Skip to main content

Hello all, I'm a mastering engineer in the NYC area and have been in self debate for many years whether to roll off below 20hz or not to. My personal experience has been that when I do roll off, the low end breaths and almost jumps out which makes sense because the amps aren't working to push out these sub freq's but I do read about others that believe that to do so is a sin and one should let the low end extend further down. Any views?

Comments

KurtFoster Fri, 09/13/2002 - 09:24

Welcome to RO,
I usually roll off stuff between 50 and 70 hz just to tighten up the low end, it seems to reduce the flab in the mix. But I wonder if there is some kind of psycho acoustic effect to this? I know the human ear can hear out to 20K hz but studies have shown that we can still sense frequencies even higher than this. Hence the move to 96 k and higher sample rates for digital. I wonder if this is the same in the low end also...... Fats

Michael Fossenkemper Fri, 09/13/2002 - 10:03

There is definately things going on down there but whether or not to keep it is the question. Even the modest of setups reveal a change below 20hz even if the speakers themselfs can't reproduce it, the amps and the wiring are. I have several monitors that I know very well but my main ones are the B&W N802's. These guys go way down and have no problem handling these sub freq's but I find that it's the more modest setups that are affected the most ie. a home stereo setup. I think that lopping off 50 or 70hz is too excessive in that even the cheapest systems have no problem delivering this. but when you get down below 20hz give or take a few hz, there is a definate benifit to my ears to the smaller systems.

knightfly Fri, 09/13/2002 - 11:09

Psycho-acoustics is an interesting field - I still remember a discussion I had with an old movie projectionist when I was still in High School - he showed me an optical sound track on an old dracula movie, and asked if I knew what those long-period waves were ('way too long to hear) I of course said no (I was less of a smart-ass then) so he explained it - claimed that even though we can't "hear" lower than about 20 hZ, we feel it thru bone induction, body vibes, etc, and it has a definite effect. He showed me the long-period "sounds" on that dracula flick, and said they were put in there to "pre-scare" the audience just before dracula appeared, to heighten the fear factor. Since the theater we were in had Bozak 30" Electrostatics, I assumed they could reproduce that freq somewhat...

I later read a blurb somewhere about an experiment the Russians had tried, where they built a sub-sonic generator (megawatts) to use to "pre-scare" enemy troops - the article mentioned that the experiment had to be dropped - seems that unless they remote controlled it, the generator had a tendency to turn the operator to jelly (as in DEAD) If I remember correctly, the freq was around 7 hZ.

Still, a LITTLE BIT of that might really increase the "Power factor" of some mixes - are we ready for yet another surround format, say 9.2.1 ???

I couldn't see incorporating that into normal mixes though, not too many home systems are quad-amped... Steve

audiokid Fri, 09/13/2002 - 11:11

Welcome Michael

Hi, after many years in writing and playing I'm finding mastering to be the most rewarding and definitely the next step in my career. My first comment is, without having the basics down it sure can be frustrating. I wish I could sit in with a few of you guys one day. This thread may solve a long time mystery to me.

If you roll off these subs does it not give you a higher SN when burning to disc? What's going on there? Is it a trade off and is that important?

I sense subs and when doing an A/B it drives me nuts as what to keep in. It seems to effect not only lows but the sub harmonics all the way up. Am I right?

I think knightfly has touched on something. Our hearts feel and are effected by freq way before our conscious mind does. This being said I would have to aggree that if we can keep it all in (20 t0 20 plus) then we are closer to alive or death. :eek:

If our media however, can't grasp it "flat" then this is where we are forced to trade off. Am I close?

Michael Fossenkemper Fri, 09/13/2002 - 12:10

Yes a trade off, but what are the costs on both sides. Am I saving people by not having 7hz or am I taking away some kind of sub-conscious effect of the material. I think that by lopping off these subs that you are mostly hearing the effect it's having on the amps and of coarse the drivers not having to work as hard. I'm not sure to what extent these low freq's are interacting with other freq's before they hit the amp stage. I remember engineering and mixing on LL cool J's "mama said knock you out" album and I was subjected to 808's from hell all day every day. I of coarse wore ear plugs most of the time cause the volume was at 11 but by the end of the day I felt like vomiting from all these low freq's. I actually sensed it changing my heart beat. Maybe I was just an 808 away from being jellied.

knightfly Fri, 09/13/2002 - 13:48

I know the feeling - I went to an Earth Wind and Fire concert in Portland Or last monday, and overall the sound was pretty good for live (converted theater) but the kick drum was from absolute hell. After the first 15 seconds I had my plugs in, then I spent half the nite trying to figure out if it was a young stupid subwoofer addict on FOH, trying to run 2 ohm loads across a bad Sub amp, no pillow, or the riser-on-a-riser drumkit with no dampening inside the risers, or D. all the above ??? Anyway, picture the absolute worst abuse of car subs on the planet, and you got what the bottom sounded like. When they did stuff minus drums in between songs, it was great. Or is this sound just what the pubic (sic) now expects? Gawd, I hope not...

audiowkstation Sat, 09/14/2002 - 07:55

Depends on the program material (other things going on at the same time). As a general rule, I let everything down to 8hZ pass. I have been known to boost 14 hz even though there was no fundamentals there..just sub harmonics. Whatever I do, it must pass the cheap boom box test. As long as the longest waves do not intefere with the vibe and they do not mees up the sound of lesser system (compromising volume etc)..I let the wool hang out.

audiowkstation Sat, 09/14/2002 - 11:44

8.175 hZ is the frequency of the note ''''''C, Produced by the 64 Foot Pipes of the Diaphone/Dulzian, Contra Bourdon, and the Ophicleide pipes. With as much air motion as you experience with 64 pipes and 32 foot pipes, you can clearly hear the difference of the 8 and 16 hz notes(8.175 and 16.35) respectively and know they are the same note..only octaves. Since music can be used down there...I go for it..when appropriate. I did a rap song that had a pedal 12HZ for 2 measures.

Some of the finest sounding Organs on the Planet are built by the Nichols & Simpson firm. Hand made. Since "C" (reg pedal C) ''''C 32.7 hz note is easly reproduced and appears on thousands of recordings, it is only fitting to have at least one octave below ..if not two. Reproduction of below 7HZ is utterly useless IMHO. My monitor system is 6dB down at 13hZ so there is plenty of info down there. Proper loudspeakers give no hint whatsoever that they can reach with total authority those fundamentals until they are present in the recording. One thing I cannot tollarate is a boomy subwoofer system. Totally unmusical..but at high level when all the bass is crystal clear and a few 16HZ or lower frequencies (sub Grave') are in the recording..the floor simply moves, as does your shirtsleeves and collarbones..with none of the roar or boom that is apparent in the so called "Boom systems". Hope this sheads some light. [[url=http://[/URL]="http://www.nicholsa…"]Fantastic instruments hand built and tuned.[/]="http://www.nicholsa…"]Fantastic instruments hand built and tuned.[/]

Michael Fossenkemper Sat, 09/14/2002 - 13:02

Bill, thanks for the info. It did shed light on the matter. I don't have the priviledge to work on many records like this so it does build a case to extend further in these situations. Most of my experience for rolling off higher is with pop music expecially when a turntable was involved in the material. But it does give me food for thought on how these sub-consiously might impact the listener on other material. Thanks

audiowkstation Sat, 09/14/2002 - 19:44

Ah yes. When mastering for RIAA equailized vinyl, the rule of thumb is the 32 hZ barrier on program material...soft shelf of -12 at 20HZ and a full 18 to 24 dB/octave (36db works too) below 20 hZ. Their is enough trash from 4hZ to 20 hZ generated by the vinyl itself that vinyl playback must incorporate at the very lowest a 16HZ cutoff as well. Essentially the digital medium does afford us flat response to 2HZ (read two) thus my rationale for the 8HZ limit. But remember this: We are artist..and perhaps for artists sake, their are few limits or rules. Guildlines can be appreciated and sharing of works welcomed for any achieved advancement, any of us can gain.

My Mastering is unique..as I subscribe to whatever give maximum flexiblity to the artist wishes and maximum vibe for all systems. As a mastering engineer, Art is prevelent to science.

I would be really keen to share works with you. Low end (fq) balance is always a thumbprint of our translation and the closer both of us get to each other, the better the situations are for the end user. I agree that mastering (the ones that are getting the big deals) has reached an all time low quality standard. Their are some horrable works out there that is inexcuseable at best. Major dissapointment was the "wingspan" album. The 20th anniversary Diana Ross is another huge foul-up. I hear some artistic mastering in the New Stones releases (DSD/SACD) that could have benefited from better monitors, more skills behind the platforms. Just some real oversights. What happened to going back to the way it was supposed to sound? Too young, too much of a rush I suppose..to get them out. Does the word "shrill" not register with these folks? Probably audiophile monitors (with the huge cut a 3K and heavy damped rooms) Some super highly regarded loudspeakers simply do not have linear dynamics at 3K and the engineers compensate for room absorption and high volume and the end result is a 9dB rise at 3K. Pitiful. These folks do not have ears...or brains. Loudness countour does not function at 105dB!..or is it needed!!! How many mastering folks have studyied Linear dynamics and how it partains to Flecture-Munson? Their is an average that must be adheared to. We should all strive to learn from each other as no one is the absolute. Art for Arts sake is still the vibe here..Shrill mastering must go south.

Another 25 hrs of editing...apologize in advance for any typos..

Michael Fossenkemper Sat, 09/14/2002 - 23:26

Your right, there are no absolutes, not even with myself can I say that from year to year that i don't change my opinion about something. I try to keep an open mind about new ideas and new approaches while also trying not to loose sight of what i'm doing. I hope this forum will give insight and new ideas to all of us. I have been blessed in working with some great artists and not so great artists. Each contributes art in their own interpretation and it benifits us all and is appreciated by someone somewhere. So lets keep the reels rolling and maybe in the process we can raise the bar a few notches.

audiowkstation Sun, 09/22/2002 - 02:50

Hi Mantik! I really do not think the subject of dynamic lineararity has been touched on in the mastering field as it should be. I also want to say, their are some fine works out there that take the subject seriously. A major part of my loudspeaker studies have involved this and it is a kettle of fish that is still under study. Some reading you may find interesting can be studied in Journals for the Acoustical society. The many acoustical societies all over the globe. Your university library should have these.

Let me provide an example of dynamic linearity.

A given loudspeaker is fed by 3 signals at separate times. The signals are for this example:

80 hZ, 1200 hZ and 8K hZ

First measurement is at 1 meter Second measurement is at 2 meters and the third at 5 meters. If all 3 measurements at each distance are within a tight constant then this loudspeaker/room/amplifier/source setup exibits linear dissipation. Now we take a loudspeaker and drive it with the same three signals at the three distances above but this time their are 63 measurements. the first set of 9 are at 60 dB at the 1 meter reference, 2nd 9 at 70dB, 3rd 9 at 80 dB , 4th at 90dB, 5th at 100 dB, 6th at 110 dB and 7th at 120 dB (7th only if the "system" can achieve the 120 dB point at all frequencies at 1 meter). The compression artifacts exibited in the dynamics of the loudspeaker pertaining to the deviation in drop at the given frequencies at the given sound pressure levels and the given distance is your measure of effeceincy lineararity. Many loudspeakers and rooms fail miserably at this seriously compressing out of uniformanty and boosting at other unwanted frequencies and distances. The room is the interplay as well. When these figures measure closer to the ideal, instruments sound like instruments and music translates so much better.

Believe it or not, the simple full range driver do have some advantages here but many of them fail miserably at the extremes. I have personally been involved with this study now over 23 years but very few loudspeaker manufactures have actually picked up on it. How does dynmaic linearity sound? One thing, it is uncanny. The speakers all but dissapear in the room. Everything "fits" like a glove and since the compression (that some mastering engineers make up for..that is caused by this inaccuracy) is not happening, the mastering does not have the shrill bite and the bottom is always there. No need to really crank that bass control up in the car anymore. Unfortunantly, their is a fine line to how we master. The music still has to sound acceptible on non linear dynamic systems. Actually it ends up sounding better because problem "A" has been addressed.

Major points are, room interaction, loudspeaker inaccuracies.

Again, above... the lab rabbit approach to loudspeakers and rooms is not always the best route to go, but some things should remain constant for the art to be heard. It is amazing when I do marathon mastering on my system without using any other references, then after 18 songs are done, I go to the theater room and lay on the couch and make notes. Last session I only had two tunes that needed adjustment. Now for the car stereo today........

Calibration standards are a subject that is touchy. I share with other mastering engineers the direction I take works and they share theirs. This way we can all see where we are coming from and get a feel for each others rooms. Afteall, if you give a tune to 5 different mastering engineers, you will have one that is clearly the better of the lot. I feel all mastering engineers can use this data to achive more uniformity in the science so that their art can shine through so all 5 would be killer and no one job would be "better"..just different, each in its most enjoyable form.

One mention: If a mastering engineer is getting progressivly better results and the works are very acceptable under a broad range of circumstances, it would be heresy to tear down their room and start from scratch. Constant calibration is the key. I actually found a well respected mastering suite with the channel balance off 3dB on the main power amp due to someone not paying attention to the input gain controls. No telling how many hours of work were waisted before I discovered this. It was amusing to see the bean counters walking around scratching their head.

I wish your frind well on the journey through the archives of unrealeases.

crazy_guitar Mon, 09/23/2002 - 13:48

Perhaps that's why so many people like the vinyl sound. Cause there isn't much happening way down there.
The other day I mixed a song without the bass guitar. I had the bass on another track. I did some mastering to the mixdown without the bass, and cut off everything below 20hz, then I mixed in the bass unaffected on the 20hz area (but I did some EQing on the low mids. It was interesting.

-Joz

Don Grossinger Wed, 09/25/2002 - 04:59

When I cut vinyl on our DMM system, I do a digital roll off at 12 Hz. It has been proven to my ears that there is important stuff going down that deep. This is especially true in Dance & Hip Hop. Club systems definately can reproduce these frequencies.
As always, serve the music. Make it sound right. If the music feels better with the low freq. stuff in, leave it in. Try it both ways & make an informed decision. Keep in mind the ultimate destination & end user & do the right thing.

Alécio Costa Thu, 09/26/2002 - 11:32

wowo! I have been after this info and had previously debated HPF on another forums!!!great to know and read these pro guys!

I am a recording engineer/musical producer who also does budget mastering.

I have to tell you that I have been HPF around 32, 37 and sometimes up to 42hz and the results are quite good. The stuff plays much better on my $40 reference test stereo system!

However, I had just finished a root samba album and I felt that HPF around 39hz was a litle bit much. It is oka, but the subbass could be a little bit more in there, however the Cd is making some quite success and receiving nice reviews. But I think this is related with the boominess of my room around 135hz , too damped and not very Pro moniotrs, MONITOR ONES from Alesis, but that I had been able to mix to and that had made the most recent jobs translate quite well on the radio and in other systems. I had sent some stuff of mine, regarding even my own Cd ( soon to be released) and the guys in The USa had liked it a lot but also recommending me a little HPF under 40hz.
Thanks a lot!
keep up the great forum!
:)
studiodp@terra.com.br
:w:

Michael Fossenkemper Thu, 09/26/2002 - 15:09

After talking about and reading replies to this question, I've been really paying attention to what is down there and I'm spending much more time playing with it. I have to say that, depending on the material, there is a definate benifit in manipulating these subs but keeping them in there. Until the last 3 or 4 years, I never thought much about what was down there because frankly I had nothing that could reproduce it in a way that made much sense. Keep tweeking, listening, and learning....

audiowkstation Thu, 09/26/2002 - 15:42

Michael, you are a really nice breath of fresh air to the RO community. Thank you ever so much for taking the time to post and moderate here. Your sharing is extremely valuable to all readers.

It is a most difficult task to faithfully reproduce anything below 35hZ....for loudspeakers that is... Thank goodness we each have monitoring systems that can really do an honest sub 30hZ.

Probably (there is that word again)...well definitly, it is power band response that determines how much we can "get away with" on the sub lows. We have to take into consideration the idiot..that turns their bass control all the way up for everything. It took a painful amount of time to convince my nephew that cranking the bass and pushing the loudness button did not do anything for the sub lows other than cover them up. I showed him an experiment (on his system) where he could get much more output below 40hZ by turning the bass control all the way down! His control was centered at 100hZ so turning it down removed most of the bass except for below 45hZ. How long do we have to compensate for the folks that enjoy full bass boost? I personally include accurate bass with controls set to flat..so if they turn them up..it does get ugly..but to a point. I still compensate for the low end issue of consumerism. I also find many consumer (even high end) speakers have a tilt at 50hZ. A good 5 dB bump..because of the room. In an anechoic chamber the speaker may have a smooth response curve below 100hZ but in a room? Well you know what I am taking about.

In 1972 I started really buiding speaker systems. I collected as many parts as I could from old jukeboxes (yes the 50's boxes)McGee radio in Kansas city had the finest selection of OEM over-run drivers on the planet. You could buy an emenence 15 and a compression horn with crossover on a baffle board for 17.95 The ad stated "Loudspeaker system cheaper than a pizza and a beer" They kicked ass in a 4 cu ft box . I took several (when I was 15 and could legally drive) systems to the local hi fi shop and they shook their head in amazement. I met Paul Klipsch in 75 and we remained buddys and visitors (and friends) up until he died this year at 98 years old. In 1975 I built a pair of Klipschorns and powered them with a McIntosh 2105 in my bedroom! I used particle board for the bass horns and bought speakerlabs mid horn and university (Klipsch K55V) driver, t-350 electrovoice tweeter and custom designed my crossovers. I loved building spekers and I still do it. I mainly do consulting work for manufactures in the loudspeaker arena. I actually set out to build the worlds most accurate loudspeakers. It was a 4 piece system with outboard crossover in a 19" rack case weighing 77 LBS (the x over). They were 17 to 19K devices +/- 0.7dB. They weighed 653 pounds per channel and required 6 identical mono amps. The cost associated with the first pair was over 300,000 dollars, moist of this involved in using a military base FFT and anechoic test chamber and cray supercomputer. Rejection rate on mil spec resistors was 300+ to 1. Coils hand wound, wire was soildered using exotic solder and timed presicion soldering. Needless to say, those monsters were the best I have ever heard..but damn..no one including myself could afford them so I scrapped them and paid of the investors. We came away very clean. The drivers were insane. The enclosures are so different that I keep it confidential..hoping some loudspeaker manufacture would one day decide to buy out the novel design of them. Put it this way, each driver had the ability to operate in an encl;osure that would vary its dimentions instantly depending on frequency. This much I will say...x overs were 90hZ and 900hZ.

Well excuse me for getting off topic..

As far as mastering is concerned....

Bass translation is a sore point with me. Mastering engineers have it "all over the place" as of late. I hope to saee you on line and slide a track I just finished that I am quite happy with all around. That way...you can give me your opinion on where my bottom lies. The track I sent you before was totally exagerated on purpose to show that there is music down there..but really, looking at the power curve..even the lowest of bottom notes that are sustained cannot be more than -3dB down from normal level of impact sounds such as the kick. I inserted a 16.7hZ tone at -31dB on a track I was mastering for effect (artist approved) and it proved to be a bit much as it pushed up other areas of the bottom a shade. I settled on -34dB!

Lets discuss this some.. Okay!!

Again the standard disclaimer...I am not a typest and typos are the norm..so if you don't like it or critisize it...got hang your hat in the storm!

TKX from Bill

Alécio Costa Thu, 09/26/2002 - 16:38

Bill, I got rid of my NS10M´s very recently because they do not represent the consumer audio of nowadays.

Maybe the Ns10´s did go oka for the 70´s, 80´s and early 90´s. Today, people seem to love super/mega / x/shit bass on their small systems. They sound very artificial bumped !

Ns10´s do not have lows, too much mids ( specially between 1k/3k) and so...
Never understood why people still keep them...
What are the folks opinions on this? Oka.. I knwow that hundreds of Ns10´s are all around the world with top names....

I think that Alesis Monitor Ones or Mackie´s HR624 can be much closer.

Awful when folks push those silly Eq presets: rock, clasical, Pop, Dance...

:)
:w:

audiowkstation Thu, 09/26/2002 - 18:37

Email me, I want to hear your tracks..or go over to my forum Streming and post one or many of them.

The inverse is another reason to use NS10's. Your monitor ones have boomy bass. IF it sounds good on them..because they are boomy, your bass wil be thin..the inverse of what you want. Mid heavy (IYO) means a cut in mids...perfectly acceptable for the sound you want. Make it sound the way you want it on NS10's and watch the woofers..for ballistics and it will work. My Dossier on loudspeakers dictate that faults in the speakers will mean the reverse on the mix. It is that simple!!!

But remember, I am on your side..what works...works..and the 40 dollar system is important to the translation..a 10 to 20 dollar clock radio is cool as well

Alécio Costa Sun, 09/29/2002 - 17:24

thanks Bill! which email address shall i sent you? I also have some recently finished mixed/mastering of a Cd album I started producing in 2001.
They might sound a little weir to you at the first listen, but it is root samba.
Some friends on the forum also evaluated some classical stuff and my own cd. Ronny M, Brad B., Joe S and John St. were really great guys commenting the stuff.
I generally do 2 mp3 mix files with a piece of each song so clients as well as me can send for proper evaluation and the people on the other side already have a quick but general image of the album, without consuming much precious time of theirs neither disk space/ download time.
Genereally I do 2 files of 4 minutes/ 4MB.
Is it oka my friend?
today I finished the new version of a song of mine sang in english, from 1992, great stuff.
Thanks in advance
Nice weeK!

Alécio Costa Sat, 10/05/2002 - 21:24

great Bill! In fact there are 2 mix files of my own pop/funk/sould CD of 4MB

2 mix files of Camerata Orchestra Vol.4 also around 4MB each

And samba mix files with the same size too. It is possible to send it via ICQ to you without disturbing your job? MSN/hotmail are too small!

Besides, last night was the first presentation of one of the CD songs on a Cable TV show. Very funny because the song is called 5 Against 1. I think you guys shall now what it means...
It is a tune whose orchestra hits remember Billie Jean.

The nice girl that presents this weekly show asked me to comment the 5 against 1 issue..

I said: it is something that almost everyone had done or someday will do; It is not a sin... hehehe

Nice weekend Bill!
:w: