Position:Home >> News

Products Category

Products Tags

Understanding EQ Curves – Why Identical EQ Settings Can Sound Different

Date:2020/1/9 14:33:42 Hits:

You’ll often see posts in online forums from people who, despite using the same settings on two different EQs, prefer the “sound” of one EQ over another — even though they don’t really know why. But, there’s a way to quantify these differences and to evaluate how different EQs affect the sound. This gives you far more power to choose the right EQ for the right job, compared to just inserting plug-ins haphazardly, trying different settings, and crossing your fingers.

Filter Basics

To understand why different EQs can sound different even with the same settings, first we need to look into some basics. If you’re familiar with EQ parameters, feel free to skip ahead to the next section.

EQs have different frequency response curves — after all, EQ is about altering frequency response. All equalizers have parameters for adjusting these response curves, and each parameter affects the sound in some way. There are three main equalizer filter designs: highpass/lowpass, shelving, and parametric.

The highpass and lowpass filters, not surprisingly, pass high or low frequencies respectively (Fig. 1). The filtering action starts at a certain cutoff frequency (the frequency where the alteration is 3dB down). However, the response doesn’t just come to a dead stop; instead, the response rolls off a certain number of dB per octave, as if going down a slope. There may also be a Q parameter that adds a boost at the cutoff frequency or that alters the slope, depending on the design.

Shelving EQ (Fig. 2) also has a cutoff frequency and can sometimes add resonance at the cutoff frequency. The main difference compared to a lowpass or highpass filter is that the response doesn’t keep dropping down a slope but evens out after reaching a minimum or maximum amount of gain.

A parametric EQ (Fig. 3), sometimes called an EQ with a bell response, has a resonant frequency, and the response is either boosted or cut at this frequency. The range over which this boosting or cutting occurs is called the bandwidth, which is usually specified in octaves. Resonance, also called Q, correlates to the amount of bandwidth and gain. With some equalizers, for a given amount of gain, a higher resonance will have a narrower bandwidth, while a lower resonance will have a wider bandwidth.

Why EQs Can Sound Different

One of the main reasons EQs with the same settings can sound different is because many modern EQs emulate the analog EQs from classic consoles and outboard gear. However, analog EQ design involves multiple tradeoffs. For example, the designer may have wanted a steeper slope, and was willing to trade off frequency response or phase anomalies. It’s also worth noting that analog EQ stages typically have some degree of phase shift. Combining these stages together, for example with a four-stage EQ, can produce subtle cancellations or additions at certain frequencies. This may seem undesirable, but it accounts for the “character” of many classic EQs.

We needn’t get bogged down in technical details — suffice it to say a digital EQ that emulates one particular type of analog EQ can indeed sound “different” when compared to a digital EQ that emulates a different analog EQ design because these EQs will have different frequency response curves and slopes. Emulating a certain design doesn’t mean a particular EQ plug-in is inherently better; however, that particular plug-in might be better for a specific application.

For example, the SSL E Series EQ reduces Q as you increase a band’s level, so the bandwidth remains constant at different gain settings. With drums, that design can have a more “musical” result because it’s possible to combine higher Q with lower gain for affecting individual drums.

Passive vs. Active EQs

Another difference is that the emulated EQ may have a passive or active circuit design. Passive EQs don’t use gain in the filter sections themselves, so theoretically the filters can only cut, not boost. However, many passive designs include an output amplifier that adds makeup gain to all the filter sections, and the filters’ boost/cut controls add a constant amount of attenuation at the “0” setting that offsets the amount of output boost. Therefore, when you’re boosting at a particular frequency, the filter circuit itself is simply reducing the amount of attenuation, which results in a “boost” by taking advantage of the gain provided by the output amplifier.

Active EQs are variations on amplifiers, so they can boost as well as cut and can add resonances that would be difficult to achieve with passive EQs. Many people consider passive EQs gentler and more “musical,” and active EQs better at problem-solving.

The following examples are based around Waves plug-ins because the company offers a wide variety of EQ types. But how do you know which type of EQ is right for a particular application? Let’s come up with a repeatable approach that tells you what you need to know about different EQ types.

Enter the Noise

Trying to differentiate among EQ characteristics is difficult with program material because it’s a moving target — the distribution of energy at certain frequencies is always changing. Fortunately, there’s a simple way to evaluate the different tonal characteristics among different EQs: run pink noise through the EQ. Pink noise is a test signal with equal energy per octave, so it provides a constant, uniform way to make comparisons.

As you adjust the settings for different EQs, you’ll hear tonal changes that highlight differences among an EQ’s characteristic curves. Monitoring the output visually through a spectrum analyzer (which shows the level of a signal at specific frequencies) further quantifies these differences and confirms what your ears are hearing.

For example, the following spectrum analyzer screenshots (taken in Studio One) compare the PuigTec MEQ-5 with the V-EQ4. Both are set for a -10dB cut at 1kHz. Neither has a Q control per se, so the curve depends on the filter design the plug-ins are emulating. Figure 4 shows the PuigTec.

Now compare that to the V-EQ4 (Fig. 5), which many consider a “musical” EQ with a vintage sound. This shows the same -10dB at 1kHz settings. The top screen is the Low Q mode, while the bottom screen is the High Q mode.

Clearly, the EQ curve slopes are quite different. The V-EQ4’s Low Q mode has a very broad bandwidth; the High Q mode is a bit narrower, but still quite broad. The MEQ-5 has a still-narrower bandwidth. This implies that you’d use the V-EQ4 for general tone shaping, whereas the PuigTec MEQ-5 would be more suitable for zeroing in on a narrower range of frequencies, which wouldn’t affect the adjacent frequencies as much.

Next up: the Abbey Road TG Mastering Chain plug-in (Fig. 6). Given that the older, passive EQs didn’t have a continuous Q control, it was crucial for the designers to choose curves that would be useful in as many situations as possible. The top screen shows the BL setting (broadest Q), the middle screen is the MED setting (medium Q), and the bottom screen shows the SH setting (narrowest Q). Again, the settings are -10dB at 1kHz (well, actually 1.02kHz… but that’s close enough).

The lowest Q setting sort of splits the difference between the V-EQ4’s low and high Q settings but tends toward the lower Q. This was likely designed for general tone shaping. The medium setting is a bit more drastic — but not quite as much as the MEQ-5. However, the narrowest Q setting is the steepest one we’ve looked at so far, which implies that it was intended to deal with specific problem frequencies, to be used when you don’t want to color the adjacent frequencies any more than necessary.

Now let’s see how the Abbey Road Studios RS56 Passive EQ shapes the sound (Fig. 7). Settings are -10dB at 1.024kHz. Like the Abbey Road TG Mastering Chain, it had a choice of fixed Q settings; from top to bottom, the screenshots go from broadest to sharpest.

The curves themselves are quite similar to the Abbey Road TG Mastering Chain, presumably because both EQs are based around passive circuits. However, I think I sense a feature request in the development of the RS56 — instead of the TG’s single sharp setting, the RS56 has two sharp settings: one a little less narrow than the TG, the other slightly narrower.

It’s also worth mentioning another important point: frequency response curves aren’t all there is to the story. Modeling solid-state or tube EQs can give different results, and inductor-based filter circuits can sound different compared to filters based solely on resistors and capacitors. These differences are often subtle, but they also contribute to different EQ timbres.

Now, let’s look at narrow and wide Q settings for the Q10, a modern parametric EQ (Fig. 8). The top curve shows -10dB at 1kHz with a Q of 1 (the settings in the plug-in face panel shown), while the bottom curve shows the same settings with a Q of 30.

Compared to the previous “vintage” EQs, the low Q on the Q10 can go much lower, and the high Q can go much higher. With this kind of versatility on offer, it may seem odd that many people prefer the more “limited” EQ types. However, one of the biggest traps of working in the modern studio is the nearly infinite number of potentially distracting choices — yet inspiration often wants to move as fast as possible. The earlier EQs, due to the limited technology, forced choices that a consensus of engineers and designers deemed as “musical.” So, not only do you have an emulation of a particular technology with these EQs, but you can also draw on the background of those who literally created the sound of popular music.

But there’s also a “split the difference” option: the Waves H-EQ Hybrid EQ. It provides the usual parametric controls and also seven distinctive curves. So, you can choose a curve quickly that at least approximates the results you want, and if needed, you can further adjust the frequency, bandwidth, and boost/cut controls. The following seven curves show the H-EQ cutting by -10dB at 1kHz with the Q at midpoint. From top to bottom, the modes are US Vintage, UK Vintage 1, UK Vintage 2, US Modern, UK Modern, Digital 1, and Digital 2 (Fig. 9).

Now take a listen to what each of those curves actually sounds like. This example uses the H-EQ with a ?10dB cut at 1kHz. The audio steps through each of the filter modes in the same order as the images (Fig. 9; top to bottom).

Some of these curves look similar to what we’ve seen before, but remember, the bandwidth is variable, which can change the nature of the curves considerably. Even with the same settings, the curves have different timbres, as you’ll hear from the audio examples toward the end.

Finally, let’s look at the H-EQ when it’s boosting instead of cutting (Fig. 10). We haven’t gone into boosting much because that would have doubled the number of screenshots, and personally, I often use cutting more than boosting, so the cutting performance interests me the most. The following figure shows the same curves in the same order as above but with a +10dB boost instead of a cut.

Now hear what that +10dB boost sounds like on all the curves above (presented sequentially).

Remember — the gain and Q settings are the same within each audio example. The differences you’re hearing are due to the differences in the curves, as caused by emulating different filter topologies.

Here’s another example that steps through the same modes in the same order as above but with the Q setting at maximum and gain at -18dB.

This final example steps through the same modes with the Q setting at maximum and gain set to +18dB.

You can run the same experiments with any equalizer by feeding in some pink noise and varying the controls as you listen. Ultimately, instead of just clicking on buttons, trying different plug-ins, and hoping that you find something suitable, you’ll be able to choose the right curve for the right job — because you’ll know exactly how each curve affects the sound.

Prev:Burk Technology Plus-X 300

Next:Broadcast Tools AES Audio Sentinel Web

Leave a message

Message List

Comments Loading...

Products Category

Products Tags

Fmuser Sites

Understanding EQ Curves – Why Identical EQ Settings Can Sound Different

Leave a message

Message List