Spintricity - Towards a Hybrid Subjective Objective Review Process

[This is an ongoing research project. When complete, we will publish a checklist that people can use which will have a place to enter a score for each sound attribute described below].

As everyone knows, subjective versus objective reviews are a bunch of distracting B.S. Both have failed miserably.

Reviewers using objective approaches are starved for things to measure, and subjective approaches are abused and become dependent on the personal preferences and moodiness of the reviewer.

We propose a hybrid approach. A disciplined subjective approach.

The goal is for reviews, in aggregate, to be able to

1. communicate, with great fidelity, what the component really sounds like to a person who has not heard the component
2. communicate, with great fidelity, the strengths and weaknesses of a component so a person looking to adjust their current system in a specific manner can determine which components to replace and with what

In the disciplined subjective approach:

A. various STANDARDIZED [potentially eventually measurable] qualities are observed for the component under review.
B. These qualities are then rated and optionally described by a conscientious and competent listener.
C. These results are then normalized across the many reviews of the component.

We now describe what we think the STANDARDIZED qualities should be.

The Qualities of Sound of a High-end Audio Component

These qualities, when judged by a competent, conscientious listener, will result in the sound of a component being quantified, making it possible for others to understand the sound of the component.

This is a lot of work on the part of a listener, but it is a lot easier, and cheaper, than actually measuring these qualities with scientific instruments [should such ever exist].

Dynamic Response

The component's dynamic response behavior should be described. We will here define the accuracy of the dynamic response: a combination of the leading edge and decay of a note.

This is important because many if not most people associate dynamic response with 'liveness' and 'engagement'.

J.A. at Stereophile often shows a graph of the dynamic response of an impulse at a single frequency and also the decays across all frequencies in a waterfall plot.

These graphs are hard to translate into what we hear. They do not account for the unnatural too fast/too slow rise and falls of notes. In other words, how a component response to notes that are SUPPOSED to decay slowly and notes that are SUPPOSED to rise slowly. Similarly with notes that start rising slowly then quicken their pace and then slow down again, etc.

So what we want a reviewer to describe then is the quality of dynamic response in the lower treble, midrange, upper bass, and bass

Rate perceived accuracy of dynamic response [1 to 10, 10 being 100% accurate] and describe any major issues
1. Dynamic Response (bass. Rate both macro- and midi-dynamics)
2. Dynamic Response (upper bass. Rate both macro- and midi-dynamics)
3. Dynamic Response (midrange. Rate all macro-, midi- and micro-dynamics)
4. Dynamic Response (lower treble. Rate both and midi- and micro-dynamics)

Linearity (evenness) of Dynamic Response

Another important aspect of a component's performance is how consistent its dynamic response is.

This is important because the ear perceives it as unnatural for notes in some frequencies to sound and 'quick' and 'light on their feet' and in other frequencies to be 'muddy and slow'.

We want to know how the response varies between low notes and high notes (frequencies), between loud notes and soft notes (SPL). We also want to know how it behaves across different frequencies and SPLs as the music becomes more complex.

This is a very complex multidimensional thing to describe in detail, but some general information can be communicated quite easily.

Rate perceived linearity of dynamic response [1 to 10, 10 being 100% consistent] and describe any major non-linearities
1. Linearity with respect to different frequencies
2. Linearity with respect to different SPL
3. Linearity with respect to different complexity

Harmonic Response

The Harmonic qualities of a component should be described.

This is important because this is typically related to the 'warmth', and 'neutral'ness, as well as 'engagement', of a component. Most musicians struggle to be able to own and play an instrument that has exquisite harmonic color - from violins to electric guitars to grand pianos.

This is very hard to quantify. The differences between, say, a 300B tube and a 211 tube, or between different amps using the 833 tube, or between 'warm' solid-state and 'dry' solid-state make it difficult sometimes to rate.

Rate perceived harmonic richness [5 being real, 10 being very rich and 1 being very dry] and describe any major issues

1. Harmonic richness (bass)
2. Harmonic richness (midrange)
3. Harmonic richness (treble)

Linearity (evenness) of Harmonic Response

Across the midrange, components handle harmonics at some frequencies better than others. Although the midrange is by far the most important, as usual, consistency is important to lend a sense of realness and help the listener suspend disbelief [even thought the real world, and the ears, are not consistent, our brain has learned to compensate for them. Making it also compensate for a non-linear component is not what high-end audio is all about].

Example: Some components render 'beautiful highs' for, say, a flute but lack color when it comes to lower frequencies like that from the body of a guitar.

Rate perceived linearity of harmonics [1 to 10, 10 being 100% consistent] and describe any major non-linearities
1. Linearity with respect to different frequencies
2. Linearity with respect to different SPL
3. Linearity with respect to different complexity

Frequency Response

Although the frequency response is often known for a single SPL, it is very useful to know how the FR various with respect to the relative loudness (SPL). For example, many speakers go down to 20Hz, but very few are flat to 100dB and live to tell about it.

Again, the ears handicap us somewhat here, but our brains have learned to compensate to a large degree.

Rate perceived flatness of frequency response [1 to 10, 10 being 100% accurate] and describe any major issues
1. Linearity with respect to an average listening SPL
2. Linearity with respect to different SPL

In Part II we will cover:

Resolution Response

Separation Response

Imaging/Soundstaging

Works well with others in a typical system
* Support for a variety of connectors
* Vibration Susceptibility
* Remote control availability
* Number of chassis
* Number of power cords required

Timing (PRaT) Towards a Hybrid Subjective / Objective Review Process Audiophile dissatisfied with the current results of the reviewing community