upwards compression.
[00:14:15] You can see here it's trying to pull up some frequencies.
[00:14:18] I have to switch it on.
[00:14:24] But it doesn't sound good, right?
[00:14:31] It sounds very distorted
[00:14:33] 'cause it tries to push up all the individual frequencies.
[00:14:37] You can reduce the frequency bins here
[00:14:39] by using here the window size,
[00:14:42] but then you have more or less one bin
[00:14:45] for the frequency range from zero Hertz to 1K,
[00:14:49] one kilo Hertz,
[00:14:51] which is just too broad in my opinion.
[00:14:56] So you want to at least have here 2K.
[00:15:02] Let's reduce this here a bit more.
[00:15:04] Right, this doesn't sound right.
[00:15:13] It sounds way too digital.
[00:15:14] So you have to be very, very, very carefully
[00:15:18] with the spectral compressor on the master
[00:15:20] if you want to use it there.
[00:15:22] But you can use it like in the same way
[00:15:25] as I use here the DSEQ3
[00:15:27] just to look for things that it wants to change.
[00:15:31] So let's go back here.
[00:15:35] Right, you can see I want to take away the bass,
[00:15:41] but this is pretty fine
[00:15:42] because this is a 3D beeper octave.
[00:15:45] I don't want to have my track at 3D beeper octave
[00:15:48] as a slope or as a frequency distribution.
[00:15:50] It's just a rough guide for me.
[00:15:52] And here having some bass and some tops in there
[00:15:56] is pretty fine.
[00:15:57] I want to have this kind of smiley curve.
[00:16:00] I don't want to have a flat straight line for 3D beeper octave.
[00:16:03] So I guess I could dial in here something like a curve,
[00:16:08] something like this.
[00:16:09] But in my opinion, it's not versatile enough
[00:16:13] here, this threshold curve.
[00:16:15] So I use it more like an indicator or an analyzer
[00:16:19] and I use it mostly on drums, if at all.
[00:16:22] But my main use case for this
[00:16:26] is actually only side chaining things.
[00:16:30] I can show you how this sounds here on the drum bus
[00:16:35] here for a moment.
[00:16:36] Do I have it here on here?
[00:16:39] I have this EQ on the drums,
[00:16:41] just reducing slightly stuff, as you can see.
[00:16:44] So let's mute the drums or solo the drums here.
[00:16:49] So let's put here the spectral compressor on it.
[00:16:59] So what we can do here is we can use upwards compression
[00:17:02] that sounds a bit like OTT.
[00:17:03] But I would also say it doesn't sound natural.
[00:17:20] It sounds very digital and you have to be very carefully
[00:17:24] with using this.
[00:17:27] In my use, I try to use this on drums multiple times,
[00:17:30] but it always sounds so distorted
[00:17:32] and so artificial in a way.
[00:17:36] Maybe you can recommend how to use it on drums,
[00:17:39] but I never found a way to actually make it work
[00:17:42] for me on drums and sound great.
[00:17:45] I always have this kind of noisy sound to it.
[00:17:47] Maybe in a very subtle way.
[00:17:55] (upbeat music)
[00:17:57] And we leave some transients through.
[00:18:06] (upbeat music)
[00:18:09] And to me, it sounds always better without spec.
[00:18:25] So I fiddle around with this all the time
[00:18:28] and try to use it on drums, right?
[00:18:30] Just to see if I actually miss something,
[00:18:33] but when I'm done with fiddling with this compressor
[00:18:36] and I just mute it or disable it,
[00:18:38] it always sounds better to me.
[00:18:40] So maybe this is okay if you use very low ratios here
[00:18:46] for the drums,
[00:18:47] but if you use it drastically in this kind of way,
[00:18:52] it always sounds bad in my opinion.
[00:18:55] Yeah.
[00:18:58] Maybe if you, when you mix it,
[00:19:01] so mix in the dry signal here.
[00:19:04] Yeah, this could be something.
[00:19:09] So if you use it in parallel,
[00:19:22] this could be maybe worth a look if it works for you,
[00:19:27] but if you use it your 100% and a lot of ratio
[00:19:30] for upwards compression, downwards compression,
[00:19:32] it's always sounds too artificial
[00:19:34] because you have, like I said,
[00:19:36] all these individual bands measured and compressed
[00:19:39] and changed in volume and it doesn't sound right to me.
[00:19:42] So I don't use it on this type of things,
[00:19:48] but where I use it is here on the bass bus, right?
[00:19:52] Like in my last video.
[00:19:58] The drums coming in with sidechain compression here
[00:20:05] and then ducking basically the drum bus to the bass bus
[00:20:09] and it reduces here for the kick drum, the bass, the sub,
[00:20:12] and for the snare drum, which is just a clap
[00:20:15] or an white noise burst here at the top end from the bass.
[00:20:20] So that's how I use it usually.
[00:20:24] And also here, if you just want to reduce the bass,
[00:20:27] you just have to use a different slope here,
[00:20:29] something like this,
[00:20:30] and then you have to change here the threshold.
[00:20:33] So it's a bit off, a bit fiddly here, but it works.
[00:20:41] So let's go back here to my initial thing, 3dB.
[00:20:46] Something like this.
[00:20:49] And here you can see I have pretty fast release time.
[00:21:03] It's almost audio rate because I really want to only
[00:21:08] get the kick out of the bass more or less,
[00:21:10] also attack is here, zero MS, which is very fast.
[00:21:15] And it works for me, it sounds okay.
[00:21:18] So it doesn't change the sound too much.
[00:21:28] It just really removes the bass or the kick frequencies
[00:21:32] from the bass frequencies when they occur.
[00:21:35] So that's how I use the spectral compressor most of the times.
[00:21:40] Like I said, I try to use it on drums sometimes.
[00:21:45] Maybe as a style, if you like the style
[00:21:48] of having it sound very digital and very dubstabby
[00:21:52] or whatever, then it can be something.
[00:21:56] I also try to use it on a mid-range basis
[00:21:59] on these type of neuro bass sounds.
[00:22:01] And it also can be useful there
[00:22:04] if you just want to push up
[00:22:05] or upwards compress the bass sounds with this.
[00:22:10] But yeah, it has a certain sound to it.
[00:22:12] Then we have here in, on the master,
[00:22:18] we have here the spec craft.
[00:22:20] So this one is kind of a middle ground.
[00:22:24] So it kind of works like a compressor
[00:22:26] because we have here a real ratio setting
[00:22:29] even though it's also in percentage.
[00:22:32] We have here attack times, release times in milliseconds.
[00:22:35] So we can say really here I'm working at 174 BPM
[00:22:40] and 160 note is probably around 88 milliseconds
[00:22:44] and one eighth note is 170 MS, right?
[00:22:48] You can really tweak in the release times here
[00:22:52] if you want to, you can also leave some attacks
[00:22:55] through some transients through.
[00:22:58] But here also we have something like a threshold.
[00:23:01] We can pull this down.
[00:23:03] And then we have here a cue and we have a ratio
[00:23:07] and I have to see why this is not working at the moment.
[00:23:11] The ratio is up.
[00:23:13] Maybe it's disabled.
[00:23:14] I don't know what I did there.
[00:23:18] Spec craft.
[00:23:20] Okay, so you have probably a reduced the range probably.
[00:23:27] So yeah, you can see it reduces already frequencies.
[00:23:29] We have also here a threshold.
[00:23:31] The threshold is I think also 3 dB per octave.
[00:23:36] I don't know if you can change this,
[00:23:37] but there is here something like profiler.
[00:23:41] You can record basically all the tracks
[00:23:43] and then you have a changed threshold curve.
[00:23:48] Let's capture this here for a moment.
[00:23:51] It makes no sense.
[00:23:52] I just capture more or less the same frequency curve
[00:23:56] of this track, but you can just put it on another track
[00:23:58] and then capture this thing.
[00:24:01] You can see the threshold changed more or less
[00:24:03] in a different curve and then you can pull it down
[00:24:05] and everything that goes above this line then gets pulled
[00:24:08] down.
[00:24:09] So yeah, this is possible here.
[00:24:13] The ratio also a bit lower.
[00:24:15] So we don't want to reduce too much.
[00:24:17] So here instead of having this, what's the name?
[00:24:22] This is Q, the selectivity here, right?
[00:24:26] Here it's called Q.
[00:24:28] So if you have this at 0%, you have more like broad,
[00:24:32] broad frequency changes.
[00:24:34] And if you have this all the way up to 100%,
[00:24:36] you have individual frequencies getting pulled down.
[00:24:39] So this is possible.
[00:24:42] But here they also added some adaptive compensations, right?
[00:24:45] So we have the dynamics and the idea here is
[00:24:48] when you reduce a lot of individual frequencies here, right?
[00:24:52] You also reduce kind of the overall gain
[00:24:56] of this frequency area, right?
[00:24:59] (upbeat music)
[00:25:02] Right, so the higher it's become quieter.
[00:25:09] You reduce, of course, a lot of individual frequencies,
[00:25:13] but it gets quieter.
[00:25:14] So here they have some adaptive things.
[00:25:17] So with dynamics, you can see here,
[00:25:20] they kind of want to add back in
[00:25:24] some broad EQ changes, right?
[00:25:27] So let's pull this all the way up.
[00:25:28] You can see here, so it reduces individual frequencies,
[00:25:33] but it pushes up broader changes.
[00:25:36] So it makes basically an upwards compression here
[00:25:40] with this, just to bring back the volume.
[00:25:43] And this is very helpful sometimes.
[00:25:47] And you can change here, of course,
[00:25:49] the dynamics, the form and treble.
[00:25:51] So you have a treble here, you add more top end.
[00:25:58] But again, you want to use this very in a very subtle way.
[00:26:02] It sounds very artificial, very fast.
[00:26:06] (upbeat music)
[00:26:09] (upbeat music)
[00:26:11] But I really like the idea of this one here,
[00:26:31] and I use it more and more in certain situations
[00:26:35] because it's so versatile
[00:26:39] that you can use this curve
[00:26:40] and you have this adaptive change with the DSEQ sometimes.
[00:26:44] If I reduce too much stuff here, right, in the top end,
[00:26:49] I have to add an EQ then here after DSEQ
[00:26:54] and say, oh, bring back actually here the top end a bit, right?
[00:26:59] So with Spectraft, you have this basically in one plug in.
[00:27:04] But it more or less works in the same way
[00:27:06] as all of the other things, it's just quality of life.
[00:27:09] And it has maybe a different sound to it.
[00:27:11] And it's also that you have here this profiling thing, right?
[00:27:15] You can see I already
[00:27:16] clone some curves from different productions
[00:27:21] just to pull this down and see
[00:27:24] where I have too much content in my mix down
[00:27:28] and then push it up and correct it on the bus
[00:27:31] or maybe on the individual instrument.
[00:27:34] So also here, I use this sometimes just as a visualizer
[00:27:37] to see what's going on instead of having this on the master
[00:27:40] and just rock heavily and reduce the gain
[00:27:44] of certain individual frequencies.
[00:27:46] So yeah, this is Spectraft.
[00:27:48] I think this is pretty expensive,
[00:27:51] 150 bucks or something like this.
[00:27:54] So you don't have to use it.
[00:27:56] It's just quality of life.
[00:27:57] It does some nice things.
[00:28:00] This one here is a bit cheaper.
[00:28:01] I'm not sure how much this is.
[00:28:04] And this other one here,
[00:28:08] the Spectra compressor is completely free and open source.
[00:28:12] And I think it's also the only one that works under Linux.
[00:28:17] But you have to be very carefully with this, right?
[00:28:20] It lacks some quality of life things.
[00:28:22] And you have to be aware that you really work
[00:28:24] on all these individual frequency bins all the time.
[00:28:29] And it makes it very harsh or digital very fast.
[00:28:33] And yeah, like I said in the beginning,
[00:28:36] really try to use this in subtle ways.
[00:28:38] You can destroy your music with this.
[00:28:40] I know a lot of people like to use Soothe
[00:28:44] on more or less everything.
[00:28:45] I watch sometimes you these dubstep videos,
[00:28:48] color-based producers, and they just pull down heavily
[00:28:51] and remove everything until everything's completely flat
[00:28:55] and digital.
[00:28:56] And it's also, I get it, it's the sound.
[00:28:59] It's the aesthetics of this kind of genre.
[00:29:02] But if you do other music, sometimes drum base here,
[00:29:06] you want to have some resonances, right?
[00:29:08] You want to have some dirt.
[00:29:10] You want to have some overlaps.
[00:29:11] You want to have these sounds glued together in a way.
[00:29:17] So if you overdo this,
[00:29:20] you make these sounds stand out transparent
[00:29:22] or you single out certain sounds
[00:29:25] and you miss basically the glue.
[00:29:27] So you want to have the dirt.
[00:29:29] The dirt makes the music interesting, in my opinion.
[00:29:32] And I get it, I was at this point also in my life,
[00:29:36] I thought, oh, nice.
[00:29:38] Now we have the spectral compressor.
[00:29:40] I just pull this down.
[00:29:41] I have my curve I want to have.
[00:29:42] I pull this down.
[00:29:44] Everything is, everything sounds fine.
[00:29:46] I don't need to mix my tunes anymore.
[00:29:49] I just have this on the master and call it day.
[00:29:52] Sadly, this is not how it works.
[00:29:54] If you have a nice mix down,
[00:29:57] if you have a nice well balanced mix down
[00:30:00] and you work on these individual tracks
[00:30:03] and single out certain frequencies,
[00:30:06] it always sounds better
[00:30:08] and it always sounds more punchy, in my opinion,
[00:30:13] than having this here on the master
[00:30:15] and just put a mask with the same settings
[00:30:20] on every frequency.
[00:30:21] It just doesn't sound right, in my opinion.
[00:30:24] So try to use these very carefully.
[00:30:27] If you produce color based, then I can't help you.
[00:30:32] That's your fate.
[00:30:34] Yeah, I don't want this anyone.
[00:30:37] It's just, you know, it's type, it's part of the sound.
[00:30:40] I get it.
[00:30:41] So yeah, this is more or less my video
[00:30:45] about these type of processes.
[00:30:48] I hope it helps in a way.
[00:30:51] I'm not sure.
[00:30:52] Let me know what you think.
[00:30:53] Do you use any of these compressors
[00:30:55] or have any experiences?
[00:30:57] Maybe I got something wrong and it's with you.
[00:31:00] Let me know in the comments down below.
[00:31:02] I am open for everything.
[00:31:03] If you insult me, I just delete the comment
[00:31:08] and call it a day, okay.
[00:31:10] Yeah, that's it.
[00:31:16] I want to work on this track here today.
[00:31:18] I just started this.
[00:31:20] So that's it for the day.
[00:31:22] Thanks for watching.
[00:31:23] Leave a like, leave a subscription.
[00:31:24] Let me know what you think.
[00:31:25] See you in the next video.