the peaks on 90 Hz and 180Hz and the cut on 45 Hz and 135Hz means the distance between the mic and the speaker over the wall equal 3 m 70 cm. So the distance of the speakers to the wall can be around 1 m 85 cm. It could be the wall behind the camera or the ceiling. If you put the speakers closer the wall the peaks and cuts would shift to another frequency.
@@pulidoggy you can search for SBIR effect formulas in the Google. Sound travels in the air at speed 340-345 m/s (or ~2.8 ms/meter) depending on temperature, humidity and the given atmosphere pressure. So the frequency boost or cut depends on the time the initial sound is delayed after bouncing of the walls before mixing with the source (direct sound). Boost or cut happen due to sound wave's peaks and troughs geometrical matching depending on the time the signal (the wave) is delayed. So if you know the distance between the sound source, sound reciever and the walls, you can figure out the time the repeated signal would be delayed. If the distance is 4 meters, the delay would be 4*2.8=11.2. 1000 ms divided by 11.2 (the time delay in ms) = ~90 Hz. 90*2=180 Hz (second harmonic). So 90 and 180 would double the frequency amplitude. Half the frequency (90/2=45) would come in antiphase (180°), so it would cut down the spectrum as well as 135 Hz (90+45).
It is worth mentioning that, should you use an analog measurement mic (or an USB mic whose calibration file doesn't include also sensitivity data) the SPL "software" display reading on REW cannot be considered reliable, but needing to be checked and calibrated against the reference figure from an external SPL meter.
This sounds like good advice. I didn’t realize that the REW SPL readings are inconsistent and need to check against handheld meter. I will explore further. Thanks again. -Jon
@@ripewave Hi Jon, No problem, I'm moving my first steps with REW, and every day there's something new to learn 🙂 It is a very powerful and elaborate tool despite being, unbelievably, free (although I would also mention and encourage the good habit of making a donation to support the huge effort devolved in developing a software that is in no way inferior to other expensive tools) Not for nothing the accompanying manual is 317 pages long, that I'm trying to digest and put into practice little by little 😬 Keep up the good work, and greetings from Italy! Roberto [edit] Also, it is very important non to omit the audio interface calibration step, i.e. the one requiring a loopback between the interface's input and output (or maybe I missed that part in your video?) that is necessary to calculate the interface own response curve, so to be taken into account in subsequent measurements thus avoiding to affect them.
REW is amazing - stands up to commercial products. Yes there is a learning curve. Thanks for the tip on calibration of the audio interface. Yes - important step. -Jon
@@joaquinruiz3039 Sorry, I have no experience about the two hardware pieces you mentioned, but if you can afford them you should know better than I do 😁
When you do a measure and choose L for example. You have to disconnect the R speaker? Or you choose L and measure the Left Speaker with the two, L and R speakers connected and giving you sound?
This was really helpful, thanks a lot. I've searched youtube and wasted hours and nobody actually explained how to use REW to test the speaker till I found yours. One question I would like to ask, was it worth for your to use DAC instead of your laptop audio interface? did it make any worthy enhancement?
I am glad the content was helpful to you. Your question regarding the benefit of the external DAC is a good one. I don’t know how much it’s use impacted the quality of the analysis as I didn’t attempt to calibrate without it. I can say that I can audibly tell the difference when streaming music from my laptop with and without the external DAC - very clear difference. I can only assume it could only improve the quality of the calibration as well. How much I wouldn’t know without a formal test.-Jon
@@mhnoni try to find a video or text that talks about calibrating a sound card if you suspect yours is playing in a nuanced way. in principle, the computer will play a sound and then record whatever comes out of the sound card (not the speakers) and make a comparison between what went in and what came out of all the components that make up your sound card. a correction filter/eq is then applied. with that being said, do not get too deep into the decimal points that we see in today's equipment. there is no practical reason to achieve laboratory precision, first reason being that it's inaudible, like literally impossible to detect with our ears :). we might think we hear things, but blind tests after blind tests show that we can't in any consistent fashion detect differences in SPL of 0.5dB or between 0.001% and 0.01% distortion levels. and that is actually a good thing. it doesn;t mean things shouldn't be built properly, it just means that the "accuracy issue" has been solved for a while and now this is just fun and games when it comes to DACs, amps, software.. ==================== today's sound cards, when used without any DSP effects, are pretty linear/accurate. when it comes to external sound cars, almost anything from a known brand that is over 50$ will be transparent/linear w/ extremely low variations and distortion. also much more quieter than most recordings, listening environments and all that. a quiet fridge is exponentially louder than a sound card noise level and it's still impossible to hear the fridge while playing music at even soft volumes. for example a budget $100 DAC will offer you a "pure original" sound in within 0.5 dB deviations and well below 0.001% distortion. These are basically extremely accurate characteristics for measuring room acoustics (not to mention listening to music). there are other factors much more influential in shaping the sound than the sound cars, amps, mics: speakers themselves can produce 10% distortion and still sound clean, just as a comparison.. the shape of a room and listening position can influence the fq response well over */-5dB, the sofa in the room or a closed, open door or window has more influence on the sound than a proper working sound card -kind of impossible to detect if someone is quietly opening a door behind you when you're listening to music :). when headphones amplifiers are involved, meaning there is a load (a component is using lots of energy, very little compared to speakers but VERY high compared to a line out connection to an amp), then the actual amp in the DAC does play a role in sound quality as it can offer less distortion under 'stressful' conditions... hope this helps the topic of interest.
the peaks on 90 Hz and 180Hz and the cut on 45 Hz and 135Hz means the distance between the mic and the speaker over the wall equal 3 m 70 cm. So the distance of the speakers to the wall can be around 1 m 85 cm. It could be the wall behind the camera or the ceiling. If you put the speakers closer the wall the peaks and cuts would shift to another frequency.
@CyrilleBoucanogh
Interesting... Could you kindly provide some link to the relevant formulas and calculations? Thanks!
@@pulidoggy you can search for SBIR effect formulas in the Google. Sound travels in the air at speed 340-345 m/s (or ~2.8 ms/meter) depending on temperature, humidity and the given atmosphere pressure. So the frequency boost or cut depends on the time the initial sound is delayed after bouncing of the walls before mixing with the source (direct sound). Boost or cut happen due to sound wave's peaks and troughs geometrical matching depending on the time the signal (the wave) is delayed. So if you know the distance between the sound source, sound reciever and the walls, you can figure out the time the repeated signal would be delayed. If the distance is 4 meters, the delay would be 4*2.8=11.2. 1000 ms divided by 11.2 (the time delay in ms) = ~90 Hz. 90*2=180 Hz (second harmonic). So 90 and 180 would double the frequency amplitude. Half the frequency (90/2=45) would come in antiphase (180°), so it would cut down the spectrum as well as 135 Hz (90+45).
It is worth mentioning that, should you use an analog measurement mic (or an USB mic whose calibration file doesn't include also sensitivity data) the SPL "software" display reading on REW cannot be considered reliable, but needing to be checked and calibrated against the reference figure from an external SPL meter.
This sounds like good advice. I didn’t realize that the REW SPL readings are inconsistent and need to check against handheld meter. I will explore further. Thanks again. -Jon
@@ripewave
Hi Jon,
No problem, I'm moving my first steps with REW, and every day there's something new to learn 🙂
It is a very powerful and elaborate tool despite being, unbelievably, free (although I would also mention and encourage the good habit of making a donation to support the huge effort devolved in developing a software that is in no way inferior to other expensive tools)
Not for nothing the accompanying manual is 317 pages long, that I'm trying to digest and put into practice little by little 😬
Keep up the good work, and greetings from Italy!
Roberto
[edit]
Also, it is very important non to omit the audio interface calibration step, i.e. the one requiring a loopback between the interface's input and output (or maybe I missed that part in your video?) that is necessary to calculate the interface own response curve, so to be taken into account in subsequent measurements thus avoiding to affect them.
REW is amazing - stands up to commercial products. Yes there is a learning curve. Thanks for the tip on calibration of the audio interface. Yes - important step. -Jon
@@pulidoggy Hello, I have a Lynx Aurora (n) 8 USB, it does not have preamps, can I do it with the ISA two Focusrite preamp? thank you
@@joaquinruiz3039 Sorry, I have no experience about the two hardware pieces you mentioned, but if you can afford them you should know better than I do 😁
When you do a measure and choose L for example. You have to disconnect the R speaker? Or you choose L and measure the Left Speaker with the two, L and R speakers connected and giving you sound?
Choosing L or R makes the sound card route the signal to that output only, thus no need of disconnecting anything.
This was really helpful, thanks a lot. I've searched youtube and wasted hours and nobody actually explained how to use REW to test the speaker till I found yours.
One question I would like to ask, was it worth for your to use DAC instead of your laptop audio interface? did it make any worthy enhancement?
I am glad the content was helpful to you.
Your question regarding the benefit of the external DAC is a good one. I don’t know how much it’s use impacted the quality of the analysis as I didn’t attempt to calibrate without it. I can say that I can audibly tell the difference when streaming music from my laptop with and without the external DAC - very clear difference. I can only assume it could only improve the quality of the calibration as well. How much I wouldn’t know without a formal test.-Jon
@@ripewave Good to know, thanks again!
@@mhnoni try to find a video or text that talks about calibrating a sound card if you suspect yours is playing in a nuanced way.
in principle, the computer will play a sound and then record whatever comes out of the sound card (not the speakers) and make a comparison between what went in and what came out of all the components that make up your sound card. a correction filter/eq is then applied.
with that being said, do not get too deep into the decimal points that we see in today's equipment. there is no practical reason to achieve laboratory precision, first reason being that it's inaudible, like literally impossible to detect with our ears :).
we might think we hear things, but blind tests after blind tests show that we can't in any consistent fashion detect differences in SPL of 0.5dB or between 0.001% and 0.01% distortion levels.
and that is actually a good thing.
it doesn;t mean things shouldn't be built properly, it just means that the "accuracy issue" has been solved for a while and now this is just fun and games when it comes to DACs, amps, software..
====================
today's sound cards, when used without any DSP effects, are pretty linear/accurate. when it comes to external sound cars, almost anything from a known brand that is over 50$ will be transparent/linear w/ extremely low variations and distortion. also much more quieter than most recordings, listening environments and all that. a quiet fridge is exponentially louder than a sound card noise level and it's still impossible to hear the fridge while playing music at even soft volumes.
for example a budget $100 DAC will offer you a "pure original" sound in within 0.5 dB deviations and well below 0.001% distortion. These are basically extremely accurate characteristics for measuring room acoustics (not to mention listening to music). there are other factors much more influential in shaping the sound than the sound cars, amps, mics: speakers themselves can produce 10% distortion and still sound clean, just as a comparison.. the shape of a room and listening position can influence the fq response well over */-5dB, the sofa in the room or a closed, open door or window has more influence on the sound than a proper working sound card -kind of impossible to detect if someone is quietly opening a door behind you when you're listening to music :).
when headphones amplifiers are involved, meaning there is a load (a component is using lots of energy, very little compared to speakers but VERY high compared to a line out connection to an amp), then the actual amp in the DAC does play a role in sound quality as it can offer less distortion under 'stressful' conditions...
hope this helps the topic of interest.
@@duroxkilo definitely helps, thanks a lot for the detailed info :)
Thanks John!!
Glad it helped. -Jon
Thank you for the tutorial. How do we use REW with Emotiva RMC-1?
I haven’t explored that yet. So far I am using Dirac 100% with my RMC-1. -Jon
Help!