Maximum subjectively preferred loudness level

[sarumbear]

I hope not - hearing loss possible in two minutes at those levels.

I agree, maybe the harm happens at a bit longer than two minutes but that’s the state of live events, whether you hope or not.

 

[No. 5]

ChatGPT
...In dry air at room temperature (around 20 degrees Celsius), sound waves at 1 kHz will attenuate by approximately 0.3 dB per meter, while sound waves at 10 kHz will attenuate by approximately 1.2 dB per meter.


1,2dB is a lot
Imagine at 3m

I question ChatGPT's numbers. Can you imagine what the back row of any live event would sound like?

The reference works on my bookshelf show 2kHz attenuating by around 1.5dB at 30 meters with 10% relative humidity (or 0.05dB/m). At 50% RH, it's less than 0.25dB of attenuation at 30 meters (0.008dB/m). 10kHz is attenuated by less than 4dB at 30 meters with 50% RH.

 

[Axo1989]

I agree, maybe the harm happens at a bit longer than two minutes but that’s the state of live events, whether you hope or not.

Yeah peak levels (or even slow averages) can be deceptive, you really need to run the dosimeter to be sure ... that said, earplugs can be a good idea to have handy for live stuff: you get the body sensation without the hearing damage.

 

[Keith_W]

May I suggest that the OP adds a poll to this thread?

 

[theREALdotnet]

ChatGPT

The speed at which sound attenuates in air depends on various factors such as frequency, distance traveled, temperature, humidity, and the presence of obstacles. Generally, high-frequency sounds are more attenuated by air than low-frequency sounds. In dry air at room temperature (around 20 degrees Celsius), sound waves at 1 kHz will attenuate by approximately 0.3 dB per meter, while sound waves at 10 kHz will attenuate by approximately 1.2 dB per meter.

Can you ask ChatGPT what the absorption in air is according to ISO 9613? It depends on pressure, humidity and temperature, but for 1kHz it’s more like 0.005dB/m, or 0.2dB/m for 10kHz (20ºC and 40% humidity).

Calculation of absorption of sound by the atmosphere

 

[antcollinet]

I play the congas and it always suprised me how diferent they sound when you play them, as opposed to when you are the listener.

ChatGPT



1,2dB is a lot
Imagine at 3m

Chat GPT is not a reference work or encylopedia. It simply regurgitates what it has been trained on. We have no idea of the veracity of that data set.

We do know from what it says about "high end" audio kit, that there is a lot of audiophile folklore in there. That alone confirms that it is not reliable for giving factual information.

 

[sarumbear]

I play the congas and it always suprised me how diferent they sound when you play them, as opposed to when you are the listener.

ChatGPT



1,2dB is a lot
Imagine at 3m

The wiki scientists transformed into ChatGBP scientists...

There is less than 1dB attenuation in domestic environments at 10kHz.

 

[sarumbear]

May I suggest that the OP adds a poll to this thread?

How loud do you like your music?

I'm interested in how loud you enjoy your music. For this purpose I propose A weighted, slow SPL measuring. I'm also proposing a clean, well produced song that is not bass heavy to avoid subjective impressions of low frequencies (Hey Nineteen - Steely Dan). Since music material varies in dB, I...

www.audiosciencereview.com

 

[dasdoing]

@people_which_have_replied_to_my_chatGpt_post

I am sorry, I haven't questioned it enough. though it is wrong "it simply regurgitates what it has been trained on", sometimes it just doesn't dig enough if a question hasn't been done before.

here is a new dialog where I tried to dig a little deeper

Spoiler: dig

absorption of soundin air



The absorption of sound in air is the process by which sound energy is gradually converted into heat as it travels through the air. This occurs due to the air molecules' frictional resistance to the sound waves, which causes the sound waves to lose energy and become quieter as they propagate through the air.
The amount of sound absorption that occurs in air depends on a number of factors, including the frequency of the sound, the distance the sound has traveled, and the temperature and humidity of the air. In general, high-frequency sounds are more easily absorbed by air than low-frequency sounds, and the absorption increases as the sound travels a greater distance through the air.
Additionally, higher temperatures and humidity levels can increase the absorption of sound in air, while lower temperatures and drier air can decrease it. This is because water molecules in the air can absorb sound energy more effectively than dry air molecules.
Overall, the absorption of sound in air is an important factor to consider when designing and evaluating acoustic environments, such as concert halls, recording studios, and other spaces where sound quality is critical.

numbers



Numbers are mathematical symbols that represent quantities or values. They are used for counting, measuring,

absorbtion of sound in numbers



The absorption of sound in air is typically measured in decibels (dB) per unit distance traveled by the sound wave. This measurement is called the "attenuation coefficient" and represents the amount by which the sound energy is reduced as it travels through the air.
The attenuation coefficient of sound in air varies depending on the frequency of the sound, the distance traveled, and the temperature and humidity of the air. In general, high-frequency sounds are more easily absorbed by air than low-frequency sounds.
At a temperature of 20°C and a relative humidity of 50%, the attenuation coefficient of sound in air is approximately:

• 0.16 dB/meter for a frequency of 1000 Hz
• 0.05 dB/meter for a frequency of 2000 Hz
• 0.02 dB/meter for a frequency of 4000 Hz

These values indicate that high-frequency sounds are more easily absorbed by air than low-frequency sounds, with the attenuation coefficient decreasing as the frequency increases. It's important to note that these values are approximate and can vary depending on the specific conditions of the air, such as temperature, humidity, and air pressure.

where are these values from?



These values for the attenuation coefficient of sound in air at a temperature of 20°C and a relative humidity of 50% are based on empirical measurements and are commonly used as reference values for acoustic calculations and design. These values are derived from the ISO 9613-1 standard, which provides guidelines for the calculation of sound attenuation in outdoor environments.
It's important to note that these values are approximate and can vary depending on the specific conditions of the air, such as temperature, humidity, and air pressure. Additionally, the attenuation coefficient can be influenced by other factors such as the presence of obstacles, wind, and atmospheric conditions. Therefore, it's important to consider a range of factors when designing and evaluating acoustic environments.

can you include studies



Certainly! There have been many studies on the absorption of sound in air, which have helped to develop our understanding of this process and inform acoustic design. Here are a few examples of relevant studies:

• "Attenuation of Sound in the Atmosphere," by Robert E. Rand and David R. Stine, published in the Journal of the Acoustical Society of America in 1963. This study measured the attenuation coefficient of sound in the atmosphere over a range of frequencies and distances, and provided a theoretical model for predicting sound attenuation in outdoor environments.
• "Effects of Humidity and Temperature on Sound Attenuation in the Atmosphere," by E. A. Power and K. R. Kube, published in the Journal of the Acoustical Society of America in 1981. This study measured the attenuation coefficient of sound in air at varying levels of temperature and humidity, and found that higher humidity levels led to increased sound absorption.
• "The Absorption of Sound in Atmosphere Containing Water Droplets," by M. A. Biot, published in the Journal of the Acoustical Society of America in 1956. This study examined the effect of water droplets in the air on sound absorption, and provided a theoretical model for predicting sound attenuation in foggy or rainy conditions.
• "Measurements and Models of Outdoor Sound Propagation," by J. Paul Guyer, published in the Journal of Applied Mechanics and Materials in 2012. This study reviewed and synthesized existing research on outdoor sound propagation, including studies on sound attenuation in air, and provided guidance for acoustic design in outdoor environments.

These studies, along with many others, have contributed to our understanding of the absorption of sound in air and have helped to inform the development of acoustic design principles and practices.

so including those studies what are the numbers?



Based on the studies I mentioned, here are some general ranges for the attenuation coefficient of sound in air:

• For frequencies around 1000 Hz, the attenuation coefficient is typically between 0.10 and 0.20 dB/meter.
• For frequencies around 2000 Hz, the attenuation coefficient is typically between 0.03 and 0.08 dB/meter.
• For frequencies around 4000 Hz, the attenuation coefficient is typically between 0.01 and 0.03 dB/meter.

It's important to note that these ranges are approximate and can vary depending on a number of factors, including the temperature, humidity, and pressure of the air, as well as the distance traveled by the sound wave and the presence of obstacles or other environmental factors. Additionally, these values are specific to outdoor environments and may not be applicable in all situations.
Overall, the attenuation coefficient of sound in air is an important factor to consider when designing and evaluating acoustic environments, and should be taken into account alongside other factors such as the frequency range and volume of the sound source, as well as the characteristics of the space and its occupants.

 

[sarumbear]

"it simply regurgitates what it has been trained on"

In other words it cannot be used as a reference. There are many uses for AI but reference data is not one of them.

 

[dasdoing]

In other words it cannot be used as a reference. There are many uses for AI but reference data is not one of them.

it analised the whole internet until 2021 afaik. it just isn't ready yet to give perfect answers on the fly. the data is there. also I am on a free acount

 

[sarumbear]

it analised the whole internet until 2021 afaik. it just isn't ready yet to give perfect answers on the fly. the data is there. also I am on a free acount

I didn’t know that absorption of air was researched after 2021…

It is ready to give answers on the fly and does it with huge accuracy but not suitable for reference. Ask for software code, write from scratch or debug. Ask for processes and procedures. Ask about laws and legal issues.

 

[Philbo King]

Live venues are often are at 110-115dBSPL.

65dBSPL is a very low level. It’s almost the same as produced by air conditioning or conversation. It will be difficult to hear everything.

Yes... Earplugs. Long gone are the days I trust my hearing to a drunken (or hiphop-rave-trained) sound man. If it's loud enough to cause fatigue (>95 dB) I vote with my feet, and leave.

I have a very quiet recording studio. Around 34 dBZ SPL when HVAC isn't running (and the wife isn't blasting the TV upstairs).

 

[dasdoing]

I didn’t know that absorption of air was researched after 2021…

wtf? you give some very weird replies sometimes

 

[sarumbear]

wtf? you give some very weird replies sometimes

Your excuse for ChatGBT’s error was “it analised the whole internet until 2021.” It implied that it didn’t know the reference as it was not established before 2021. Otherwise why would you give a date?

 

[antcollinet]

I didn’t know that absorption of air was researched after 2021…

It is ready to give answers on the fly and does it with huge accuracy but not suitable for reference. Ask for software code, write from scratch or debug. Ask for processes and procedures. Ask about laws and legal issues.

How do we know the answers given are correct? Rather than just based on incorrect training data sets?

 

[dasdoing]

Your excuse for ChatGBT’s error was “it analised the whole internet until 2021.” It implied that it didn’t know the reference as it was not established before 2021. Otherwise why would you give a date?

ok, I might epxressed myself wrongly. I said it doesn't always analise all it's data on the first question. it's understandable, since it's a lot of data. after you make it dig deeper, it seams to ajust and give better replies to the same question

 

[sarumbear]

How do we know the answers given are correct? Rather than just based on incorrect training data sets?

When the training data set are the study books, ChatGBT gives you the answers that are accepted as correct. For example: It passes the bar and Harvard MBA exams and accepted as level 3 programmer at Google.

 

[Axo1989]

it analised the whole internet until 2021 afaik. it just isn't ready yet to give perfect answers on the fly. the data is there. also I am on a free acount

We covered this somewhat in another thread.

If you want a better understanding of how this large language model is assembled and trained this article on jill/txt is a good start. As you can see not exactly the whole internet, but a specific large language corpus. Also ChatGPT modifies its responses according to prompts for the duration of your chat session, but I don't think it's true to say it "learns" cumulatively based on everyone's chat sessions. The training process follows by analysing user feedback (you would have seen the dialog and likely responded yourself) which can be built into subsequent model iterations.

As for probabilistic text versus reference accuracy, answers depend on the prompt: @sonitus mirus gave it a bash on GPT-4 and we see an inaccurate response to an initial prompt, followed by corrections within the chat session. I don't have an account (they appeared to want a phone number etc) but you could try that initial prompt

Was there a large meteorite impact at Lac Wiyashakimi in Canada?

and see if a subsequent probabilistic answer differs.

The efficacy of providing an incorrect answer to a straightforward factual yes/no question, but requiring subsequent prompts in-session certainly escapes me.

 

[antcollinet]

When the training data set are the study books, ChatGBT gives you the answers that are accepted as correct. For example: It passes the bar and Harvard MBA exams and accepted as level 3 programmer at Google.

Right - but for any given question - we don't know what the training data is. So we cant' take what it says as correct.

Perhaps when it starts citing references we'll then be able to check.