Random audio trivia of the day

[normen]

Localization of sound sources is especially dependent on frequencies around 8kHz.

Try it! Cut 8kHz on your whole mix / speaker controller and hear the sound detach from the speakers. Boost 8kHz on a guitar that doesn't firmly stand in the stereo field of your mix and hear how it becomes more stationary instead of washed out. It's a quite narrow band, try to get it as focused as possible by changing the Q factor, move the frequency a bit around to hear the change in the effect.

[jimmyklane]

I’ve seen this empirically in my own mixes, especially apparent when mixing in a big finely-tuned room where you can actually hear stuff like the SPL Vitalizer, etc....This should work in Reason but only on “E” mode, as the G series EQ is really a broad brush....say a sword instead of a scalpel . I’ve always thought it to be a little lower however....in what I call the “alcohol frequencies” (a little bit is really awesome and too much makes you sick!) around 5-7kHz....to my ears this is just like the “clarity” slider in Adobe Lightroom/Photoshop....it gives what they call local contrast. Interestingly enough, this is close to the area that I cut with the deesser too, as adding some sparkle after compression gives you that hard-edge.

I like to boost these frequencies going into the sampler set to a lower sampling rate, giving a little extra grit from aliasing and ringing of the imaging filter.

Normen, do you have the scientific or psychoacoustic reasons behind this phenomena? I’d be interested in white papers, etc.

[normen]

I’ve seen this empirically in my own mixes, especially apparent when mixing in a big finely-tuned room where you can actually hear stuff like the SPL Vitalizer, etc....This should work in Reason but only on “E” mode, as the G series EQ is really a broad brush....say a sword instead of a scalpel . I’ve always thought it to be a little lower however....in what I call the “alcohol frequencies” (a little bit is really awesome and too much makes you sick!) around 5-7kHz....to my ears this is just like the “clarity” slider in Adobe Lightroom/Photoshop....it gives what they call local contrast. Interestingly enough, this is close to the area that I cut with the deesser too, as adding some sparkle after compression gives you that hard-edge.

I like to boost these frequencies going into the sampler set to a lower sampling rate, giving a little extra grit from aliasing and ringing of the imaging filter.

Normen, do you have the scientific or psychoacoustic reasons behind this phenomena? I’d be interested in white papers, etc.

The wavelength of an 8kHz wave is 4,3 cm, somewhat similar to a human ear, right? HRTF would be the keyword here.

I could dig up some papers on that but nothing I could actually give you that you couldn't google yourself or would have to un-paywall yourself. It's pretty much historic knowledge and just written as a fact in many audio "standard works" like the "Handbuch der Tonstudiotechnik" in Germany.

A colleague of mine did something about this while in University and found some little "wiggles" in that Area that pretty much universally worked as a "5 o'clock" or "7 o'clock" generator when applied to any audio material. He pretty much synthesized a certain angle into the audio instead of applying what's basically an IR of "sound coming at an eardrum from that direction" like in most HRTF systems.

[selig]

…This should work in Reason but only on “E” mode, as the G series EQ is really a broad brush....say a sword instead of a scalpel

Side note:
Yes, this is true only IF you stick to low boost/cuts. But the more you boost/cut, the more the bandwidth becomes the same with either E or G modes. At max boost/cut, Q is identical for both modes.


Sent from some crappy device using Tapatalk

[jimmyklane]

…This should work in Reason but only on “E” mode, as the G series EQ is really a broad brush....say a sword instead of a scalpel

Side note:
Yes, this is true only IF you stick to low boost/cuts. But the more you boost/cut, the more the bandwidth becomes the same with either E or G modes. At max boost/cut, Q is identical for both modes.


Sent from some crappy device using Tapatalk

Sure, proportional vs constant Q

[selig]

Side note:
Yes, this is true only IF you stick to low boost/cuts. But the more you boost/cut, the more the bandwidth becomes the same with either E or G modes. At max boost/cut, Q is identical for both modes.


Sent from some crappy device using Tapatalk

Sure, proportional vs constant Q

Yes, and as it happens neither mode is a scalpel…

But back on topic, this would imply you can find this frequency “by ear”, if you will, because it’s related to our ear’s physical design?

And if so, would one test be to sweep a narrow cut on white noise and find the “magic frequency”?

I just happen to have an EQ that does ridiculously narrow cuts, probably too narrow for this test. @Normen: What bandwidth do you suggest for this effect?


Sent from some crappy device using Tapatalk

[jimmyklane]

I’ve seen this empirically in my own mixes, especially apparent when mixing in a big finely-tuned room where you can actually hear stuff like the SPL Vitalizer, etc....This should work in Reason but only on “E” mode, as the G series EQ is really a broad brush....say a sword instead of a scalpel . I’ve always thought it to be a little lower however....in what I call the “alcohol frequencies” (a little bit is really awesome and too much makes you sick!) around 5-7kHz....to my ears this is just like the “clarity” slider in Adobe Lightroom/Photoshop....it gives what they call local contrast. Interestingly enough, this is close to the area that I cut with the deesser too, as adding some sparkle after compression gives you that hard-edge.

I like to boost these frequencies going into the sampler set to a lower sampling rate, giving a little extra grit from aliasing and ringing of the imaging filter.

Normen, do you have the scientific or psychoacoustic reasons behind this phenomena? I’d be interested in white papers, etc.

The wavelength of an 8kHz wave is 4,3 cm, somewhat similar to a human ear, right? HRTF would be the keyword here.

I could dig up some papers on that but nothing I could actually give you that you couldn't google yourself or would have to un-paywall yourself. It's pretty much historic knowledge and just written as a fact in many audio "standard works" like the "Handbuch der Tonstudiotechnik" in Germany.

A colleague of mine did something about this while in University and found some little "wiggles" in that Area that pretty much universally worked as a "5 o'clock" or "7 o'clock" generator when applied to any audio material. He pretty much synthesized a certain angle into the audio instead of applying what's basically an IR of "sound coming at an eardrum from that direction" like in most HRTF systems.

I’m familiar with ORTF, jecklin, Schneider, and binaural techniques (HRTF) but haven’t really thought about using EQ to stimulate the natural resonance frequency of our ears. Interesting discussion.

[normen]

I just happen to have an EQ that does ridiculously narrow cuts, probably too narrow for this test. @Normen: What bandwidth do you suggest for this effect?

I'd say around 2 you should have enough of an effect without doing too much other stuff but depending on the speaker (I mostly use it to make speakers "vanish" or monitors "be there") you can go as far as 10. You also don't really have to cut much most of the time. But yeah, just try

[normen]

Sound needs about 3 milliseconds per meter to travel through air

Consider this real world value when dealing with and handling latency in systems, be it computers or digital desks.

..and of course PLEASE consider it when setting up loudspeakers in your school or for local town festivities to avoid the typical train station feel

[normen]

Sound waves have a length of about 1,7 centimeters at 20kHz to 17 meters (!) at 20 Hz

Consider this when placing microphones in stereo or multi-mic setups etc. 2 centimeters can change the sound but they won‘t solve your low frequency canceling issue.

[WongoTheSane]

..and of course PLEASE consider it when setting up loudspeakers in your school or for local town festivities to avoid the typical train station feel

Can you elaborate on that? What yields the typical train station feel, a comb filtering due to a regular spacing of loudspeakers or is it something else?

[normen]

..and of course PLEASE consider it when setting up loudspeakers in your school or for local town festivities to avoid the typical train station feel

Can you elaborate on that? What yields the typical train station feel, a comb filtering due to a regular spacing of loudspeakers or is it something else?

Not having a delay on additional speakers that are standing further away from the stage. You need to put a delay according to the distance they stand off from the main speakers. Thats why those speakers are called a „delay line“. If you don‘t do that the wave front of the delay line speakers arrives before the main speakers, causing an echo/phasing effect depending on the distance. Add more speakers and you get more and more echoes and diffusion. In german I’d say the sound „klappert“.

[WongoTheSane]

Can you elaborate on that? What yields the typical train station feel, a comb filtering due to a regular spacing of loudspeakers or is it something else?

Not having a delay on additional speakers that are standing further away from the stage. You need to put a delay according to the distance they stand off from the main speakers. Thats why those speakers are called a „delay line“. If you don‘t do that the wave front of the delay line speakers arrives before the main speakers, causing an echo/phasing effect depending on the distance. Add more speakers and you get more and more echoes and diffusion. In german I’d say the sound „klappert“.

Gotcha, thanks! I think I'm going to have a train PA system in my next song...

[MarkTarlton]

my favorite pultech eq band is 8khz, granted it's a crazy curve and I set it pretty wide, it helps instruments pop out better than most other tools I have.

[aeox]

my favorite pultech eq band is 8khz, granted it's a crazy curve and I set it pretty wide, it helps instruments pop out better than most other tools I have.

So it's just a regular eq right? What makes it your favorite? Saturation or something?

[MarkTarlton]

my favorite pultech eq band is 8khz, granted it's a crazy curve and I set it pretty wide, it helps instruments pop out better than most other tools I have.

So it's just a regular eq right? What makes it your favorite? Saturation or something?

it's not just a regular eq, especially if you have the hardware unit. it's pretty magical.

[aeox]

So it's just a regular eq right? What makes it your favorite? Saturation or something?

it's not just a regular eq, especially if you have the hardware unit. it's pretty magical.

What do you mean by magical?

[MarkTarlton]

it's not just a regular eq, especially if you have the hardware unit. it's pretty magical.

What do you mean by magical?

to my ears it's the most pleasing high end. it's like a soft pillow instead of a hard spikey bed made of nails

[selig]

What do you mean by magical?

to my ears it's the most pleasing high end. it's like a soft pillow instead of a hard spikey bed made of nails

At least SOME of the Pultec “magic” is based on the fact the total available boost isn’t that much compared to other EQs. So even for an EQ with similar boost/cut, the Pultec’s boost (and cut) is on ONE knob, so you have the entire range of the knob to cover only about 10dB or so total boost. Because of this, most folks end up using less EQ, which always sounds better and more “subtle” (less harsh).

[jimmyklane]

So it's just a regular eq right? What makes it your favorite? Saturation or something?

it's not just a regular eq, especially if you have the hardware unit. it's pretty magical.

Do you have the 40 y/o original or one of the reissues by Pulse? Or another company’s clone?

I’ve never had a huge use for Pultec EQP-1A unless I’ve got access to a pair of them. If I’ve got access to a pair then they DO have a really sweet flavor to them. Driven hard you can get some lovely harmonics from the inductor saturation artifacts as well as the gain makeup circuit. The boost/cut situation creates its own sound....which is admittedly extremely sweet and musical. The MEQ-5 on the other hand is perhaps my favorite “broad brush” EQ of all time when it comes to bringing out the “reality” of a recorded source. I’ve found it able to either push a source back or bring it forward by dipping at 200, boosting at 300 and 5kHz.

I tend to reach for a compressor (of which I have far more) to alter tone before EQ....you can dull the high end, boost the bass, bring up transient (high frequency) energy, and more using a compressor or chain of compressors. Then again, I also like subtle saturation and tend to dislike phase shift if I can avoid it.

[MarkTarlton]

to my ears it's the most pleasing high end. it's like a soft pillow instead of a hard spikey bed made of nails

At least SOME of the Pultec “magic” is based on the fact the total available boost isn’t that much compared to other EQs. So even for an EQ with similar boost/cut, the Pultec’s boost (and cut) is on ONE knob, so you have the entire range of the knob to cover only about 10dB or so total boost. Because of this, most folks end up using less EQ, which always sounds better and more “subtle” (less harsh).

great points! I do like the fact that the cut and boost are that way. I think to put it into more musical terms...I try and turn the knobs til I hear what I want the sound to do....with a neve it's 7.2khz and I definitely am not afraid to crank it. they have a sound that is different, but in terms of the pultech sound it still has it's flavor...the circuitry does stuff, but that curve and the sound that comes out is pleasing to my ear.

[MarkTarlton]

it's not just a regular eq, especially if you have the hardware unit. it's pretty magical.

Do you have the 40 y/o original or one of the reissues by Pulse? Or another company’s clone?

I’ve never had a huge use for Pultec EQP-1A unless I’ve got access to a pair of them. If I’ve got access to a pair then they DO have a really sweet flavor to them. Driven hard you can get some lovely harmonics from the inductor saturation artifacts as well as the gain makeup circuit. The boost/cut situation creates its own sound....which is admittedly extremely sweet and musical. The MEQ-5 on the other hand is perhaps my favorite “broad brush” EQ of all time when it comes to bringing out the “reality” of a recorded source. I’ve found it able to either push a source back or bring it forward by dipping at 200, boosting at 300 and 5kHz.

I tend to reach for a compressor (of which I have far more) to alter tone before EQ....you can dull the high end, boost the bass, bring up transient (high frequency) energy, and more using a compressor or chain of compressors. Then again, I also like subtle saturation and tend to dislike phase shift if I can avoid it.

I use the original ones a lot and recently, and have a few newer ones at home. I like to use them after compression on kick and bass...plus on buss's...and a lot of other things...vocals and ac gtrs....I don't have enough time to go on about how important these units are to me.

[aeox]

in terms of the pultech sound it still has it's flavor...the circuitry does stuff

I'd be interested to know what that circuitry is doing!

[MarkTarlton]

in terms of the pultech sound it still has it's flavor...the circuitry does stuff

I'd be interested to know what that circuitry is doing!

a lot of hardware that was made back in the day has a sound. I like to put the fairchild on the mix buss and leave the comrpression circuit off while it passes through all the tubes/modeling...it's pretty cool try it out! make sure the threshold is off.

[aeox]

I'd be interested to know what that circuitry is doing!

a lot of hardware that was made back in the day has a sound. I like to put the fairchild on the mix buss and leave the comrpression circuit off while it passes through all the tubes/modeling...it's pretty cool try it out! make sure the threshold is off.

Unfortunately I can't afford that stuff :/

I tried with a PuigTec EQ once and it didn't phase cancel. So, something is happening (not volume) Could just be mild saturation TBH. I'm just looking for more of a scientific answer as to what is happening.