I asked ChatGPT (AI) if cables matter…

[GrubTheHedgehog]

 

I asked an AI chatbot if cables can make a difference. I got a pretty good answer considering how esoteric this topic is!

 

Q: Can different cables make a stereo system sound better?

 

Yes, different cables can make a stereo system sound better, but the extent to which they make a difference is a matter of debate among audiophiles.

Some people believe that high-quality cables with better shielding and materials can reduce interference and improve signal transmission, resulting in a clearer and more accurate sound. Others argue that the difference is negligible or even non-existent, and that expensive cables are nothing more than a marketing gimmick.

It's worth noting that there are many factors that can affect the sound quality of a stereo system, including the quality of the components, the room acoustics, and the listener's preferences. So even if different cables do make a difference, it may not be the most significant factor in improving the overall sound quality of a stereo system.

Ultimately, whether or not different cables make a difference is subjective and dependent on individual preferences and perceptions. The best way to determine if different cables make a noticeable difference in your stereo system is to try them out and see if you can hear a difference that is worth the investment.

[botrytis]

ChatGTP is only as good as the information put into it. If bias information is loaded in, then it will be biased. 

 

Basically, garbage in garbage out with current AI.

[MarcelNL]

Agree, this sounds as if it's regurgitating the mean of the info it has access to, hardly intelligent IMO.

You should ask ChatGTP how you make the best sounding cables and try them...

[botrytis]

5 minutes ago, MarcelNL said:

Agree, this sounds as if it's regurgitating the mean of the info it has access to, hardly intelligent IMO.

You should ask ChatGTP how you make the best sounding cables and try them...

 

Stellar idea.

 

People need to realize - OpenAI's ChatGPT Powered by Human Contractors Paid $15 Per Hour (gizmodo.com)

 

This is how ChatGTP works, still human have to tell it what to scan, there is bias there.....

[DuckToller]

32 minutes ago, GrubTheHedgehog said:

and see if you can hear a difference

Obviously chicken shit  - oops - language model 😇

[DuckToller]

Even I have got a more than satisfying answer on MQA from ChatGPT, I find this tool unusuable in relation to technology, the audio hobby and journalism.
What the developer call halluzination, the permanent attempt of AI to provide "factual information" from sources that doesn't exist on a probability bet is simply annoying. means inventing inexistant subpages on homepages that exist.
And the limited understanding about the imformation it provides.
You can ask i.E. about as description of an open back dynamic headphones, and it gives you a correct description. Ask about  10 of such headphones in a certain price bracket, AI will provide you then 10 headphones, 3 of them dynamic, 4 of them open back, 5 of them in the price bracket.  sometimes you will have double entries. Go figure ..

[botrytis]

AI can only make good decisions and explanations if ALL the information about a topic is scanned in. If not, it cannot make a valid answer. The cable explanation is obviously one of them.

 

This is why Facial Recognition software has the issues it does. It is built in bias from the programmers.

[GrubTheHedgehog]

I don’t understand how this is not impressive to you guys. It gave a clear and concise answer to a very esoteric question. I think it accurately presented a very concise summary of the arguments that are made on both sides of the issue. Then it suggested adopting an empirical approach since the question cannot be resolved objectively. It cut right through that question.

[MarcelNL]

4 minutes ago, GrubTheHedgehog said:

I don’t understand how this is not impressive to you guys. It gave a clear and concise answer to a very esoteric question. I think it accurately presented a very concise summary of the arguments that are made on both sides of the issue. Then it suggested adopting an empirical approach since the question cannot be resolved objectively. It cut right through that question.

it does, but regurgitating information is something vastly different than intelligence, with intelligence I'd expect that system to come back with new ideas rather than coughing up a synopsis of existing information.

A database will keep all the info, a query allows a selection to be drawn, does it become intelligence when the query is adaptable?

[stefano_mbp]

… it’s not A.I. but only a quite sofisticate (?) search engine , in other words it can just help people who isn’t able to search on its own

[Archimago]

27 minutes ago, MarcelNL said:

it does, but regurgitating information is something vastly different than intelligence, with intelligence I'd expect that system to come back with new ideas rather than coughing up a synopsis of existing information.

A database will keep all the info, a query allows a selection to be drawn, does it become intelligence when the query is adaptable?

 

Let's be honest. Most debates about cables involve audiophiles arguing about what they heard said by fellow audiophiles, what manufacturers said, and what they supposedly subjectively "heard" in an uncontrolled fashion. How much intelligence is there in the usual arguments and bickering?

 

IMO, the most intelligent of us will actually bother to run measurements and comparisons. Try our best to figure out parameters like LCR of speaker cables. Try our best to detect differences in the DAC output when using different digital cables, etc. Find the data while of course also having a listen with A/B comparisons. (a plug 😉).

 

I agree with @GrubTheHedgehog, ChatGPT is impressive. Here's what it told be when I asked the question in another way (as discussed here):

 

Quote

 

Arch: What is the current understanding of the value of high priced audiophile cables more than $1000? Is there enough evidence that they sound different to be worth putting much money into purchasing them?

 

ChatGPT: The current scientific understanding is that there is no evidence to support the claim that high-priced audiophile cables, costing more than $1000, can improve the sound quality of an audio system beyond what can be achieved with well-designed, standard cables.

 

Audiophile cables are often marketed with claims that they can improve the sound quality of an audio system by reducing distortion, noise, or electromagnetic interference. However, these claims are not supported by scientific studies or measurements. There is no evidence to suggest that expensive cables can provide any audible benefits over properly designed and constructed cables.

 

Some studies have been conducted to test the claim that high-priced audiophile cables can improve sound quality, but they have consistently failed to find any significant difference between the sound of expensive cables and that of well-designed standard cables. In fact, blind listening tests have shown that even trained listeners are unable to distinguish between different types of cables when they are not aware of which cable they are listening to.

 

In conclusion, based on current scientific understanding, it is not worth spending large amounts of money on high-priced audiophile cables in the hope of improving the sound quality of an audio system. Well-designed, standard cables are likely to be sufficient for most users.

 

 

That answer certainly sounds more "intelligent", more nuanced, and more incisive at least than many responses I've seen over the years!

 

Of course ChatGPT cannot do its own research and depends on what was fed into it. But at least it looks like it was fed with enough information to synthesize a reasonable answer to a question like this.

[Blackstone]

.

[stefano_mbp]

… but it’s not A.I.  … which is the definition of Intelligence?

[sphinxsix]

1 hour ago, botrytis said:

ChatGTP is only as good as the information put into it. If bias information is loaded in, then it will be biased. 

 

Basically, garbage in garbage out with current AI.

 

IMO that's a highly biased post.

[sphinxsix]

1 hour ago, botrytis said:

People need to realize - OpenAI's ChatGPT Powered by Human Contractors Paid $15 Per Hour (gizmodo.com)

 

This is how ChatGTP works, still human have to tell it what to scan, there is bias there.....

 

Just one more biased post ;-) 

[botrytis]

It is not - it is fact. One that is based on studies of Facial Recognition Software (and this is also AI software BTW) and how they were developed.

 

Report finds racial bias in facial recognition technology - CSMonitor.com

 

Racial Discrimination in Face Recognition Technology - Science in the News (harvard.edu)

 

facial recognition software racial bias - Google Scholar

 

How is Face Recognition Surveillance Technology Racist? | ACLU

 

(I have plenty more on this topic - I do RESEARCH for a job!)

 

Bias is nothing new.

 

I am not as 'GOO GOO EYES' on AI as others are - more realistic as if there is misinformation in AI, people will take it as truth, without doing any research on their own.

 

Even the CEO of company that developed Chat GTP says it is limited and is a dead end.

[sphinxsix]

1 hour ago, MarcelNL said:

it does, but regurgitating information is something vastly different than intelligence, with intelligence I'd expect that system to come back with new ideas rather than coughing up a synopsis of existing information.

 

Some say AGI may be a year or two away (and these are claims by highly professional guys from the AI field).

 

[sphinxsix]

1 hour ago, GrubTheHedgehog said:

I don’t understand how this is not impressive to you guys.

 

You're dealing here mostly with old-school guys who can not even use their smartphone, what do you expect - they are simply jealous..

 

😉

[sphinxsix]

35 minutes ago, stefano_mbp said:

… it’s not A.I. but only a quite sofisticate (?) search engine , in other words it can just help people who isn’t able to search on its own

 

At the same time its English (both understanding and writing) is IMO impressively good!

 

[GrubTheHedgehog]

Exactly. It is a sophisticated search engine. It does not need to be "intelligent" to be useful and game changing. You have to evaluate it against existing tools, like Google.

 

If I want to know the spec differences between two speakers, I need to go through these steps: (1) go to manufacturer's website; (2) pull the spec sheets for both speakers; (3) compare the two specs; (4) determine which specs are relevant.

 

Here is how ChatGPT responded to a question about two speakers:

 

Q: What is the difference between a pair of bowers and Wilkins 802 D3 speakers, and a pair of bowers and Wilkins 802 D4 speakers?

 

ChatGPT:
The Bowers & Wilkins 802 D3 and 802 D4 speakers are both part of the flagship 800 Series Diamond range, and they share many similarities in terms of their design and engineering. However, there are some key differences between the two models.

One of the most significant differences is the use of an updated driver technology in the 802 D4. The 802 D4 features a new midrange driver that uses a biomimetic suspension system, which is designed to improve accuracy and reduce distortion. The 802 D4 also features a new Continuum cone bass driver, which is designed to deliver improved low-frequency performance.

Another difference is the cabinet design. The 802 D4 features a new reverse-wrap cabinet design, which is said to improve the stiffness of the cabinet and reduce resonance. This new design is also visually distinct from the 802 D3, with a more sculpted and streamlined look.

In terms of specifications, the 802 D4 has a slightly wider frequency response range than the 802 D3, with a claimed range of 14Hz to 35kHz compared to 16Hz to 35kHz for the 802 D3. The 802 D4 also has a higher sensitivity rating of 90dB compared to 90dB for the 802 D3.

Overall, while the 802 D3 and 802 D4 are both high-end speakers from Bowers & Wilkins, the 802 D4 represents a significant update over the previous model, with improvements in driver technology, cabinet design, and overall performance.

 

Q: Further explain the differences in specifications between the two speakers.

 

ChatGPT:
Sure, let me explain the differences in specifications between the Bowers & Wilkins 802 D3 and 802 D4 speakers in more detail:

Frequency response: The frequency response range of a speaker indicates the range of frequencies that it can reproduce accurately. The Bowers & Wilkins 802 D3 has a frequency response range of 16Hz to 35kHz, while the 802 D4 has a wider frequency response range of 14Hz to 35kHz. This means that the 802 D4 can reproduce slightly lower frequencies than the 802 D3, which may result in slightly more detailed and extended bass response.

Sensitivity: The sensitivity rating of a speaker indicates how efficiently it converts power into sound. The higher the sensitivity rating, the louder the speaker can play with the same amount of power input. The Bowers & Wilkins 802 D4 has a higher sensitivity rating of 90dB compared to the 802 D3, which also means that it can play louder with the same amount of power input.

Impedance: The impedance of a speaker indicates the amount of electrical resistance it offers to the amplifier that is driving it. The Bowers & Wilkins 802 D3 and 802 D4 both have a nominal impedance of 8 ohms, which means they can be driven by most amplifiers.

Recommended amplifier power: The recommended amplifier power rating of a speaker indicates the range of power output that is best suited to drive the speaker. The Bowers & Wilkins 802 D3 and 802 D4 both have a recommended amplifier power range of 50W to 500W, which means they can be driven by a wide range of amplifiers, from low-power tube amps to high-power solid-state amps.

Dimensions and weight: The Bowers & Wilkins 802 D4 is slightly larger and heavier than the 802 D3. The 802 D4 measures 1203mm x 390mm x 583mm (47.4in x 15.4in x 23in) and weighs 86kg (190lb), while the 802 D3 measures 1165mm x 390mm x 583mm (45.9in x 15.4in x 23in) and weighs 79.5kg (175lb).

Overall, while the differences in specifications between the Bowers & Wilkins 802 D3 and 802 D4 may seem minor, they can add up to a significant improvement in overall performance, particularly in terms of frequency response and sensitivity.

 

 

The amount of time I expended to generate this information about the two speakers using chat GPT was probably less than one minute, total. What other tool in the world is capable of doing something like this? Of course, this needs to be fact checked, but even if all it did was pull information from the manufacturer's website, the fact that it was able to do this comparison for me in such little time is simply astounding to me.

 

Also remember that a lot of us want to make these types of inquiries on cell phones, not computers with keyboards. So how long would it take me to generate this kind of information on a cell phone if this bot didn't do the heavy lifting for me? I think the whole thing is pretty incredible!

[GrubTheHedgehog]

7 minutes ago, sphinxsix said:

 

At the same time its English (both understanding and writing) is IMO impressively good!

 

Exactly. It is the way it presents the information that is so impressive. I don't need a computer to do my thinking for me, but if it wants to do some research and note taking for me, I'll take it!

[sphinxsix]

31 minutes ago, Archimago said:

Let's be honest. Most debates about cables involve audiophiles arguing about what they heard said by fellow audiophiles, what manufacturers said, and what they supposedly subjectively "heard" in an uncontrolled fashion.

I've been repeating this for years - IMO experience as an ex dealer ca 75% of customers could hear differences between cables, regardless of a system quality and its price range (begining with a basic 'audiophile' level eg cheap NAD gear).

Since I read exactly the same estimate quoted by Steve Guttenberg ('Stereophile') based on his dealer experience, I'm personally convinced this is not an accident.

It's quite difficult to say to someone - there is something wrong with either your ears or your system..

 

37 minutes ago, Archimago said:

That answer certainly sounds more "intelligent", more nuanced, and more incisive at least than many responses I've seen over the years!

I had the same thought when I asked it about the rules of shaping a listening room acoustics - a I got a better, more complete reply than most that I've seen on audio forums for years.

 

39 minutes ago, Archimago said:

Of course ChatGPT cannot do its own research and depends on what was fed into it.

In other words it can only get as good as we, humans are.

 

The bottom line - let's wait for AGI  😎

 

[sphinxsix]

2 minutes ago, GrubTheHedgehog said:

Exactly. It is the way it presents the information that is so impressive. I don't need a computer to do my thinking for me, but if it wants to do some research and note taking for me, I'll take it!

Haven't encountered a situation (yet?) in which this research done by GPT would be obviously wrong.

[GrubTheHedgehog]

31 minutes ago, sphinxsix said:

Haven't encountered a situation (yet?) in which this research done by GPT would be obviously wrong.

 

It is can be VERY wrong. I asked it to give me a summary of specific legal precedent. Not only did it utterly fail to give me an accurate summary of the facts of a legal case, even though I gave it a citation, it gave me wrong information in a format that made it look like it really knew what it was talking about. The facts were wrong. The result that was reached was wrong. It seemed to pull facts in from other unrelated cases and inject them into its analysis.  It could not have been more useless as a research tool. But you can kind of see how if this thing were improved, it could be a useful tool. 

 

[Iving]

chat GPT's currency is language. Its I/O is language. What goes on in "the black box" between I and O is at the heart of the matter.

 

Language is species-specific ... A respectable philosophical position holds that non-humans cannot process/produce language (including say chimps demonstrating ASL) except by (Pavlovian/Skinnerian) conditioning. All "intelligent" behaviour in non-humans actually comprises conditioned responses/behaviour. What's more, there is no consciousness behind that behaviour.

 

I dare suggest that chat GPT is much more "non-human" than human - wrt language underpinning consciousness - and with it experience of "pain" as we know it - and appreciation of e.g.s aesthetics, mathematics and music.

 

Whether humans should be frightened of chat GPT/AI anyway is another matter. Perhaps the "intelligence" associated with AI in the future will have a quality of its own - neither better nor worse than humans - just different.

 

I've always thought that if digits get too much "in your face", we can just unplug at the wall. That's what I do with my Hi-Fi.

 

Even cats can't produce electricity.