Urban Legends: Pultec Kills High, and Fairchild Kills Low?

[Arys Chien]

Recently I mastered my friend's instrumental album. Not that I'm a good mastering engineer, but he couldn't find anyone better that he could afford and trust. :? He loves what I did though.

I used the following plug-ins, not all of them at the same time, just when I thought were appropriate:

UAD-1: Pultec Pro, Fairchild 670, and the Precision EQ, MBC and Limiter
PowerCore: Sony Oxford EQ and Inflator

That's why I'd like to bring out two things to discuss here, which I'd like to call \"UAD-1 Urban Legends\", since there has never been any conclusions drawn about them.

------

The first thing to discuss is that, does Pultec kills high?

From what I learned during the mastering job, yes. The moment that the mix passes through a Pultec (Pro) plug-in, there's an obvious roll-off above 16khz. More obvious to the ears than to the eyes if you're using a spectral analyzer.

That caused a dilemma. Some of my friend's mixes were already lack of the \"air\" above 16khz. Placing a Pultec Pro there just made things worse. On the other hand I just had to use the Pultec Pro to breath some life into his somewhat sterile mixes, due to not so hi-end analog hardwares.

In the end I just had to turn to the Oxford EQ to boost above 16khz. But you know, to EQ over EQed stuff is not a good idea, especially over the same frequency.

I couldn't replace the Pultec Pro with Precision EQ, since the latter is only smooth and transparent and could not \"color\" a sterile mix.

So, does it only bother me, or it bothers some of you too? I'd really like to know. Also I'd like a new Pultec Pro that could \"turn off the high roll-off\", if it does happen that way.

------

Now, the Fairchild 670.

People are claiming that the F670 is a \"sub stealer\". Earlier I thought it was subtle and didn't find it a problem. During this mastering job, however, I've found it more serious than I thought.

The moment a mix passes through the F670, the whole picture \"floats\". It's not as simple as \"stealing the low frequency\". The mix really \"floats\", like it's not standing on the ground, as other good mixes do.

That's another dilemma. The F670 \"opens up\" a mix even when it's not compressing. It's good for some of the narrow mixes my friend did. Yet for those already-too-less-low mixes, it's pretty bad, to a point that it's not acceptable. Boosting some low frequncy back with any other EQ plug-ins just didn't help, since it's more complicated than just \"stealing the sub\".

I think the F670 does something more than just changing some of the frequencies.

I see people using the F670 plug-in during their mastering jobs. I'd like to know that, is this making the mix floating thing ever bothers you? How do you fix that?

------

Any info will be appreciated. Thanks.

 

[Eric Dahlberg]

The Pultec high-end roll off is annoying, enough so that I personally don't ever use it for mastering. The Fairchild bugs me more because I don't understand it, not because of the low-end roll-off. Most mixes people send you will have too much bottom so it may even be a good thing in some cases.

 

[NuSkoolTone]

I have found the same things regarding the pultec and the fairchild.

I use the pultec to \"warm up\" something, and the fairchild to \"Tighten up\" things. I look at them as tools, and use them as colors to their advantage.
I also find the pultec seems to bring the mids forward a bit.


Since these are modeled after the originals, it's probably how the real ones work. I believe the fairchild was originally made for vinyl, and I think there is a limit on how much low end vinyl can take before it causes problems. Correct me if I'm wrong, but I think the fairchild was built with those limitations in mind.

Since the pultec and fairchild are resource hungry, perhaps it's a form of blessing because if there was a \"No high end roll-off\" switch for the pultec and a \"No bottom roll off\" switch in the Fairchild, I'd just might want to use it for EVERYTHING! :twisted:

I think the new NEVE plug-in is GONNA BE the killer that has the color, \"warmth\", and high end in tact that we're lusting after. Hopefully it will not require a card of it's own to run. Be nice if we can master with it for \"Flavor\" and keep the basic mix in tact. Here's hoping...

 

[A Gruesome Discovery]

I noticed the Fairchild's \"Bottom-end roll-off\" is directly proportional to the threshold, and that the different time constant modes affect it differently. I think the attenuation is really just the compressor compressing, but the controls are so bizarre that it's tough to figure out how to set it for a more \"transparent\" sound. I usually set the threshold pretty low and then adjust the bias to what sounds good; this seems pretty low-end friendly in recent experience.
The Pultec definitely does bite off a sharp chunk of highs somewhere in the neighborhood of 16k, though. It's never really bothered me, but like Eric, I avoid it for mastering, as well as a few other applications where I want some things to poke through there (drum overheads and such).

 

[bob humid]

Re: Urban Legends: Pultec Kills High, and Fairchild Kills Lo

From what I learned during the mastering job, yes. The moment that the mix passes through a Pultec (Pro) plug-in, there's an obvious roll-off above 16khz. More obvious to the ears than to the eyes if you're using a spectral analyzer.

hey! you've just discovered one of the secrets of analogue voodoo! most pleasant colorations are about unlinearities! you are now entering level 2

That caused a dilemma. Some of my friend's mixes were already lack of the "air" above 16khz. Placing a Pultec Pro there just made things worse. On the other hand I just had to use the Pultec Pro to breath some life into his somewhat sterile mixes, due to not so hi-end analog hardwares.

In the end I just had to turn to the Oxford EQ to boost above 16khz. But you know, to EQ over EQed stuff is not a good idea, especially over the same frequency.

ah. sorry. you are set back to level 1

reason: IF IT SOUNDS GOOD, IT IS GOOD. the UAD-1 has enough resoultion to cascade different EQs. go ahead.. a lot of people are piling different EQs on each other...

here is another NO COMPRMISE solution: buy yourself a copy of Algorithmix RED EQ or try the demo. use this one for THAT incredible high sheen. you´ll drop your jaw all day long

I see people using the F670 plug-in during their mastering jobs. I'd like to know that, is this making the mix floating thing ever bothers you? How do you fix that?

i don´t know exactly what you mean with "floating"... the fairchild has a strong effect of "tightening" the music. it works only good on slightly compressed mixes or better said on slightly "under-produced" mixes, its better on acoustical tracks. on these one it has a drastical beneficiant effect.. BUT: it does take away energy in the low-end. you can avoid that by taking out the bassrange of the mix out of the fairchild filtering. it also has a roll-off in the high-end since it comes from a time where vinyl was THE media. high frequencies are difficult to cut 2 vinyl (still aplies), hence the roll-off. you can get a good impression of what the FAIRCHILD does to the frequency range if you watch the grafic response of the Voxengo Curve EQ in the "FAIR" setting (GearMatch).. it simulates the coloration of a fairchild...not so bad too...

the following post is not my writing, but that of SoundSmiths Mastering, who posted this guide on the "unofficial" UAD-1 forums, so I'm sure he won't mind us quoting it here for all. It has been done quite some times...

 

[bob humid]

Re: Urban Legends: Pultec Kills High, and Fairchild Kills Lo

>>
Please be aware that the Fairchild 670 was widely (almost universally) used in mastering, expecially in the days of vinyl where the Lat/Vert mode allowed mastering engineers to control the factors affecting groove dimensions. So it was every bit a mastering device.
It is no less useful to us today for mastering digital media, but it is not the most intuitive device. A good deal of time is required to master the art of using this peice of gear. That is as true of the hardware version as of the software. The engineers who invested 30 grand in this thing did not expect it to run itself. They were tweakers to the max. They spent hours listening to the audible effects of changing parameters. Should we choose not to do so, we will get proportionately less usefulness from it.
For mastering a full mix with the 670, you will almost certainly need to use something other than the default settings or factory presets. You will probably want to start by backing off on the Bias setting to about 2-3 o'clock.
That will get you in the stadium. To find your row and seat, you'll probably want little (or no) Input Gain, and a low (or zero) Threshold. You'll probably want to use a Time Constant of something other than 1-3 (most likely one of the Program Dependent modes, 5 or 6, or occasionally 4, rarely 3). The ideal Time Constant setting will vary greatly with the above parameters and the source material. So it may be necessary to change this and go back to square one and start again if you don't like where you're heading.
You'll probably be most happy on a full mix when you see little or no activiy on the Gain Reduction meters, even in loud passages. That may be counter-intuitive, but it works. You are trying to acheive a smoothing and sweetening effect, not a brute force taming of dynamics. Today, we have very good brickwall limiters to use after the 670, such as the L2, etc. Use them.
Much experimentation is needed to familiarize yourself with the wide variety of possible outcomes. For example, did you know that tweaking the \"Balance\" screw has a measurable and audible affect on the sound? Originally this would tweak the voltages across the 6386 tubes, which does not produce a straight-forward effect on stereo level when the two screws are set slightly differently.
Rather it causes low-level dynamic phase variations that can be quite euphonic. I find that on mono signals and some stereo mixes, it is best to have these two screws set exactly the same (with fine increments of the keyboard), preferably so that the little paint dot is at the 12 o'clock position. But for many mixes, I have preferred the sound of one screw set fully clockwise and the other fully counter-clockwise, or somewhere in between. Try it and see what you hear.
Again, I say that this is a complex and amazing piece of gear, and it will be some time before the user gains enough familiarity with it to know instinctively where to go to get the desired result. But for those of you who have not had much time to spend with the 670 on a full mix, I urge you to do so. I warn you though, that you may end up like me. I agree that for most material, it doesn't sound right any more until it's been through the Fairchild! And I've never had ANY piece of equipment, let alone software, about which I could say that!

This should get you up & running nicely with the Fairchild. If anyone else has tips from the days when they used to use the real thing, then please post them here. It's a complex beast, but well worth getting to know.
Also, try this on guitars for that Tom Petty sound:

Time Constant - 1
Input Gain - 0 to 4 (try 0)
Threshold - 0 to 4 (try 0)
Factory calibration - around 2:00 (turn anticlockwise unti the highs become crisp again)

The harder you hit them, the better these plugs sound too. If you are after the mastering \"sheen\" then use the method above. If on a solo track, then hit it hard.<<

 

[geekeye]

Re: Urban Legends: Pultec Kills High, and Fairchild Kills Lo

the UAD-1 has enough resoultion to cascade different EQs. go ahead.. a lot of people are piling different EQs on each other...

"resolution" has nothing to do with it. the problem with using sequential eqs on the same frequency band is phase distortion. if you don't understand that, back to level 0 for you (sorry...could not resist).

 

[Dan Duskin]

The high-end roll off on the pultec is so slight i really don't worry about it. Also, the high-end roll off lessens with higher sample-rates. The roll-off is directly related to the sample-rate conversion. But again, it's so slight that most users never hear it... and it's so high (above 16kHz). And because the roll-off is caused by sample-rate conversion it is really only warming the ditital fizz (high-frequency square waves) normally associated with 44.1kHz digital recordings.

The low end roll off on the Fairchild doesn't bother me either... because in mastering I end up rolling off something down there anyway. And again, it's so slight most users never hear it.

 

[tiger vomitt]

most of the time im using the fairchild on a parallel bus anyway. it sort of negates the LF rolloff. and i agree that most of that is just the compressor working anyway.

btw the GateComp compressor is my new favorite, esp. if im blending on a parallel bus like that. it pumps and squishes even better than the fancy UAD compressors. at least that's how im hearing it lately, these things have a tendency to come and go in my umm practice.

 

[bob humid]

Re: Urban Legends: Pultec Kills High, and Fairchild Kills Lo

the UAD-1 has enough resoultion to cascade different EQs. go ahead.. a lot of people are piling different EQs on each other...

"resolution" has nothing to do with it. the problem with using sequential eqs on the same frequency band is phase distortion. if you don't understand that, back to level 0 for you (sorry...could not resist).

yeah. yeah. give it to me. i´ll survive that... but I didn´t want to sound arrogant. just trying to help

but maybe I was not quite precise with the term "resolution"

for example:.. the RED EQ from algorithmix, set to XTRA settngs, analyzes about 1 full second of audio material.., forwards and backwards.. the two signals are then compared with each other and the differences are compensated algorithmically. .. you can pile a linear phase EQ on top of a phase distortion classical EQ without problems - just one of the examples .. if you don´t understand that ..

it´s usally a totally normal procedure to set a pultec for this cool transformator sound it makes and boost some treble while fine-tuning the high sheen above 15KHz wiht a more minimal phase EQ.

so ok. I´ll try it again:

1)
IF IT SOUNDS GOOD - IT IS GOOD (joe meek)

NO, REALLY!

2)
YOU CAN PILE HIGH-QUALITY EQs ON EACH OTHER - GO AHEAD but don´t overdo it, otherwise you´ll get a post from geekeye!

robert

 

[ByronSanto]

Algorithmix RED EQ

for example:.. the RED EQ from algorithmix, set to XTRA settngs, analyzes about 1 full second of audio material.., forwards and backwards.. the two signals are then compared with each other and the differences are compensated algorithmically
robert

Robert

How did you find out about this feature?

 

[geekeye]

Re: Urban Legends: Pultec Kills High, and Fairchild Kills Lo

so ok. I´ll try it again:

2)
YOU CAN PILE HIGH-QUALITY EQs ON EACH OTHER - GO AHEAD but don´t overdo it, otherwise you´ll get a post from geekeye!

robert

once again, quality is a subjective term, as is over doing it, and i must protest the recommendation that people to dive into dangerous waters without understanding what they are really doing.

i myself do sometimes use eqs in sequence, but the important thing to note is not to avoid "over doing it", but to avoid using sequential eqs on the same frequency band. using one eq to add colour at high frequencies, and another for a surgical shaping of the lows is a perfectly acceptable practice that can yield excellent results.

but using multiple non-phase linear eqs on the same frequency band (or even nearby bands) will make your music sound like crap, guaranteed. even if you boosted with one eq, and cut with the next, so that they "cancelled" each other, the frequency spectrum might come out looking identical to the input, but the frequencies in that band will be smeared over time, and the result will sound flat and unfocussed.

and i don't see how me replying to anyone's post is a bad thing to be avoided. i'm just trying to help, as most everyone here is.

 

[Arys Chien]

I know people are piling up eqs, but:

- some of them don't know what they're doing
- some of them have really high quality analog eqs
- some of them are mixing, not mastering
- some of them are doing mastering for music that sounds better when distorted

My friend's album, though not well recorded and mixed, is closer to \"natural\" music, i.e. when he uses a snare he expects the listeners seeing a snare in the middle of the speakers.

Then piling eqs can be a problem sometimes. Even to my level 1 ears. :wink:

Even though, I still tried what and could when I had to pile up eqs. So I'm not saying that we can't use two or more eqs in a row when mastering. I'm just saying that it's \"a trouble that can be avoided\".

------

I don't really care what the gears sound like in their real life form. I'm not using these hardwares anyway. Even if I have two hardware Pultec EQs, I still have to learn the different character of the two to use them wisely. So why sticking to the 100% reality becomes so important? Why not giving us an extra switch to make our good UAD-1 life better?

------

The F670 makes things \"float\". That's what I hear all the time. Maybe it's subtle, but then again mastering is subtle too.

This side effect is good on those mixes with too much low end, but really bad for those with too less already. And really, adding some low end before the F670 just doesn't help. I don't know why. And that's why I think the F670 plug-in does more than simply \"stealing the sub\".

As for the F670 plug-in opening up the mix a little bit, now it really sounds a bit close to a BBE Sonic Maximizer plug-in, which opens up a mix too. The BBE plug-in has a \"lo end make-up\" knob. Don't you find it interesting?

 

[Eurocide]

I know people are piling up eqs, but:

- some of them don't know what they're doing
- some of them have really high quality analog eqs
- some of them are mixing, not mastering
- some of them are doing mastering for music that sounds better when distorted

That's the reason why I started the "what's your music style" thread. This could clear up a lot of misunderstandings here.

 

[Dan Duskin]

Piling EQ's is bad, unless one of them is linear-phase, and/or unless the two EQ's aren't adjusting anything within the same two (or 3) octaves.

But far more inportant than this is to get it right in the first place! How? Move the microphone around until you get the sound you want (instead of EQing it to death later)... and once you find the right spot to put the mic, try different mics to get that sounding right... and once you get the right spot and the right mic, choose the right polar pattern... then once you get the right spot, the right mic and the right polar pattern, angle the mic where it sounds best... then once you get the right spot, the right mic, the right polar pattern, and have angled the mic where it sounds best, choose the preamp that sounds best for the task... then once you get the right spot, the right mic, the right polar pattern, angle the mic where it sounds best, and have choosen the preamp that sounds best for the task, get the settings right on the preamp. Then and ONLY then can you start to think about EQ. But guess what! Once you do all that right, 8 out of 10 times you won't need ANY EQ!!! And if you can get the instument right (prior to the mic, and still following the above steps), 10 times out of 10 you won't need any EQ

Now, if you still want some EQ... use the highest quality EQ you can... do it once (never twice), and cut before boost. But like I said above, you won't need to EQ if you get all that above stuff right.

Boosting frequencies always sounds \"fake\". It never sounds like it came from the instrument that way. This is why you have to get it right to begin with... then cut, and never boost unless absolutely needed.

 

[Ken Rash]

I'm with you until the advice about cutting EQ instead of boosting.

Absolutely necessary?

Some EQs, the Pultec for example, sound great and take on an attitude you won't get otherwise when using to boost frequencies.

 

[Dan Duskin]

I'm with you until the advice about cutting EQ instead of boosting.

Absolutely necessary?

Some EQs, the Pultec for example, sound great and take on an attitude you won't get otherwise when using to boost frequencies.

The point is that if you can get that boost out of the instrument to begin with it will sound better than the best EQ ever made.

And in regards to cutting before boosting... it is all too common for beginers to only boost frequencies and never cut them. I very rarely boost anything. For example, if it sounds dull... it's more likely that applying a high-pass filter or cutting some low-mids will sound better than boosting the highs. It's hard for most to grasp this, because when you boost things get louder, so you instantly think it just got better.

 

[jr213]

The Pultec does not have the HF rolloff problem at higher sample rates. I mastered something at 88.2 (upsampled from 44.1) and it sounded great. Then, just for fun, I processed the original 44.1 files with the exact same settings and it sounded like someone put a blanket over everything. The Pultec is absolutely gorgeous when freed from its 44.1 handicap. I really wish they'd fix that! Such a waste!

 

[Ericcc]

The Pultec does not have the HF rolloff problem at higher sample rates. I mastered something at 88.2 (upsampled from 44.1) and it sounded great. Then, just for fun, I processed the original 44.1 files with the exact same settings and it sounded like someone put a blanket over everything. The Pultec is absolutely gorgeous when freed from its 44.1 handicap. I really wish they'd fix that! Such a waste!

100 % agree on this !!!!!

 

[Eric Dahlberg]

I know people are piling up eqs, but:

- some of them have really high quality analog eqs

Boosting with a Massive Passive can sound incredible. For me, it's not worth even trying with digital EQ's, even for the best ones, because there's just too much opportunity to mess something up. Why take the risk?