How Many Of You Have Eq I.q?

[blueroommusic]

If you are going to EQ something, be cognisant?? of your mixing environment...duh!

and as Falken and many of you said...get it at the source. New strings on an electric guitar, mic placement, tone choice, etc...More than any other instrument IMO, electric guitar is something that to me anyways, sounds contrived, and unnatural when you put an EQ on it...unless you have a Neve, or any other nice channel strip to work with.

a little EQ can go a long way...be aware that as you get older you start to lose much of your high frequencies (even though I'm young)...I am constantly aware of making something sound too brittle, so often times if I attenuate an EQ in the highs, I will take the gain of those bands down a db or so, depending....some of us don't realize how bad our hearing loss is...go get tested.

 

[RDMSstudio]

I'm trying to follow what you are saying here. Is the track that you are lowering in volume being offset with the constant volume track or are you saying it's own timing is changing? Is the track you are lowering getting slower or faster?

The timing is changing as you move the particular fader up and down.
As you lower a tracks volume it appears to speed up the tracks timing. Raising the fader slows the track down.

 

[NYMorningstar]

The timing is changing as you move the particular fader up and down.
As you lower a tracks volume it appears to speed up the tracks timing. Raising the fader slows the track down.

I tried to measure this in wavelab using a click track. I recorded the click on two mono tracks and then exported them panned left and right to a stereo track using automation to drop track two panned right down 40db on the tracks fader. I then looked at all 3 tracks in wavelab and didn't find any timing differences down to the tenths of a ms. What is interesting though is as the wave gets lower in volume, it starts later and ends sooner. I need to go at the moment but I'd sure like to discuss this more

Later...

 

[masteringhouse]

I tried to measure this in wavelab using a click track. I recorded the click on two mono tracks and then exported them panned left and right to a stereo track using automation to drop track two panned right down 40db on the tracks fader. I then looked at all 3 tracks in wavelab and didn't find any timing differences down to the tenths of a ms. What is interesting though is as the wave gets lower in volume, it starts later and ends sooner. I need to go at the moment but I'd sure like to discuss this more

Later...

Huh?

If changing the fader position changes the timing of your track I would suggest that you have a problem with your DAW, or are you speaking psychoacoustically?

 

[NYMorningstar]

Huh?

If changing the fader position changes the timing of your track I would suggest that you have a problem with your DAW, or are you speaking psychoacoustically?

That's kinda the reaction I had when I read the following quote by RDMS. Huh?

There is a little known effect that allows you to fine tune the timing of tracks with the volume control.
If you listen to two tracks together then slowly lower the volume of one, you will hear the timing of it change subtly. Find the volume that makes your track sound most syncronised, then adjust the volume with EQ...

Clever isn't it ?

 

[Middleman]

Let's face it, when you are solo recording at home it is difficult to track sound accurately. You need to be separated from the sound in another room and listening through speakers to make accurate adjustments to mics and compression or even EQing while tracking. When you are in the same room with the sound source, recording your performance, you have make your best guess or record, stop, listen then repeat until it works.

If you record others you eventually learn where instruments generally reside in the frequency spectrum and you can make some EQ adjustments while tracking without too much risk. All this said, your best bet if your recording at home is to come close to the required sound while tracking then learn how to pan your tracks (the first step in EQ actually) and EQ last unless something is really bothering your ears.

I rarely use 2 EQs on a track but have done it. When I first began recording I only cut frequencies but when using samples or softsythns you need to boost a lot of the time. I use Drumkit from Hell Superior and most of the snare samples require a boost in the midrange to rival authentically tracked drums.

Narrow Q is generally applied on a key vocal track where you are trying to squeeze the voice into the track and around other things. It also can be used for obvious problems, primarily from samples once again.

My EQ experience evolved over time as follows:

Initially I was just using hi pass and lo pass to see what they would do. Later I played extensively with Q at various frequencies to see how it affected the sound. If you do a lot of reading and talking to the people that actually make records you find that hi pass is the most often used filter, with various slopes. Most engineers will also dip a little 250-350 on almost every track because this is the mud zone. This will leave a hole in the mix which can be back filled with snare or vocal meat to bring them out of the mix.

One suggestion is to Hi pass on the kick, bass and most any other instrument somewhere around 30 Hz because its all rumble down there and the mastering engineer is going to make this same cut. You might as well take them out early so your compressors are not being fed this data and overreacting. On occasion if I don't need drum tom information I will severely hi pass the overheads up to 500Hz.

On the opposite side you can low pass these same instruments down to 1K or below to keep things clear in the midrange for vocal, guitars, keyboards etc.

Before I touch EQ however, as stated before, I will work the pans first because a lot of the time this reduces the amount of overall EQ required. Next, one of the last things I learned about EQ is finding the sweetspot of the vocal and using narrow Q to fine tune the vocal at various frequency spots (depends on the mic and singer) to make the voice center staged and adding the edge required to sell the song.

A final piece of EQ I learned is that EQ is not always required if you learn and know the sounds that different compressors provide. A compressor is an EQ device too in that in can limit specific frequencies, especially effective if you run an EQ before or after, but many times as a stand alone effect. A little light compression can tame run-amuck frequencies that you might otherwise cut with an EQ. The overall effect may sound better too.

Last, there is EQ automation another whole topic, which can be used to take things forward and back in the mix in real time. This to feature a part or push it back in the mix.

 

[SouthSIDE Glen]

Huh?

If changing the fader position changes the timing of your track I would suggest that you have a problem with your DAW, or are you speaking psychoacoustically?

It's part physical and part psychoacoustic, depending upon how you look at it.

What's happening - or at least what should be happening - is that as one decreases the volume of the wave, it's slope changes. While the wave starts and stops at the same time, the speed at which it attains a given amplitude is slower than it is with the louder wave, causing an apparent delay at that amplitude.

In the attached figure, the blue wave has an amplitude of 10dB (on a positive scale where the horizontal line is 0dB, or silence.) The red wave is the exact same wave with the volume reduced by 4dB. The two green vertical lines represent the points in time where each wave's attack slope reaches an amplitude of 6dB. The distance d is the delay in time it takes the lower volume wave to reach the same amplitude.

This can be perceived by the ear as an actual delay in the signal. The actual length of the delay is both frequency-dependant and amplitude-dependant, however. The longer the wavelength (the lower the frequency), the longer the delay effect. Also, the more the difference in amplitude, the longer the delay effect. I'd venture that this effect is really only useful on the low end of the spectrum, the amount of delay at higher frequencies would be microscopic.

G.

 

[masteringhouse]

G.

Thanks for the explanation, but I'm still not sure that it holds water.

A delay has to be measured relative to something. If you lower the volume of a track it sounds lower in amplitude, you are not hearing it's relative peak next to the track at it's original volume. If you play the tracks together at two different volumes they are additive and you just get a louder track.

Maybe if you played a track twice quckly and immediately lowered the volume it might have this effect, but I kinda doubt it. At least I don't hear it that way, but I have a short-term memory.

To take another example, if you played a low note at precise intervals and different volumes would it sound out of time? If so, that's a good excuse for bass guitarists and a drummer's kick.

 

[SouthSIDE Glen]

G.

Thanks for the explanation, but I'm still not sure that it holds water.

A delay has to be measured relative to something.

I agree completly.

In fact the timing does remain the same because the wave is still peaking at the same time as it was before. If you look at those two waves again, you'll see that the location at which they reach their peak volumes remains the same, therefore there is no perceived change in timing.

The key to the above example was that the delay is in the occurance of equal volume between two waves. It's like Special Relativity; the difference is only measurable relative to another wave.

But I also have several problems with this idea. The first is that the precedence effect would also come into play, which I would think would be the dominant effect here. The second is that my "explanation" has exactly the opposite results as RDMS describes. He says lower volume increases apparent timing speed, whereas my example shows lowering the volume flattens the slope, slowing down the timing of the attack.

I'd sure like more explanation from RDMS when he wakes up down under . When I first read his first post it sounded like he was describing Haas precedence in reverse; instead of changing timing to change apparent volume, you change volume to change apparent timing. I thought he was just describing the other side of the same precedence coin.

But the more I think of it, the more that doesn't make sense either, especially considering his proposition that lowering volume advances the timing, which is really the opposite of what Haas says, where advancing the timing psychoacoustically increases the apparent volume. They are not two sides of the same coin, they are two opposite coins.

I guess I was looking at my explanation more in light of NYMorningstar's observation:

What is interesting though is as the wave gets lower in volume, it starts later and ends sooner.

I think maybe what I had might have been a way of describing what NYM had observed.

G.

 

[NYMorningstar]

I agree completly.

In fact the timing does remain the same because the wave is still peaking at the same time as it was before. If you look at those two waves again, you'll see that the location at which they reach their peak volumes remains the same, therefore there is no perceived change in timing.

The key to the above example was that the delay is in the occurance of equal volume between two waves. It's like Special Relativity; the difference is only measurable relative to another wave.

But I also have several problems with this idea. The first is that the precedence effect would also come into play, which I would think would be the dominant effect here. The second is that my "explanation" has exactly the opposite results as RDMS describes. He says lower volume increases apparent timing speed, whereas my example shows lowering the volume flattens the slope, slowing down the timing of the attack.

I'd sure like more explanation from RDMS when he wakes up down under . When I first read his first post it sounded like he was describing Haas precedence in reverse; instead of changing timing to change apparent volume, you change volume to change apparent timing. I thought he was just describing the other side of the same precedence coin.

But the more I think of it, the more that doesn't make sense either, especially considering his proposition that lowering volume advances the timing, which is really the opposite of what Haas says, where advancing the timing psychoacoustically increases the apparent volume. They are not two sides of the same coin, they are two opposite coins.

I guess I was looking at my explanation more in light of NYMorningstar's observation:

I think maybe what I had might have been a way of describing what NYM had observed.

G.

I think I needed to be more clear in my description. When I was writing about the wave starting later and ending sooner as the volume was being turned down I was talking about the waves representing the whole click and not each individual wave, my bad.

The reason the click starts later and ends sooner is because as you turn the volume down, the software is inserting silence on each end of the sound when the SPL reaches below a detectable level. It has much more effect on the release than it does the attack also, at least visually. In other words the delay at the beginning(the attack) of the click was only about 1/3 of a millisecond but it ended about 8 milliseconds earlier than the click with the constant volume.

Glen, could you please explain how that would fit into the Haas precedence and the psychoacoustical stuff you were speaking of? Could the click ending so much sooner cause it to sound faster psychoacoustically as RDMS is saying?
Thanks

 

[SouthSIDE Glen]

The reason the click starts later and ends sooner is because as you turn the volume down, the software is inserting silence on each end of the sound when the SPL reaches below a detectable level. It has much more effect on the release than it does the attack also, at least visually. In other words the delay at the beginning(the attack) of the click was only about 1/3 of a millisecond but it ended about 8 milliseconds earlier than the click with the constant volume.

NYM, I'd really love to see screenshots of that if it were possible. I could be wrong, but I think it's not a matter if the software "inserting silence" as it is what I was explaining as the initial attack slope just being made more gradual so it may not have been visible at your viewing resolution until 1/3rd ms later. As to why it's 8ms on the trailing end, again I can only speculate, but I'd guess that there was a natural decay in the click sound; that decay was quickened by the lowering of volume for the same slope change reasons given before. But I'd like to, if possible, see before and after shots just to see if I"m understanding your description properly.

Glen, could you please explain how that would fit into the Haas precedence and the psychoacoustical stuff you were speaking of? Could the click ending so much sooner cause it to sound faster psychoacoustically as RDMS is saying?
Thanks

No, I can't explain it . That's why I was hoping and waiting for RDMS to wake up down there in the land where it's winter in June and throw his additional two cents in here. His description *sounds like* it's exactly the opposite of Haas presidence, which is why I don't get it. Maybe I'm misunderstanding him.

The difference in attack timing of 1/3rd ms - which at 44.1k is only about 14 samples - awfully short to be of much signifigance either way, I think. Tests have shown that precedence doesn't really begin to take effect until at least 2ms, and that real domination by the leading source doesn't happen until about 8ms, according to the NIH.

The 8ms difference in the speed of the decay of the signal is *I think* irrelevant to the presidence effect becaue it's all about the timing of the main impulse of the sound (someone correct me if I'm wrong about that.) That 8ms shortening of the decay serves only to make the lowered volume click sound shorter and more punctuated (as well as lower in volume, of course.)

Unless that increase in decay speed is what RDMS is talking about? Perhaps he's saying that by making the sound more punctuated it makes things sound like they're happening faster or earler? I don't know, that would be a totally new on on me. I'm only guessing at this point; none of it seems to quite add up for me just yet.

G.

 

[RDMSstudio]

Huh?

If changing the fader position changes the timing of your track I would suggest that you have a problem with your DAW, or are you speaking psychoacoustically?

Obviously psychoacoustically. The timing of the track does not change at all, how your ears relate the timing of the tracks is definitely influenced. I'm not sure a click track is a good way to judge this. Complex waves perform better.
The big problem is that once you are aware of this phenomina your mixing becomes even more of a balancing act.

This is actually a published phenomina. In nature, echo's and reflections are alway quieter than the original. So a late sound is better off being the quieter of the two sounds.


It's a bit like with panning positions effect. Start with all your tracks at center and get a reasonable mono mix. Then adjust each tracks pan position purely on the basis of sound clarity. As you SLOWLY pan left and right on each track you will hear the sound of the ENTIRE mix change subtly. Find the position that gives best definition to all sounds(compromise in other words), then move to the next fader and repeat.

All this is doing is micro-tuning relative volumes to achieve the same effect.

 

[NYMorningstar]

These are shots of the stereo click file before I lowered the right channel 40 db.
https://homerecording.com/bbs/attachment.php?attachmentid=32449&stc=1
https://homerecording.com/bbs/attachment.php?attachmentid=32448&stc=1

This is after.

https://homerecording.com/bbs/attachment.php?attachmentid=32450&stc=1
https://homerecording.com/bbs/attachment.php?attachmentid=32451&stc=1
This particular click ends about 14ms earlier with the volume reduced.

Sorry I obviously need to go back to html school

 

[SouthSIDE Glen]

This is actually a published phenomina. In nature, echo's and reflections are alway quieter than the original. So a late sound is better off being the quieter of the two sounds.

OK, that's nothing but the Haas precedence effect, and is again also not only in reverse to what you originally described, but mathematically not possible to acheive through volume change alone.

G.

 

[SouthSIDE Glen]

NYM,

Thanks for publishing those screenshots. Yeah, what you are seeing at the end appears to me to be nothing more than the volume of the natural decay envelope being decreased below either viewing resolution or bit depth resolution (probably the former more than the later, is my guess). Nothing unexpected there.

There is a bit of that same thing happening at the start as well. There does appear to be the slightest of shifts (by about 2 pixels) in the reduced volume one like you say, but I honestly think that's more of an artifact of the visual display itself than it is a real shift in the audio.

I personally and honestly don't see any relation at all between what you're seeing there and anything that RDMS is describing, and there is certainly no relation whatsoever between what you're seeing or what RDMS is talking about with the Haas presedence effect.

I think he has his explanation inadvertantly turned inside out.

G.

 

[RDMSstudio]

OK, that's nothing but the Haas precedence effect, and is again also not only in reverse to what you originally described, but mathematically not possible to acheive through volume change alone.

G.

That's the trouble with digital recording mentality ..... This effect was discovered in the seventies when all people used was their ears... Too many people now rely on screens to tell them what the ears should be doing.

All I can say is that it does work, I have used it many times and ALWAYS had the client tell me that somehow their mix sounded tighter.

You can dismiss the concept if you like with mathematical arguments and vector diagrams to prove how it's not possible, or choose to try and hear the effect and maybe learn something about your ability to discern micro detail in a mix.

If you accept that a compressor can be used to tighten timing of a track why do you have a problem with this concept?

 

[masteringhouse]

I don't think that Glen is dismissing it. Actually Glen's answer makes sense to me. With the precedence effect you have two sources to measure from, it's basically a pan knob if we're just playing with volumes, e.g. if you have the same source going to left and right speakers at the same volume the sound will appear to be coming from the center, if you lower the volume of one of the speakers it will seem as if the sound moved to the louder side.

Likewise you can play "tricks" by delaying one of the sides by less than 40 milliseconds and it will seem like the sound moved to one side, but this isn't a volume trick, it's a delay trick.

It's also true as you mentioned that lowering level and rolling off top end (getting back to uses for EQ) makes things sound more distant. I wouldn't equate this with timing though if it's a mono track or you are applying equal amounts to a left and right channel. I you're talking about using more/less on the left or right I get where you're coming from.

 

[SouthSIDE Glen]

If you accept that a compressor can be used to tighten timing of a track why do you have a problem with this concept?

Because i believe you have it backwards, my friend. Please don't just dismiss me with that digital newbie label either; I have been working with analog tape and analog everything else for some 27 years now.

I am very familiar with the effect you're talking about, but you have it's cause and effect backwards. Like Masteringhouse says, it's a volume effect caused by delay, not a delay effect caused by volume.

The Haas presedence effect, which is well documented in audio engineering circles and not a "little known effect", is a psychoacoustic effect that states that when you have two sounds of equal volume that are seperated by anywhere from ~5 to ~30ms (or maybe as Tom says, up to 40ms, the numbers are only fuzzily agreed upon), the human ear tends to hear the earlier sound as being the louder of the two. This is because at those delay times it tends to interpret the second sound as being a reflection or echo of the first sound, which it naturally expects - as you correctly say - to be the quieter of the two.

This effect can be used to great advantage in a couple of ways; it can be used to bring one instrument forward in the mix without having to increase it's volume by sliding it a few ms forward (e.g. with competing kick and bass). It can also be used as a virtual pan like Tom said; if the two sounds (often in this case the same sound double-tracked) are panned L-R, a "ghost image" can be rotated anywhere in between by playing with the delay timing between the two signals. Stuff like that.

But it's all a case of using the timing to create psychoacoustic volume effects, not using volume to create pychoacoustic timing effects. It doesn't work in reverse .

G.

 

[NYMorningstar]

You can dismiss the concept if you like with mathematical arguments and vector diagrams to prove how it's not possible, or choose to try and hear the effect and maybe learn something about your ability to discern micro detail in a mix.

I think the diagrams support what you are saying. There is definitely a timing change in the sense that the click at the lower volume starts a little later and ends alot sooner. My understanding is that hearing it would sound like the track is speeding up some.

If you take complex waves, lower the volume and then increase the volume by adding more bass(longer waves), I can see where there would be a shift in the timing which could be perceived as a tighter bottom end. I hope I'm understanding this correctly and will give it a try soon. I can understand why this would be little known

 

[SouthSIDE Glen]

NYM and RDMS, there is one possibility that I can see as veing potentailly valid. If one frontloads the majority of the energy towards the attack side of the envelope via volume to cut the delay time or via EQ to emphasize the initial attack, it *might* have the psychoacoustic effect of slightly advancing the sound; the average location of the main energy might be a couple of ms ahead of where it would be otherwise. But I do have some serious questions about it:

- the attack is only delayed by 1/3ms (14 samples). The human ear just can't discern - let alone interpret - that as a signifigant delay. It's not until a delay at least six times longer than that where presedence becomes barely discernable. So I think that difference in attack is irrelevant.

- if the advance in apparent energy is moved forward by enough of a delay to be discernable, then the Haas principle will have jurisdiction; I don't see yet how the two are reconciled.

- if this is the case, I don't see it as being a "little known effect" for as long as it has.

I'm just waiting for a proper explanation, which hasn't arrived yet. RDMS's vague reverence to human echo location sounds more like precedence to me than anything new.

G.