The reasoning behind low passing at about 10-12KHz: Why not simply lower?

[Harry Hughes]

Some people are a fan of that classic super fizzy Pantera "Cowboys from Hell" or "...And Justice for All" guitar sound but it's not my thing at all.
Listening to a lot of my favorite productions, like Linkin Park's Hybrid Theory, Evanescence's Fallen, the last two Katatonia albums, every Paramore........well there's practically no harsh fizz anyway.
Unlike all that old fizzy as fuck 80s metal stuff, all the stuff that gives us the perception of the brightness in the tone seems to be really around the upper mids and a lot of the really trebly shit is not really there in any significant dose.
It just makes me wonder why we don't simply say "Fuck it, gonna low pass at 7-8KHz" and just get rid of some of that really gross fizz straight away so you've got more room for the cymbals and such.
So for the guys with more developed ears than me, what is the reason that it's often suggested to low pass at 10-12KHz?
Does some of that really high up stuff actually add anything to the tone?

 

[Morgan C]

Up to 12khz is useful audio information, and sounds good, past 12khz is just noise. If you were to record to tape you might not even need to lowpass at all. Fizz is not really the issue I think, that shit starts at like 2khz. Lowpassing Pantera or Metallica at 8khz isn't going to get rid of the fizz, it'll just make it sound dull.

 

[Fragle]

the reasoning of lp at 12khz is to get rid of some HF hiss that doesn't really add a lot to the guitar sound, but does a great job at masking the cymbals hehe.
it's got next to nothing to do with fizz....as morgan already pointed out the fizz can be found much lower, around 5-8k or something, so you're better off eq'ing that out instead of drastically lowpassing and killing all the overtones in the process.
hell, some guys (devin townsend) don't even use the LP at all.....actually it sometimes adds a nice air to the guitars if you leave it in.

 

[slashvanyoung]

I like me some air. I seldom low-pass guitars if it's not absolutely necessary. Could never imagine to go down to 7k, that just takes out way too much life.

 

[deLuther]

Low pass does not mean cutting of frequencies, -3 db at low pass frequency and another -6 db every octave above (for first order filter), not exactly "too much", I think.

 

[Melodeath]

I don't typically low-pass guitars either

 

[BLUElightCory]

You can eliminate most fizz with proper amp-tone and mic placement. It helps to dial in the amp while listening at mic/speaker level.

 

[mva801]

Low pass does not mean cutting of frequencies, -3 db at low pass frequency and another -6 db every octave above (for first order filter), not exactly "too much", I think.

Right. You don't have to use a 24 db/oct filter on everything you low pass. The important thing here is to USE YOUR EARS. Try lo passing with a high and low db/oct, and try at different frequencies. Use what sounds best. You should be listening to your source in the mix, not arbitrarily picking numbers to lo pass at on the internet.

 

[Harry Hughes]

No offense guys, I don't think any of you guys have really read my first post properly to be honest.
What I'm asking is, why in this modern time of guitar tones where the brightness is mostly concentrated around the upper mids, is the content of 8-9KHz and above really necessary at all?
That's what I'm asking, not asking "how do eliminate fizz?" or "how do low pass filter work?".

The stuff I listed in my first post (except Metallica and Pantera obviously) really seem to have virtually nothing happening in the guitar tones above about 8-8.5KHz.
Seriously, find a way to wider the stereo image of the tracks so all you can hear is the rhythm guitars, strap on a low pass filter, and sweep around.
I found that pretty much putting a low pass filter on 8.5KHz made no perceptible difference to the guitar
And I attribute this too, as I said before, the fact that modern tones are much more centered around the upper mids for brightness, whereas some of the older Metallica stuff and Pantera stuff was eating into the treble regions for brightness.
To me it sounds like a lot of the "air" type thing we perceive might actually be the treble based stuff, happening from perhaps about 5.5KHz to about 8.5KHz or so in the guitar tones.

It just seems to me, that apart from obviously good tracking in the first place, but in this modern day that when it gets sent to the guy mixing it, a guitar tone is made in mind with having it's absolute own region (which is obvious enough).
But that, from what I can hear, also seems to mean, if perhaps they aren't low passing, then they are certainly EQ-ing out a lot the high end muck that effectively seems to make the stuff above 9KHz seem fairly inaudible as to not compete with cymbals (which was a huge problem with Cowboys from Hell, not only is that tone just harsh and grating but it really clashed with the cymbals).
It seems that it makes sense to just low pass a bit lower than this seemingly standard way of going for 10-12KHz or just leaving the low pass out entirely and dealing with those frequencies above about 6.5/7KHz or so very effectively with EQ because I'm still not convinced that anything past about 8KHz in guitar tracks is really useful for a modern, full mix

Again, maybe I could be entirely wrong with this and maybe I just can't hear all the detail with my Behringer monitors and zero room treatment, but is there anyone that can confirm how close/how far away to the truth I could be that has vastly superior monitoring to myself ? (and that actually wants to read my first post a bit more carefully haha)

 

[nwright]

I agree with you for the most part. I was experimenting with some reamped tracks I had been working on and I threw my HP and LP on the group just as usual course of action. I started messing with the LP and brought it to 12K, and then 10K. I didn't really notice a difference, so I kept taking it a bit lower. It wasn't until around 8.5K that I noticed anything REALLY happening. Out of curiosity I put up a frequency analyzer to visually see if what my ears heard was correct. In my signal, it started tapering pretty severly right around where you are saying. There were 2 little spikes in a couple higher frequencies, and even with the LP set (using Waves Q10), those spikes didn't change much at 12K or 10K. Bringing it down to 9 effected them, but I think I liked it. It gets rid of the stuff I don't like and almost brings the guitars in a bit more in your face. I guess I could see it perhaps making the tone seem boxy or the stereo spread smaller, but I really dug it.

 

[ahjteam]

What I'm asking is, why in this modern time of guitar tones where the brightness is mostly concentrated around the upper mids, is the content of 8-9KHz and above really necessary at all?

It's most likely the same reason why you don't lowpass vocals at 2khz where the speech definition stops. Male voice definition is roughly between 500-1750hz (I've tried it, multiple times) and if you narrow it down anymore, it becomes really shitty sounding (kinda like muffled version of 1930's radiobroadcasts), as it's the first few harmonics above the attack of the highest notes (4khz *2 = 8khz, 4khz * 3 = 12khz) that make the guitar sound better, after that it's usually just useless garbage

 

[Morgan C]

No offense guys, I don't think any of you guys have really read my first post properly to be honest.
What I'm asking is, why in this modern time of guitar tones where the brightness is mostly concentrated around the upper mids, is the content of 8-9KHz and above really necessary at all?
That's what I'm asking, not asking "how do eliminate fizz?" or "how do low pass filter work?".

The stuff I listed in my first post (except Metallica and Pantera obviously) really seem to have virtually nothing happening in the guitar tones above about 8-8.5KHz.
Seriously, find a way to wider the stereo image of the tracks so all you can hear is the rhythm guitars, strap on a low pass filter, and sweep around.
I found that pretty much putting a low pass filter on 8.5KHz made no perceptible difference to the guitar
And I attribute this too, as I said before, the fact that modern tones are much more centered around the upper mids for brightness, whereas some of the older Metallica stuff and Pantera stuff was eating into the treble regions for brightness.
To me it sounds like a lot of the "air" type thing we perceive might actually be the treble based stuff, happening from perhaps about 5.5KHz to about 8.5KHz or so in the guitar tones.

It just seems to me, that apart from obviously good tracking in the first place, but in this modern day that when it gets sent to the guy mixing it, a guitar tone is made in mind with having it's absolute own region (which is obvious enough).
But that, from what I can hear, also seems to mean, if perhaps they aren't low passing, then they are certainly EQ-ing out a lot the high end muck that effectively seems to make the stuff above 9KHz seem fairly inaudible as to not compete with cymbals (which was a huge problem with Cowboys from Hell, not only is that tone just harsh and grating but it really clashed with the cymbals).
It seems that it makes sense to just low pass a bit lower than this seemingly standard way of going for 10-12KHz or just leaving the low pass out entirely and dealing with those frequencies above about 6.5/7KHz or so very effectively with EQ because I'm still not convinced that anything past about 8KHz in guitar tracks is really useful for a modern, full mix

Again, maybe I could be entirely wrong with this and maybe I just can't hear all the detail with my Behringer monitors and zero room treatment, but is there anyone that can confirm how close/how far away to the truth I could be that has vastly superior monitoring to myself ? (and that actually wants to read my first post a bit more carefully haha)

I don't know why you think above there is not necessary. I'm almost ALWAYS boosting with a highshelf at 8khz. I think it must be your system, but even 8khz is low enough that most systems should deal with it fine.

 

[Fragle]

well, if you have a problem with LP@12k then just do it differently. it's audio engineering, there are no set rules, do what sounds good. you've obviously set your mind on this one, now go with it.

having that said, to all the guys who can't hear the difference between no LP/LP@12k/LP@8k should probably start to listen closer or get better monitoring. EDIT: just noticed you said behringer monitors and no room treatment. that's your problem right there. i can assure you, there IS a difference. and i'm positive many guys here will agree.
i mean, there's obviously not gonna be a night and day difference, like no LP = Bright, LP@10k = dull or something. it also depends on the guitar tone (obviously), but i'd say that 99% of the times a LP around 11-14k makes some room for the cymbals without noticably reducing the air of the guitars. and lower than that and you're definitely cutting into the higher overtone range, and the sound goes downhill. no LP works for some tones/mixes, but usually there's less clutter and the guitars have more of their own defined space with a mild lp going on.

and to get even more into the subject, there's a huge thread over at gearslutz where people actually agreed that a LP is just as necessary as a HP in the digital domain. think about it, you're pretty much always hipassing stuff to get rid of the LF rumble and create more space down there, increasing definition blabla.
now, the basic statement of that thread was something along the lines of this: in the digital world, everything up to 22k (iirc) get's sampled accurately. when recording to tape, there's pretty much always some sort of reduced highend going on. that's why we identify tape recordings as warmer, and digital ones as sterile (IF you want to go down that road.....). hence, consequently lowpassing the super highs out of your tracks (coupled with saturation etc, but that's not the point here) sort of recreates the HF rolloff tape provides. and think about it, modern mastering usually will add a lot of top end so it holds up to modern commercial bright-as-fuck stuff. if you haven't LP'd out some of the ultra highs on the individual tracks, you'll end up with a LOT of high end which can get harsh quickly. LP that shit out in the mix, and you've got way more headroom to add that top end sheen without getting harsh so fast.
certainly worth a thought if you ask me.

oh, and btw, does anyone know the steepness of the hp/lp filter on the waves ssl e channel?

 

[gabriel g.]

well, if you have a problem with LP@12k then just do it differently. it's audio engineering, there are no set rules, do what sounds good. you've obviously set your mind on this one, now go with it.

having that said, to all the guys who can't hear the difference between no LP/LP@12k/LP@8k should probably start to listen closer or get better monitoring. EDIT: just noticed you said behringer monitors and no room treatment. that's your problem right there. i can assure you, there IS a difference. and i'm positive many guys here will agree.
i mean, there's obviously not gonna be a night and day difference, like no LP = Bright, LP@10k = dull or something. it also depends on the guitar tone (obviously), but i'd say that 99% of the times a LP around 11-14k makes some room for the cymbals without noticably reducing the air of the guitars. and lower than that and you're definitely cutting into the higher overtone range, and the sound goes downhill. no LP works for some tones/mixes, but usually there's less clutter and the guitars have more of their own defined space with a mild lp going on.

and to get even more into the subject, there's a huge thread over at gearslutz where people actually agreed that a LP is just as necessary as a HP in the digital domain. think about it, you're pretty much always hipassing stuff to get rid of the LF rumble and create more space down there, increasing definition blabla.
now, the basic statement of that thread was something along the lines of this: in the digital world, everything up to 22k (iirc) get's sampled accurately. when recording to tape, there's pretty much always some sort of reduced highend going on. that's why we identify tape recordings as warmer, and digital ones as sterile (IF you want to go down that road.....). hence, consequently lowpassing the super highs out of your tracks (coupled with saturation etc, but that's not the point here) sort of recreates the HF rolloff tape provides. and think about it, modern mastering usually will add a lot of top end so it holds up to modern commercial bright-as-fuck stuff. if you haven't LP'd out some of the ultra highs on the individual tracks, you'll end up with a LOT of high end which can get harsh quickly. LP that shit out in the mix, and you've got way more headroom to add that top end sheen without getting harsh so fast.
certainly worth a thought if you ask me.

oh, and btw, does anyone know the steepness of the hp/lp filter on the waves ssl e channel?

correct!!!

about the ssl channel. Not easy to answer, because it is not a linear eq+some harmonic content. The steps are very soft, as the other filter Q´s .

The v-eq works most of time perfect for lp and hp.
hp150hz lp12khz....very soft and "creamy".

As you mentioned it gives the sound that warm character and takes away the "steril" digital sound.

 

[Wolfeman]

oh, and btw, does anyone know the steepness of the hp/lp filter on the waves ssl e channel?

18db/octave

 

[gabriel g.]

18db/octave

sure?

 

[Fragle]

i'm asking because i did a shootout between all my eq's regarding HP/LP on the guitar bus, and the ssl won hands down. probably the modeled analogue stage...

 

[KillFrenzy]

Absolutely ALL guitar tones I deal sound lo-fi if I lo-pass under 10k (8k for say). Not to mention that a filter @ 10k touches much lower freqs. I usually lo-pass at about 12k or more to remove unnecessary garbage and then notch out any strange fizz carefully.
Not saying it's a bad idea to lo-pass that low. Nickelback's guitar tone sounds that way. It fits the production.

 

[Ermz]

I'm usually attenuating a ton of stuff around the 10k region, even when lowpassing at 11kHz.The only reason I don't LPF lower is because of some innate security that I'll lose 'air'. Honestly though I can't see much of a practical use for those frequencies at all in guitar sounds. IMO they are best left to cymbals and the air of vocal tracks to dominate. Most of my favourite guitar tones don't have much going on up there at all.

 

[Jackal_Strain]

I tried putting a lowpass at 8,5 kHz on the guitar bus today, and I really liked it. I'll have to experiment further...