watchnerd
Grand Contributor
Every couple of years I re-examine what I believe to be true in audio, and then use these thoughts as a set of guiding principles when it comes to total system design. To be reassessed as new evidence or technologies emerge.
Here are mine. I'm sure yours will be different, or even contradictory.
1. The recording matters the most.
2. The transducers, speakers, headphones, & cartridges, are next most important.
3. The room is third, but barely.
4. Analog electronics are mostly a solved problem; past a mid-tier price point, returns are seriously diminishing.
5. Digital is rapidly approaching 'solved problem' state. $3000 DACS from 5 years ago are bettered by $500 DACs today. Don't spend too much on a component category that changes this quickly.
6. I cannot pass DBT ABX test between any kind of lossless, whether standard definition or high resolution, for playback.
7. Balanced connections are better. Use them wherever possible.
8. There is no reason to invest in Class A power amplifiers any longer; the best Class D is now good enough, and without the inefficiencies.
9. Tubes can be fun in phono stages and preamps. Using them as power amps carries too many other sacrificies in terms of matching speakers. The price is too high for the benefit.
10. Power is now essentially limitless. It's okay to trade sensitivity for bass extension.
11. Active speakers with DSP crossovers are the future, regardless of driver type. There is no real reason to invest in anything else at the moment, even if it means waiting.
12. Networked lossless streaming trumps direct-attached computer systems, both for convenience and noise isolation.
13. Digital EQ is not as bad as audiophiles think it is.
14. You can't trust sighted listening tests, especially if self administered. Confirmation bias is unavoidable, even if conscious of it.
15. Multi-bit/R2R/Delta-Sigma, doesn't really matter -- all that matters is the implementation.
16. Vinyl is not superior to digital. But I like it.
17. There is no replacement for displacement.
Edit: added more, based on in-thread discussion
18. [Tentative] @Cosmik 's rule: Room correction is fundamentally not valid.
19. Cables: For analog cables or digital coax, anything more expensive than Canare or Belden is a waste of money. For digital data, generic ethernet and USB are fine. For Toslink, Mitsubishi Eska is fine.
20. If you use passive speakers, keep the speaker cables short and use balanced interconnects. This is one of many reasons why monoblocks are better.
21. Tube rolling will drive you nuts, aggravate symptoms of audiophilia nervosa, and costs money.
22. MQA is partially lossy compression in a fancy container, a technology in search of a use case, and can be ignored for now.
23. Active speakers with the playback software embedded in the speaker (e.g. KEF LS50) appear, as of this writing, to have no defined upgrade path to new formats or protocols. Proceed with caution or use an external streamer.
What are yours?
Here are mine. I'm sure yours will be different, or even contradictory.
1. The recording matters the most.
2. The transducers, speakers, headphones, & cartridges, are next most important.
3. The room is third, but barely.
4. Analog electronics are mostly a solved problem; past a mid-tier price point, returns are seriously diminishing.
5. Digital is rapidly approaching 'solved problem' state. $3000 DACS from 5 years ago are bettered by $500 DACs today. Don't spend too much on a component category that changes this quickly.
6. I cannot pass DBT ABX test between any kind of lossless, whether standard definition or high resolution, for playback.
7. Balanced connections are better. Use them wherever possible.
8. There is no reason to invest in Class A power amplifiers any longer; the best Class D is now good enough, and without the inefficiencies.
9. Tubes can be fun in phono stages and preamps. Using them as power amps carries too many other sacrificies in terms of matching speakers. The price is too high for the benefit.
10. Power is now essentially limitless. It's okay to trade sensitivity for bass extension.
11. Active speakers with DSP crossovers are the future, regardless of driver type. There is no real reason to invest in anything else at the moment, even if it means waiting.
12. Networked lossless streaming trumps direct-attached computer systems, both for convenience and noise isolation.
13. Digital EQ is not as bad as audiophiles think it is.
14. You can't trust sighted listening tests, especially if self administered. Confirmation bias is unavoidable, even if conscious of it.
15. Multi-bit/R2R/Delta-Sigma, doesn't really matter -- all that matters is the implementation.
16. Vinyl is not superior to digital. But I like it.
17. There is no replacement for displacement.
Edit: added more, based on in-thread discussion
18. [Tentative] @Cosmik 's rule: Room correction is fundamentally not valid.
19. Cables: For analog cables or digital coax, anything more expensive than Canare or Belden is a waste of money. For digital data, generic ethernet and USB are fine. For Toslink, Mitsubishi Eska is fine.
20. If you use passive speakers, keep the speaker cables short and use balanced interconnects. This is one of many reasons why monoblocks are better.
21. Tube rolling will drive you nuts, aggravate symptoms of audiophilia nervosa, and costs money.
22. MQA is partially lossy compression in a fancy container, a technology in search of a use case, and can be ignored for now.
23. Active speakers with the playback software embedded in the speaker (e.g. KEF LS50) appear, as of this writing, to have no defined upgrade path to new formats or protocols. Proceed with caution or use an external streamer.
What are yours?
Last edited: