TOPIC:

Guys this is driving me crazy. You know when you're searching for an answer and can't find anything?!

So I was watching a video on YouTube about the Geddes approach. In that video the presenter explained that it takes our ear/brain a certain amount of time to register a tone. The example given was it takes 30ms to register 100hz. The point they were making was that reflected sound then is perceived as direct sound. 

They went on to say the lower the frequency, the longer we need to perceive it. Great OK - so begins my search. I want to know how to calculate how long different frequencies take before we perceive them. I'd settle for a table or graph of frequency vs time if someone has already worked it out. 

Anyone know where to look?

Thanks in advance. 

I saw that video…..  if we need 1/4th the wavelength to our ears to hear it as a tone then we should be able to make a spreadsheet and logmerithicly (sp?) turn a downwards phase response…

it would be easy to make an fir to compensate for that, but I bet it would sound kinda funny. 

but if our ears sence pressure changes, we only need 1/4th of the wave and nature does the rest. Just the 1st peak really. But it needs to be consecutive I would guess, the rarefaction and compression cycles to hear the “tone” otherwise it would just be a pulse of sort…. I don’t know how many of the cycles needed at different frequencies it would take, it would seem obvious that it is frequency dependent. 

if you can figure that one out it would be pretty easy to build an FIR to remove the delay. (As long as it’s less than one cycle at any given frequency below 100hz) it would be a lot of phase that would have to be removed otherwise and that might be impossible with hardware fir 

Hi,I don't want to know so that I can compensate. I just want to know because I'm a bit sad and i'm just interested! 🙂 Since posting I've been thinking about it some more. I found a site where you could select how many full cycles you wanted to play. If you select one cycle it sounds like a click. If you change to hear 3-4 cycles it starts sounding more like a tone. And by 6 repetitions the tones can be heard clearer.  So from that I'm thinking the way you might be able to work it out would be, knowing the frequency's period then let's say it takes 'X' full cycles:

 Wave Period multiplied by 'X' = time to register.

E.g. Assuming 6 cycles needed.For 200hz with a period of 0.005 sec that's 0.03secs to register the tone.

If that theory is right we just need to know what the correct number of cycles might be but suspect that may be different from person to person and for different frequencies. Still the formula above could put us in the right ballpark. 

Interesting….. so you can be the 
Guinea pig. 

in REW you can also put how many cycles in each tone….. I would just use a rectangular window so all cycles come out at same level. 

so that phenomenon I wonder if it attributes to why a real world impulse response is hardly a needle shape impulse and more a triangle shaped impulse…. The width of the bass in time in our environment (actual air) and mechanical reaction time.  Because in the electrical world it can be that needle shape impulse…. It’s like low frequencies are just slowed down… even if there in time with the HD they just move slower or something….

I bet someone on here (that’s smart) has the answer…..  very interesting for sure 

asx77 wrote:

If you select one cycle it sounds like a click. 

It will, because whilst you might reasonably think one cycle at say 80 Hz only has 80 Hz content, it doesn't. To get the sharp start and stop at the beginning and end of the single cycle you actually need a lot of much higher frequency content too. As you change the number of cycles you change this higher frequency content. Your speakers also won't be perfectly recreating the intended but very challenging single-cycle signal.

I'm afraid I don't believe that the test you've been trying is actually be testing what you want. 

What really matters here is that the first/early arrival part of a very low frequency signal essentially doesn't matter, but rather what does is the result you get once the signal has bounced around the room many times and all the reflections have added up. This is why we hear room modes dominate over the direct signal. The good news is that what is easiest to measure (the steady-state in-room response once you've had all these reflections) is what you need to use to make appropriate corrections and make the sound better. Trying to measure early arrivals in-room at low bass frequencies is basically impossible but fortunately this doesn't really matter.

Hi, the whole thing that got me thinking was the fact that because it takes us a long time to register bass tones, by the time we do the sound has bounced off multiple surfaces. So we're hearing the direct and reflected sound and our brain hears it as the same thing. But there's a point where the direct sound is registered before we start receiving reflections. If reflections are withing a certain time window to the direct sounds then that adds muddle. If they're separated by enough time our brains can interpret them as reflections. 

Just found it interesting. You're probably right about the website. If I could link both I've used I would but they keep getting bounced as spam. 

Anyway, it takes a number of cycles which varies upon frequency  for us to register the tone let's say in either open space or an anechoic chamber. So if we know the number of cycles it should be easy to work out how long to register.

Didn't know about the other frequencies being present in a tone test but yes makes sense they might be there. 

I'll play with REW generator. Didn't realise it could do that. Though not sure I'd make a good Guinea pig. Sample size of me might not withstand scientific scrutiny!

asx77 wrote:

Hi, the whole thing that got me thinking was the fact that because it takes us a long time to register bass tones, by the time we do the sound has bounced off multiple surfaces. So we're hearing the direct and reflected sound and our brain hears it as the same thing. But there's a point where the direct sound is registered before we start receiving reflections. If reflections are withing a certain time window to the direct sounds then that adds muddle. If they're separated by enough time our brains can interpret them as reflections. 

Just found it interesting. You're probably right about the website. If I could link both I've used I would but they keep getting bounced as spam. 

Anyway, it takes a number of cycles which varies upon frequency  for us to register the tone let's say in either open space or an anechoic chamber. So if we know the number of cycles it should be easy to work out how long to register.

I'm not sure what website you're referring to?

Yes in outside spaces what you're asking about become more relevant since there will be few reflections, but then the same is also true of real instruments etc outside, so it's still not really something you'd want to do anything about.

I don’t think we so much wants to do anything about it from the sound of it 

but, would be Interesting to see if it could be plotted…..

you guys are familiar with the geddes talk on this ? 

Hi guys, That's right just interested. Familiar with Geddes approach. Watched a video with him explaining his approach but this vid is pretty good too. It's the one that got me thinking:

Here are links if this works for the other two sites for the tone repetitions and a calculator that tells you periods and other wave info. Just select sound in air and choose your frequency. Remove the spaces:

auditory neuroscience (dot) com (Slash) pitch (Slash) repetitions (hash)

And 

Omni calculator (dot) com (Slash) physics (Slash) frequency (hash)


 You get posts here about American football without problems but I try and post links and it's spam 🙄You'll have to reconstruct them.