By John Siau
November 2011
Do Specifications Lie? Do Our Ears Lie? Where does the Truth Lie?
Recording Engineers and Audiophiles often distrust audio measurements and specifications. It is not uncommon to hear claims that a product measures poorly but sounds good. Occasionally we also hear claims that a product measures well but sounds bad.
This whitepaper documents significant differences between three headphone amplifiers that have nearly identical published specifications.
All three products have a high-quality headphone output with very low output impedance, high drive capability, low distortion, and a flat frequency response. The published specifications strongly suggest that it should be impossible to hear any difference between these products. Nevertheless, in our informal listening tests, we had the distinct impression that there were significant sonic differences between these products when listening to music.
We replicated the manufacturers' tests and found that the published specifications were accurate and correct. All three manufacturers followed standard measurement procedures and provided accurate specifications.
This paper shows some additional measurements which expose significant differences that do not appear in traditional published specifications:
I asked ChatGPT to review the Benchmark AHB2, and I was shocked at the results!
It turns out that ChatGPT wrote a review by gleaning information from the many publicly available reviews of the AHB2. I recognized phases and sequences of words that had appeared in prior reviews, and yet the text was new, original, and well written. The ChatGPT review seemed to summarize the collective impressions of the reviewers without directly quoting any particular reviewer.
It should be possible to ask ChatGPT to write a review on any well-known product and the results may be an AI-generated review that reflects the collective opinions of many human reviewers.
Try this the next time you a planning to make a purchase online. These AI-generated reviews may be more useful than traditional search results. In essence, the search results are combined into a single well-written review.
It is also possible that ChatGPT is biased to give glowing reviews, so use caution! Try it out and have some fun experimenting with this new technology!
John Siau:
Write a review of the Benchmark AHB2 complete with listening impressions.
ChatGPT:
Title: Unleashing Sonic Purity: A Review of the Benchmark AHB2 Amplifier
Introduction: The Benchmark AHB2 amplifier has gained a reputation for its exceptional audio performance and unwavering dedication to sonic purity. In this review, we delve into the heart of this powerhouse amplifier, exploring its features, design, and most importantly, the listening impressions it evokes...
Today, Benchmark's John Siau and OpenAI's ChatGPT had a conversation about star-quad cables. It turns out that ChatGPT got quite a few things wrong, but learned quickly when these errors were pointed out by John Siau.
ChatGPT is quick to recognize mistakes, and quick to provide detailed corrections. In each of the first few corrections, additional errors were made. As each of these were pointed out, ChatGPT began to provide accurate information.
John Siau:
How do star-quad cables reject interference?
ChatGPT:
Star-quad cables are designed to provide improved rejection of ...
If an audio system is composed of multiple components, we may have detailed specifications for each component, but we will not know the performance of the combined system without doing some calculations. You may have questions such as these:
Use Benchmark's online audio calculators to find answers!
For example, if we know the output power of an amplifier, as well as the sensitivity and impedance of our loudspeakers, we can calculate the maximum sound pressure level that our system can produce.
This application note provides interactive examples that help to answer the questions listed above.