Do you MIX with headphone or monitors?

[synthz]

Just wondering what the feedback in regards to this topic.

I've read that some people say we must never mix with headphones. However i don't have the luxury of thick wall for my noise isolations. So in order to be a 'good neighbour' i have to use headphones every time :(

I'm using Grado SR 80 & 125 for my audio at the moment.

What's your opinion?

[bobbykunkle]

I use headphones usually because you can hear each part better and can use noise cancellation to prevent outside sounds from distracting you. Even though I use headphones, I always check and see how stuff sounds both on my speakers at home and in my car so I know what the sound to most everyone else will be and if a certain part is sticking out too much and balance needs to be fixed.

[kwstakhs]

I usually record & arrange with headphones, but the mixing has to be done with a pair of good monitors.
k.

[Grey Loki]

I agree - I prefer to mix as much as possible on monitors, at several different volumes (to take into account the Fletcher-Munsen curve), and then translate that across to headphones to see how it sounds, rather than working the other way around.

[King_Ellesar]

a lot of headphones don't reproduce a totally neutral EQ balance like most studio monitors do. this is essential for mixing. if you balance a song based on headphones (or non neutral monitors), as soon as you put it on a system that isn't set up like that, it isn't gonna sound like you expected.

i love to arrange and write stuff late at night and headphones work great for that if you need to keep noise down. but unless you have a really nice set of headphones that are made good enough for mixing, I'd stick with mixing with a good set of monitors. they'll give you a better representation of the sound so you can mix a track so it sounds good anywhere.

[el mae de las teclas]

Well I don't do any serious mixing, cause I'm no pro or anything, but when I do mix some of my stuff, I use the KRK RP6 monitors, and then use Sennheiser headphones to help me round the whole picture.

This topic reminds me of a question I've ever had. If almost every sound system used to reproduce music, color the sound, most commonly by boosting the bass, how are we supposed to use the flat response monitors to have a trustful mix?
My question specifically is, we need to try to make the mix sound nicely on the monitors and that's all there is to it?, ... or we need to keep in mind facts like for example that the bass doesn't sound as strong on studio monitors as it does on reproduction systems, and thus we need to avoid boosting too much on the flat monitors mix?

For example, on a mix I might feel the bass is not as present and so I boost it, but then when I play it on a regular reproduction system I find out is too loud. So I must learn to compensate those differences?

I hope you understand what I'm saying.
Thanks,
Andrés

[Grey Loki]

el mae de las teclas - the idea is that professional studio monitors give you a high fidelity sound so that you can accurately record and mix instruments and all their subtlties. Once you've done that, it's normal to listen to your mix again on reference monitors such as Yamaha NS10s (which are, in effect, nothing more than high quality hifi speakers), and then on any other sound systems that you have handy - your cheap walkman headphones, your car stereo, your home hifi system, and so forth, taking note of what sounds good and what doesn't, and going back into the studio the next day (with fresh ears) to remix until you find a happy balance.

[RobD]

I usually use both to get something to reference to and get the detail right.

[synthz]

Thanks for your feedback guys!

I really don't think i can afford the 'noise' in our living environment, i mean.. when you monitor a mix, you probably gonna have it fairly loud, and sometimes quiet.. I'll see if my Grado is not flat enough, i'll get more higher end headphones, preferably made for mixing perhaps...

I do check in different environment everytime i finish a mix as best as i can..

Grey-Loki - Funny that you mention Fletcher-Munson curve, i just read that on my way to work this morning, but it was in EQ topic? it must be quite a broad theory then.. They always say "If you can a good sounding mix with NS-10, it'll sound good everywhere" probably due to it's 'crappyness' hehehe... It's super flat perhaps..

el mae - i think you are on the right track, checking in different environment prevent you from those problems. THat's why they like their monitors to be as flat as possible.

[Grey Loki]

Synthz - yeah, the Fletcher-Munsen curve is basically the frequency response of the human ear, and how it changes (flattens, IIRC) at higher SPL levels.

The NS10's aren't really 'flat as a ruler' monitors - they're called reference monitors purely because they give you a hi-fi reference of what your track would sound like on someone's home stereo.

If you're looking for super-flat frequency response in studio monitors, check out the Earthworks Sigma 6.3 studio monitors ;)

EDIT: http://en.wikipedia.org/wiki/Fletcher-munson_curve <For those who were more curious about the Fletcher-Munson curve (I've been spelling it 'Munsen' for years :S)

Also, the 'optimum' level for monitoring is around 85dB, which is around the level of conversation. That's actually a very good rule of thumb for conserving your ears in longer studio sessions - if you have to turn down the monitoring bus master to have a conversation with somebody, you're monitoring too loud - the ideal level is just under 'conversation level' :)

[synthz]

Grey Loki : Cool! I was reading this EQ article this morning, cause i want to 'demistify' EQ :) It's gonna be funny mixing and tweaking EQ, and referrencing to the chart heheh.. i guess i need to memorize them eh...

Just had a look at your monitor recommendation. Hell, do you have any idea how much they are? :P http://www.bayviewproaudio.com/Sigma_63-p-21775.html
AU$8000 (sale price, discounted from $10,000)... that's like a price of a car! :)
They look really slick though, the shape is unique.

Thanks very much for the 85db tip! I never knew that, i guess if it's just slightly below 'conversation level' then it wouldn't be THAT loud... i will think if i can really afford the noise level....

[el mae de las teclas]

I need to thank you all too aswell, cause I learned some stuff too!

[Grey Loki]

Grey Loki : Cool! I was reading this EQ article this morning, cause i want to 'demistify' EQ :) It's gonna be funny mixing and tweaking EQ, and referencing to the chart

I personally wouldn't do that - mixing to a chart rather than with your ears is just not a good way to work. My personal favourite EQ'ing technique, that i've seen a LOT of engineers use (Including Paul Borg, and the editor of Sound on Sound magazine) is to dial up the narrowest Q you can, boost that band by about 18dB, then sweep through its entire frequency range, looking for areas where it sounds AWFUL, then widen the Q slightly and notch out that area by however much is needed - usually 3-5dB. I like to use mics to 'add frequencies' and get a good sound, and then use EQ to remove any horrible ones - EQ (in my opinion) should be thought of as a corrective tool, rather than something that should be used as a matter of course - get a good sound with the mic before you reach for the EQ, it's much easier to mix with :)

EDIT: And yeah, the earthworks stuff is expensive, but boy oh boy does it sound good!

[synthz]

GreyLoki: I see.... thank you for the tips again... you know a lot about audio engineering! Last night I was watching tutorial DVDs about EQ & Compressor.. it's really hard to stay awake.. :) I know they're speaking english.. but it feels like out of this world lol.. I'll get there slowly,....

I wouldn't be using mics for now, so i'm kinda safe in that area... It's just line levels of keyboards & VSTi.

That's right! They always say... 'cut before you boost' :)

Let me try to digest your tips here...
1. Adjust Q (using 1-2 perhaps?)
2. Boost band +- 18db
3. Sweep for unwanted freq
4. Adjust Q to fix, plus freq (3-5db)

Have you had a chance to listen to the Earthworks in the showroom or something? Don't tell me you actually own a set! :O

[Grey Loki]

No, I don't own a set, but my uni has a few matched pairs of their QTC50s, and one of the studios is fitted with the Sigmas - rather niiiiice!

Yeah, you've got that down - in case you didn't know what a Q is, i'll explain it here:

EQs come in several different 'flavours' - graphic EQs and parametrics (or sweepable) are the most common. Graphic EQs usually have 31 bands of ISO-standard freqencies that you can notch out individually - they're mainly used for removing feedback in live situations.

A graphic EQ, from THE makers of graphics - Klark Teknik. http://www.tmc.ltd.uk/acatalog/dn370sm.gif

A parametric EQ is the one that's featured on desk channel strips (as well as mic preamps and outboard EQ units). You'll usually have a dedicated HF and LF attenuator pot, that can cut or boost the extreme HF and extreme LF by whatever amount. Between these two are (usually) two 'sweepable' EQs - a Low Mid, and a High Mid, which (again, usually) consists of three pots (potentiometers - knobs, basically :P). One is used to select what frequency you're boosting or cutting. The next is how much you're boosting or cutting this frequency (usually in dB, but IIRC some desks just use 1-10).

The last knob is the 'Q' value. If you imagine a Logic parametric EQ plugin (which I use as an example because it has a graphical representation), such as the one in the link below, you can see the 'shape' of the EQ running across the middle - this particular EQ curve has a little bit of extreme LF roll-off, followed by a bass boost at around 76Hz. Take a look at the curve there - it's quite wide (especially in comparison to the cuts at ~500Hz and ~1KHz) - that's the Q value being set wide (which is depicted by a low number) - if you look at the two cuts, you can see they're like needle points - a high Q in effect.

http://www.soundonsound.com/sos/nov03/images/logic01eqsettings.l.jpg

I'm sorry if you already know this; I don't mean to patronize you at all, but my teachers were good enough to sit me down and explain this stuff, and i'd repay them poorly if I didn't take the time to explain it to anyone else who needed a bit of help.

I know a -bit- about sound engineering beacuse i've been working as a live engineer for four years, a studio engineer for three, and am currently studying for a degree (a BA, to be precise) in Music Technology.

Hope this helps - if not you, then someone else, perhaps :D

EDIT: Misc. grammar corrections.

[SerFox]

I don't use professional gear to mix because hardly anyone ahs professional gear to listen.

I listen with headphones to make sure theres no clip and the levels are okay, then I eq it slightly.

Then I put it onto my rokit speakers and mix some more.

Then to my 20 quid pc speakers

Then to my cheap 2 pound cd player speakers

then back again to the rokit and then play through them all again. if it sounds decent on all of them I've made it good. You can mix somehting on professional speakers and it sounds great then it goes into a consumer hifi and sounds terrible, because your speakers could handle the bass and the hifi couldnt.

pro gear for mixing = big nono.

[Grey Loki]

Erm....dude....

Are you saying that EVERY pro studio in the world is wrong?

[el mae de las teclas]

Grey Loki, one question.

When we sweep the freq range for the offending frequencies, are we supposed to do that for each specific musical pasage? Cause the music, and specially progressive has many changes within a song, and many sections with different instruments that need different EQ. So I suppose its logical to do that sweep like for every different segment within a song where different instruments interact right?

[SerFox]

Erm....dude....

Are you saying that EVERY pro studio in the world is wrong?

Nope, I'm just answering the topic title:

Do you MIX with headphones or monitors?

emphasis on 'you' and since everyones been going on about how they do it I thought I'd state also. To summarize: I do it with both in my own ways.

[Grey Loki]

Then you ended your post with a sweeping statement that sounded like, if you could, you'd get every studio to ditch their NS10s, their 1038's, their Sigma 6.3s, or whatever, and buy the cheapest crap that their local Argos had for sale :S

el mae de las teclas - i've never been that specific with EQ, and I doubt that many other people are. If you listen to (for example) a bassoon passage, and you decide it sounds to 'honky', even in just one passage, you can just EQ out the honk - it (instruments in general) doesn't (in my experience) get more or less 'honky' depending on what note is played.

EDIT - found this earlier - I think that you'll find it useful - http://www.independentrecording.net/irn/resources/freqchart/main_display.htm

[synthz]

Grey Loki - Whoaaaa!! No wonder you're very knowledgable in this area!! :) You've had the real life experiences and you also study formal education again.. that's really excellent man..

Yes, of course you've been helpful, thank you very much for the lengthy explanation. EQ is such a tough thing to master (we can fill up a whole new topic just on this), the more i can learn the better.
OK, just try to re-iterate what i've digested....
Graphic EQ is much more the standard EQ we've seen on our hi-fi system, probably w/ more bands perhaps.. Which situation are we gonna be using graphic EQ?
Parametric EQ is the one w/ the visual curves. The HP & LP filters must be realllly useful in a track bus situation? Should this be the ideal one to be used in a track bus?
Which type of EQ is ideal for master bus?

The Q knob - Would it be correct if i just see this as 'the amount of the curve for each node'?

Logic EQ Example - This is a good example. I can see the differences between the wide and narrow curves with different Q amounts. Is Q just used to 'smoothen out' the transition only?

I'm gonna try to guess what they're doing here, it would be good if i know what sort of track this is. Please correct me if i'm wrong :P
1. 20-40khz rolled off to eliminate sub bass
2. 4o-250khz boosted up to fatten the kick drum?
3. i don't know what the 2 extreme dips in 500 & 1000khz does :\
4. 1.2k - 12k boosted .. Is this the upper mid range area? where most of the instruments freq lives?
OK, that's my crap analysis hehehe... i bet i'm wayyy off! :P

Thanks again for your kindness to spread your knowledge Grey Loki!

Wow.. that chart from independentrecording is reallllllllly good, looks like we've got all the instruments in there... But just remembering your prev tips, should we use a chart or not when mixing? Or perhaps use this when starting out until we memorize them?

[Grey Loki]

You usually find parametric EQs on HiFis, unless it's a very high end one that has its own outboard GEQ.

On the subject of Q again, i've updated that sound on sound image - hopefully this makes it more clear - http://img169.imageshack.us/img169/492/60301388js2.png

I have no idea what that EQ curve is for, so I wouldn't even try examining it without knowing exactly what the track was.

The assorted frequencies of different instruments inhabit the entire frequency spectrum, but different boosts or cuts will yield different effects, depending on what band you're cutting.

The chart that I linked to is a very useful tool - obviously, you wouldn't mix with it, but if you're (as an example) in the middle of a mix and can't find what's honking on your whatever, you can just take a look on the chart there to find out what the different frequency bands do in say, a lead guitar sound, and cut or boost as required.


Parametric EQ's don't -usually- have a visual curve that you can 'see' the effect of the curve with - I linked Logic's channel EQ purely because it's really useful as a teaching tool, but the affect the sound in that 'shape' (usually) - long, rounded curves, as opposed to a graphic EQ which needles out specific frequencies - that's why it's used in live sound so much; if you have feedback somewhere right in the top end, you don't want to cut all of that, or you'll lose the definition of the sound, and it will sound crap to the audience. Instead, you find the frequency on the graphic EQ, and notch it out there instead.

If you're going to be mastering, (again, usually, because in sound engineering, i've found there are no real rules - just guidelines and 'what normally works') a track, you would usually use a very nice outboard parametric EQ - something like the stuff found on an SSL desk, which IMO is very neat and musical, but still punchy.

I hope I answered all of your questions - let me know if you have any more :)

[synthz]

Hey thanks for the updated image... i'm more of a visual guy, so that makes it really clear for me.
Low Q gives you narrow shape, and high Q gives you wide shape (smoother transitions)

"The assorted frequencies of different instruments inhabit the entire frequency spectrum, but different boosts or cuts will yield different effects, depending on what band you're cutting."
-> I see, so they're 'inter-dependent'

"The chart that I linked to is a very useful tool - obviously, you wouldn't mix with it, but if you're (as an example) in the middle of a mix and can't find what's honking on your whatever, you can just take a look on the chart there to find out what the different frequency bands do in say, a lead guitar sound, and cut or boost as required."
-> Ahhh.. that must be the real life situation where we can use the chart. Thanks!

Yes seems like people always have their own faves of EQ and SSL emulations... i need to experiment more to find which one i like, and learn from all the mistakes that i'm gonna do...

I have more questions if you don't mind :)

What is Paragraphic EQ, and when we should use it?

Yeah, i would have to master myself, nobody's gonna do tracking, mixing & mastering for me.. it's all 1 man band :)

p.s. I'm gonna try to record this weekend and hopefully do some rough mixing. I hope you would be able to give some feedback about the mix, i will post it later in the Show Off section.

[el mae de las teclas]

Synthz, actually the Q works the other way around. High Q values, produce narrow curve, and low Q values produce wide curves.

Also you made a mistake when describing the different freq ranges.
The audible estimated range for humans goes from 20hz to 20.000 hz or 20khz, as k stands for kilo (1000).

Grey Loki, ok so this sweep is supposed to be done to solo instruments? Not with the whole thing playing?
At first I thought this sweep should be done to find bad sounding freq but in the whole mix! I guess that wouldn't do much sense right?

Another question just jumped into my mind. Do you use spectrum analyzers? How are they commonly used?

I hope you don't get tired of answering questions too soon, hehe.

Thanks again.
Andrés

[Grey Loki]

No, I don't get tired of questions - it helps to hone my own knowledge as well, by putting what I think I know into a form that's readable to other people

synthz - paragraphic EQs are quite rare; so much so, that i'd never heard of them until you asked :D

A quick google turned up this link, which explains it in pretty good detail - http://www.sweetwater.com/expert-center/glossary/t--ParagraphicEqualizer

I guess that by that definition, Logic's channel EQ (the one pictured in the links I posted) is a paragraphic EQ - that is, a parametric EQ, with a graphic representation. You'd use it wherever you'd use a parametric EQ.

I would honestly never master your own tracks yourself - mastering is supposed to take place by engineers with decades of experience, using super-high-quality monitors, in a specially treated room, with the best outboard gear that money can buy. They'll also have a fresh set of ears, and will hear things that you missed, because you're so 'into' the track - if you can, get someone else to master.

el mae de las teclas - yes. If you hit (as an example) middle A on a piano, you won't just get the fundamental frequency of that note - 440Hz. You'll get overtones (or partials, or harmonics (NOT guitar-style harmonics - that's something different) - to my understanding, they're different names for the same thing) all the way up (and down!) the frequency spectrum.

Different parts of this spectrum have different effects - looking at that chart, I can see that if I had a piano lacking a bit of 'oompfh', I could dial in a parametric EQ with a low Q to boost the 80-120Hz band by a few decibels, which would give me a fuller sound.

The human ear's frequency response is billed at 20Hz to 20Khz, but that's not totally accurate - most people's hearing will go from 20Hz to around 16KHz - obviously, the older you get, the lower this figure gets - the same as if you do lots of loud gigs. I've met people who can't hear anything about 12KHz. It's heartbreaking - protect your ears! As a little 'test' you can do for your hearing, if you mute your television (assuming you have a normal one, rather than a TFT or plasma screen), and can hear a very high-pitched 'eeeeeeeeee' sound, cool - that 'eeeeee' is around 15KHz.

Yeah - (in the studio) I solo each instrument, and do what I described a few posts ago, just to see if there's anything absolutely awful sounding in there, then notch it out a smidge to improve the overall sound. Doing it to the whole mix....Hmmm. I've never done it, or even TRIED it, but there's no real reason why it wouldn't work. That decision would be up to the mastering engineer (and his fresh ears), though.

I personally don't use spectrum analyzers. There are probably several uses for them. The one that springs most readily to mind is in a live situation, you could dial in a spectrum analyzer being fed by a reference mic (one with a ruler flat frequency response). That way, if you got some feedback, you could see what it was on the spectrum analyzer, and notch it out on the relevant graphic EQ.

Hope i'm helping :)

[el mae de las teclas]

Absolutely, thanks!

Actually is quite funnny what you said about that "eee" produced by the Tv's! Because I have always heard that, I can say that I hear that high pitched sound even when there is volume, or for example when I get home, I can tell when someone is watching TV even if the volume is very low to tell, because of that high pitch. I have noticed that I have quite sensitive hearing, because of other things aswell.

Anyways, I was about to open a thread regarding that high pitch sound made by the tv, because I have asked my parents about it, and even my girlfriend I believe, and none of them is able to hear it, so I wanted to make sure that I wasn't imaging it, hehe.

Regarding the spectrum analizer, I guess is quite helpful for people that is more visual. Or if someone can't tell which instruments might be competing on the same freq range, I guess it will be easier to detect that and then take decisions for EQing.

Cheers!
Andrés

[synthz]

el mae: you're right! :P I got it mixed up... and the hertz thing as well..

Good question about the spectrum analyser! What i've seen is people open up the track and solo it w/ the analyser to find the actual freq of the instrument, then tweak it after that... Easier to said than done though :)

Grey Loki : Thanks for the tips about the paragraphic EQ! It's much more clearer to me now...
I know mastering supposed to be done by ME, but if it's just my own fun projects and nothing serious i can't afford the cost to hire studio & the engineer... that's just not ideal for me.. unless i'm doing a serious project.

Talking about TV high pitch. Glad to know ppl have this syndrome as well. Sometimes at work i suddenly have slight ear pain like this annoying frequency suddenly comes up... it's from the TV. I also can tell when someone just turn on the TV from that pitch (without seeing them)... are we turning into dogs? :P