I was recently shooting out a Pultec emulation against Tokyo Dawn's and Variety of Sounds Slick EQ. The latter is not an attempt to model one particular hardware device, but just a very well-thought out and straight-forward to use console style EQ that works particularly well on channels and group-channels.

The classical Pultec design (to be more precise: the EQP-1A module) is based on controls, that allow you to boost and attenuate the low frequency range and the high frequency range at the same time to achieve the desired EQ curve. Originially this was due to the technological limitations (passive circuits) of the time when the real Pultec was first introduced.

The boost for the low frequency is at the indicated frequency (eg 100 Cycles Per Second = 100 Hz) while the attenuation is centered somewhere above that. Here comes the major issue: It's hard to tell where the attenuation will take place, especially since most Pultec emulation plugins have no visual representation of the resulting EQ curve. It's pretty much guesswork from here. I found that the attenuation can take place way above, in the upper mid frequency range (and I'm talking about the low frequency attenuation control here). I would have expected it to be more in the 250 - 500 Hz range, where often you would want to remove mud, while at the same time boosting the bass and subbass area. But it's way above that!

Using Slick EQ I can just make a low shelf boost at 100 Hz and below and then use the middle bell to attenuate mud at 260 Hz (just as an example). If I tried the same using the low frequency boost and attenuation of a Pultec emulation, maybe I will succeed in doing a 100 Hz low shelf boost, but I will definitely not manage to reduce the area around 260 Hz, since the attenuation takes place much higher, as said earlier.

Why would I use a tool that is less capable to achieve a desired result, when I can use something like Slick EQ or even stock EQ to dial in exactly what I want. One argument against stock EQs and general purpose visual EQs like Fabfilter: they are sometimes overload and it's quicker to use something with a limited set and range of parameters. I'm totally with that idea, hence I love Slick EQ.

With Pultec emulations you never really know which EQ curve the combination of boosting and attenuating at the same time will result in. It's like lottery. Good luck folks!

I understand that the real Pultec is a legendary device in the history of audio engineering, being the first widely available program EQ. I'm sure that the real classical hardware has it's own boxtone. And software emulations? All of them sound different, they all have slightly (or sometimes drastically) different behavior and EQ curves. And I'm not convinced that they can deliver even an approximation of the classical Pultec boxtone.

So what's the advantage of using Pultec emulations, when you can't dial in the EQ curve you want with certainty and also won't get the genuine Pultec boxtone? Why wouldn't you rather use something like Slick EQ where you can dial in the curve you want and still add analog tone by driving the saturation stage (with various saturation types)?

 

LOL I once told an analog gearhead that unpredictability can be considered an issue, even if a small one. His reaction was spectacular.
I'll watch this thread.
You don't have to dial in the EQ with certainty tho - you're meant to mix with your ears and not with your calculator. You can tell what it does to your sound - and whether it's beneficial or not.

 

Pultec emulations... are great, because ...

I don't have access to the hardware and never used hardware.
And there are free pultec emulations and people like them and those vsts can be usefull to do some creative stuff, different workflow.
So I like them too.

 

I agree. Not with surgical precision. But at the very least I want to roughly know which area I'm attenuating when I use the attenuation dial. Not in exact Hz, but as in "lower mids" or "upper mids".

It's really hard to predict in what kind of EQ curve the combination of boost / attenuation (especially on the low frequency section) will result in, since dialing in 2 on the attenuation and 2 on the boost will also not result in an even degree of boost and attentuation. That's the kind of uncertainties I'm referring to.

This and the fact that attenuation in the low frequency section will affect the frequency range way above the range that you would expect and way above the range that I would consider to be useful when trying to dial in your low frequency response, make me doubt the usefulness of Pultec emulations.

Sure, I understand the principle of "ears first", but as an experienced engineer you often have an idea of the EQ curve you would want to apply to achieve an improvement in the context of a mix. A tool that allows you to quickly dial in the EQ curve that you envision is imho more useful, than something that comes with a lot of guesswork or even behaviour that differs from what you would expect dialing in a specific parameter would do.

 

You can measure it.

To me it's basically a tilt filter, so I simply use a tilt filter instead of pultec.

 

"...are not perticularly useful" to me. But can see it being useful for other people.

Fun thing, i was trying out UAD's emulation with the typical boost/cut trick at 100hz, and the dip ended up being way up in the 1k-2k area. But it worked, 100hz felt more pronounced even though it was a result of what's going on a lot higher.

Point is, pultecs are far from quirurjical but hey, if a certain movement of knobs result in something you know will work for what you're doing, who cares about the label?

That said, i do think they are overrated.

 

I know! That's what I did using Bertom EQ Curve Analyzer. The curves are way off if you use the frequency labels as an orientation. A 30 Hz boost turns out to be closer to a 300 Hz boost. So there is a lot of unpredictability, given this and the other odd behaviours I mentioned. I'm sure you can still dial in a good sound if you basically ignore the frequency labels completely and don't expect a 100 Hz boost to actually be a 100 Hz boost. Maybe I'm too far on the autism spectrum to understand such an idiosyncratic paradigms, so I just use my beloved Slick EQ GE (which btw has a tilt function inclued as a seperate control).

 

You really can't train your ear with a pultec-style EQ, cause it lies to you.
But once you do have your ears trained, you might as well label every knob with percentages and it won't change a bit.
Also, following that logic, you should consider every compressor ever made, VST or not, to be bullcrap, since their Attack and Release labels never correspond with their actual attack and release times.

 

Last edited: Jul 18, 2023

I'm not gonna defend pultec-style EQs here tho, I don't see any merit in "pultec trick", however complex the curve ends up being, because the same sonic result is easily achievable with a much simpler curve.
I just see a flaw in OP's reasonings. Pultec EQs are overrated for a different reason.

 

one big issue of "analog emulations" is the mouse

 

Care to elaborate? What is that different reason that I have failed to mention? No front, just curious!

 

I always thought main point of analog emulation plugins was to provide random results unrelated to parameter values,
(old) analog hardware is about usage simplicity, plugins are (should be) about accuracy and versatility - mimicking both at once is rarely superior if at all,
of course you can use any tools to achieve desired sound to your ears,
pultec EQ trick is just one of thousands of psychoacoustic things you can do, nothing more...

bEcAuSe iTs tHe iNdUsTrY sTaNdArD

 

Um, I mentioned it?
The sonic character of "pultec trick", the only thing these things exist for, is achievable with a way less complicated EQ curve, and you don't need any special plugin for it.

 

A big part of the problem I feel is that we have so much visual feedback in our tools these days, that people don't develop trust in their ears, and always want visual confirmation of what is happening. And without that visual confirmation, many people get uncomfortable and find it difficult to make decisions.

You also see a lot of younger people who've never used the older gear, or done any traditional audio engineering, also find it incredulous when *analog emulation plugin of classic studio gear* doesn't boost or cut at the same frequencies or gain markings around the knob (which actually works *against* their visual feedback).

So, useful as they no doubt are, I think there's a lot of value in EQs *without* that analyser/curve graph, and learning to make things sound good without worrying what they *look* like. And Pultecs are kinda nice for that, when you want a little bit of sweetening, not doing forensic audio fixes etc. But other EQs will do similar things, even if their interfaces might differ.

 

My argument is not that the labels are inaccurate and therefore its crap. My main argument is that the overall way of how the dials work in interaction with one another (eg boost and attenuation simultaneously) results in hardly predictable behaviour. The very inaccurate labels are just one aspect here.

With Pultec style EQ its very hard to realise a EQ curve that you have on your mind. Eg you listen to the track, then realise that it could use more bass, some dip at 300 and a broad boost of the upper mids: to achieve ecactly that with a Pultec requires much more guesswork and trial & error than with Slick EQ (which behaves as expected when you set it) or stock EQ / general purpose EQ.

Admittedly, with every EQ some degree of trial and error is involved, since you have to set it, correct and fine tune the settings. But if an EQ behaves oddly from the get go, how are you supposed to realise the EQ curve you had envisioned? Not to mention that Pultec is very limited in the EQ curves that can be achieved with it. And I'm not even talking about very unconventional or super narrow surgical EQ moves here, just broad strokes stuff...

 

Personally, I believe in 'blind' mixing. All EQs lie to some extent. What should a HPF's Hz reading be? The centre of the transition band? The beginning of the passband? The end of the stop band?

If you put your HPF at 147hz, can you tell with certainty at which frequency it'll stop attenuating? Does your OCD make you fiddle with filters until they land on a nice round number? What if the plugin gave you more decimal numbers? Trigger alert: 150.31Hz.

Not to defend the Pultec - because those interactions are crazy - but mixing by Hz numbers or to achieve visually pleasing curves is meh.
I don't love or hate the Pultec. It just does a thing and you'll get used to it after half an hour and it'll be as fast or slow to operate as any other EQ.

For myself, I've found that mixing 'blind' makes me make more EQ moves and IMO better EQ moves. Still need a surgical EQ from time to time of course.

 

Well, that is implied in my original argument already (did you fully read it?): it's limitations in the kind of curves it can produce (aka "Pultec Trick").

Not only can you easily achieve a Pultec-style curve with any generic EQ, but also better, more accurate Pultec-style (and other curves), where the dip is not way above the frequency range you would typically want to reduce, but exactly in that range (eg 250 - 500 Hz). I can dial exactly that in with Slick EQ in 5 seconds, while it's not possible with Pultec (EQ1P-A module) emulation at all.

 

Visually pleasing EQ curve is irrelevant to me. But a curve that resembles what I would expect after setting the parameters? Yes, please!

I get that even or even decimal numbers phenomena, but I often set HP to 37 Hz for example, so I can handle some oddness (must I screenshot my BPM settings? 83.07 bpm just sounded like the perfect tempo to me hahaha).

If anything I think of Hz in relation to 440 Hz as a reference. So, if I want to boost the 100 Hz area, I often end up boosting 110 Hz, since I believe that this locks in more accurately with the natural frequency distribution, if you think of it as a grid with every note being played and its overtones being somewhere on that grid. Some EQs allow you to set the position of an EQ node in semitones (musical scale), which is the exact same concept

 

you would think holding a button for slow scrolling for each increment would be standard after all these years in analog emulation plugins. I wonder why that's so hard to implement

 

Well, it turns out the odd EQ curves that the Pultec produces when boost and attenuation are used simultaneously (aka "Pultec Trick") are not really intended. In fact the original Pultec manual discourages using both at the same time!

https://www.thehistoryofrecording.com/Manuals/Pultec/Pultec_EQP-1A_Manual.pdf