DCAudioDIY.com

I was referring to using digital processing, a DAC, and separate amplifiers versus one amplifier channel and a passive crossover, for the same speaker drivers and enclosure. I wish there was a reasonable way to do an A verses B test for a speaker system by switching between a passive and active configuration quickly for the same speakers. The only speakers I know of where there are active and passive versions of the same speakers are some of the Klipsch Heritage models (analog active crossover). I really wish Genelec would come to Washington Audio Fest and demo some of their SAM speakers -

https://www.audiosciencereview.com/foru ... tor.28039/

_________________
Walt

Dealing with prejudice like what Paul shows about DSP is basically a cultural problem. There are many folks out there who are adverse to change so they find excuses to trash new things. The idea that DSP adds noise and distortion is totally wrong with good quality equipment. I am saying that as a result of many years of experience with not so good DSP and excellent DSP like the Danville Signal dspNexus 2/8. And, I often go to live, unamplified music like local concerts in the area as well as the Philadelphia Orchestra who will be having what will probably be a block buster of a performance of Mahler's Second in March.

I have had many people come to listen to my system that is optimized with DSP and there are two common threads. The people are a mix of audiophiles from both the NJ and Philadelphia audio societies, people not involved in HiFi but listen to a lot of live music, musicians, and lay persons. The first thread is that they all say it is the first time they have ever heard the performers right in the room. The second is that they have heard the particular music being played many times in there life but they are hearing things from the recording including nuances by the musicians that they had never heard anywhere before.

So before you make incomplete assumptions about DSP and musical quality, I invite you to come here for a listen. You will be surprised.

As to Walt's question as to why more audiophile speakers do not use active crossovers and DSP, part of the reason is that same culture that rejects change and also the need of many audiophiles to constantly try new tweaks or they feel unfulfilled. Also, they would get absolutely slammed by manufacturers of add on components that would then be at risk. A quick search did show some that are embracing DSP such as Dutch&Dutch, Dyanaudio, Legacy, and Martin Logan. As the old, unmovable audiophiles disappear from the scene, I will bet that more active speakers with DSP will appear. The audible advantages are too great to ignore.

At the end of the day it is your prerogative to take whatever path you want because it is your hobby. But, you don't need to do that by sticking your head in the sand or trashing other, well founded alternatives.

Tom,

DSP can yield good results. In this field, there are no absolutes.

We can be perfectly satisfied our system. But one day, we change on cable, capacitor and immediately we sense the improvement.

Take your experience for example. You were singing praises of your old dac (dont recall the name) for years until one day, Rich demonstrated how dspNexus can improve your system. And you were sold.

The point is, we need a frame of reference make judgements. We cannot claim any one price of equipment as the "best". If Charlie brings his DAC to your home, I bet you would switch the DAC in an instance.

Paul's opinions are the result of his experiences. If you are able to demonstrate to him that DSP can improve his system, I am sure he would change his mind.

Unfortunately, I cannot plug Charlie's DAC into my system because it only has two channels. The Nexus has 8 channels of AKM DACs for my quad amped system. I agree with you that there are no absolutes. That is why it is not productive to claim an absolute problem with some piece of equipment or technology. As far as the Nexus, I can say it is the best processor I have had, but that does not mean there is not something better, but at this point, as with many pieces of equipment, the result is so good that I would guess it would be difficult to get an improvement, especially when you consider not only the acoustic quality, but the extreme degree of corrections available.

There you go. You have developed a bias towards dspNexus and Paul has developed a bias against DSPs. Your bias is no more valid that Paul's.

BTW, I am with Paul on this subject.

Well then, both of you need to come here and make your decision based on listening.

I came across the following quote on the "Audiophiles - America" facebook page today -

“If it can’t be expressed in figures, it is not science, it is opinion.”
Robert Heinlein

Being someone who read "Stranger in a Strange Land" in the 8th grade when it was first published it warmed my heart.
I take it we have moved the conversation on to expectation bias. That is something I don't know how to get rid of as
long as I know what equipment I am listening to. However, I also don't think my ears are particularly golden (they are
imperfect instruments) which is why I am willing to give measurements serious consideration. I also think that it is particularly
telling when I see which speaker brands detailed technical measurements of their products.

Here is how AI replied to that question -

Brands known for transparency
Several manufacturers stand out for their technical transparency, often employing and publishing data from Klippel Near-Field Scanners (NFS), which provide a comprehensive view of a speaker's performance.

Revel: As part of the Harman group, Revel benefits from extensive research into speaker design and listener preferences. It has been noted for publishing detailed CEA-2034 "Spinorama" measurements for many of its speakers,
which provide a complete picture of a speaker's anechoic and in-room performance.

Genelec: Known for high-quality professional studio monitors, Genelec is a highly measurement-oriented company. It openly provides extensive documentation and polar maps that show how its speakers disperse sound.

Ascend Acoustics: This brand is transparent about its engineering process, and has published Klippel NFS measurements for its speakers, demonstrating confidence in its designs.

Neumann: Another brand focused on professional studio monitors, Neumann's products are known for their accuracy and are backed by extensive technical data and measurements.


Other notable brands
Many other companies are highly regarded for their engineering, even if they don't publish the full suite of "Spinorama" data for every product on their website. For some of these brands, measurements are often available from third-party sources.

KEF: Known for its innovative Uni-Q drivers and advanced engineering. The brand has provided detailed white papers and measurements for new product lines.

Focal: A high-end French speaker brand whose products often rely heavily on measurements and engineering. Independent measurements are also available for many Focal products.

Paradigm and PSB: These Canadian brands are noted for performing significant research and measurements in their design process.

ATC: A British manufacturer of studio monitors and hi-fi speakers known for a transparent and balanced presentation. ATC is praised for its honest engineering and provides extensive data on its products.

Brombo,

You are using AI tool as a search engine aggregator. The response any AI tool gives you based upon the questions we ask. Often, such questions are very comprehensive and details.

Since you are giving so much importance to measurements, I slightly changed your question. Here is the interaction:

In audio, why are there often discrepancies between objective measurements and subjective listening impressions?

In audio, there are several well-known areas where objective measurements (things like frequency response, total harmonic distortion, noise levels, etc.) don’t always line up neatly with subjective listening impressions (how people actually perceive sound). Here are the main ones:

1. Frequency Response vs. Perceived Tonality
Measurement: A flat frequency response is often considered “ideal.”
Listening: Many listeners actually prefer a slightly “colored” response — for example, a gentle bass boost (the “loudness” curve) or a warmer midrange. Our ears are not equally sensitive across all frequencies (see the Fletcher–Munson curves), so a ruler-flat line may not sound natural.

2. Distortion (THD) vs. Musical Pleasantness
Measurement: Total Harmonic Distortion (THD) quantifies added harmonics as a percentage. Lower is “better” on paper.
Listening: Some types of distortion (especially low-order harmonics, like 2nd and 3rd) can sound warm, rich, or even pleasing (e.g., tube amplifiers). Meanwhile, small amounts of high-order distortion may measure “better” but sound harsher.

3. Dynamic Range & Compression
Measurement: Greater dynamic range is objectively better in preserving quiet-to-loud detail.
Listening: In casual environments (cars, earbuds, background listening), heavily compressed audio often “sounds better” because quiet passages don’t get lost. This is one reason behind the so-called loudness wars.

4. Noise Floor
Measurement: Signal-to-Noise Ratio (SNR) and noise floor are precise numbers.
Listening: Once noise is below a certain threshold (e.g., -80 dB), most people can’t perceive it. Extra improvements in measurement (-110 dB vs. -120 dB) may have no audible difference.

5. Specs vs. Real-World Listening
Measurement: Specs like “20 Hz – 20 kHz ±0.5 dB” or “0.001% THD” suggest perfection.
Listening: Room acoustics, speaker placement, psychoacoustics, and listener bias play a much larger role in the actual impression than these lab numbers.

6. Phase and Time Alignment
Measurement: Phase distortion, group delay, or impulse response irregularities can be measured.
Listening: Many listeners won’t notice unless the issues are severe, while others may describe subtle timing errors as a “lack of coherence” or “smear” even when distortion figures look fine.

7. Subjectivity and Expectation Bias
Measurement: Two devices with nearly identical measured performance may measure “transparently.”
Listening: People’s expectations, aesthetics (e.g., tube glow, brand reputation), and even sighted vs. blind testing can radically change impressions.

✅ Bottom line: Objective metrics are crucial for engineering and consistency, but human perception is non-linear, context-dependent, and shaped by bias. That’s why high-end audio often debates measurements vs. “what sounds good.”

There is a big difference between preferring alterations to the sound on the recorded medium and stating that DSP increases noise and distortion. You have to go with what is pleasing to you regardless of the accuracy. But to denigrate a technology with false claims because it doesn't please you is, in my opinion, going down the rabbit hole.