Hmmm... I thought we were talking about The Carver Challenge, apologies if I misread. I jumped in the middle (again).
That said, a null of only 36 dB between two "identical" production amps would surprise me.
Objectivist or Subjectivist? Give Me a Break
- Thread starter Jeff Fritz
- Start date
Hmmm... I thought we were talking about The Carver Challenge, apologies if I misread. I jumped in the middle (again).
That said, a null of only 36 dB between two "identical" production amps would surprise me.
Nobody said you should. If you like the obvious and easily-measured degradation of vinyl records or analog tape, more power to you. But if you believe something sounds better or worse or just different, and a null test proves the two sources are in fact identical, then you have to question your own perception.
--Ethan
---But Ethan, then why are the super high-end stereo sound systems, many of them, use LPs for music reproduction listening?
It's not a simple number, and you also have to factor in Fletcher-Munson. This graph of jitter noise is from The Art of Digital Audio by John Watkinson. It shows the amount of noise you can expect at various frequencies, with the noise floor of various bit depths as a reference:
Yes, and that can almost border on being audible in highly favorable situations. But -76 is not a typical amount of jitter. It's a specific worst-case result from an immature technology. Good thing I run only video through my HDMI connections!
Understood and agreed. But you still have to factor in Fletcher-Munson, and also the masking effect depending on how far away in frequency the artifacts are.
As always, to obsess over gnats while ignoring the elephant in the room seems silly.
--Ethan
It proves that in all the time that article has been on my web site, not one person has had the courage to email me their choice for which file was recorded at which level. Most of my comparisons get a lot of emails, and I save them all for eventual tabulation. Note that my 80 dB figure is for a guarantee that two devices sound identical. In many cases, sources that differ by / null to only 40 dB may well be indistinguishable. If you believe my record level test proves nothing, please email me which files you think are which. While you're at it, here are two more tests I'd love for you to chime in on:
Converter Comparison
Converter Loop-Back Tests
Not just you, but anyone else here who has confidence in their hearing acuity.
--Ethan
Lol, that was an easy 'escape' answer!
I've already explained many times here and elsewhere that some people enjoy the degraded sound of vinyl and tape. This is not new info. The extra "crunch" can be perceived as clearer.
--Ethan
Perception is a whole different matter. The mind is prone to wild imagination. That's often a good thing. How does one go about questionig his own perceptiion?
For example I claim the sound the sound of an amp improved during breakin. As Amir pointed the measurements out of the box and say two weeks remain virtually the same. How do I alter my perception to account for this discrepancy?
greg
In an ideal power amp the voltage at the speaker terminals will remain constant regardless of the load. Obviously this rules out tube amps, but good solid state amps come awfully close to this ideal. Or maybe I misread your point? Of course, the resistance of the voice coil is the main limiting factor for achieving a super high damping factor.
--Ethan
When I think of a null test on a power amp, I think of comparing the input to the output, to see how much the amp changes the sound. As opposed to comparing two different circuits. Maybe some day I'll make such a test rig, with a single very high quality op-amp for summing / inversion and a ten-turn trim pot for exactly matching levels.
--Ethan
For that matter many prefer the "steely" sound of digital and percieve that as clearity. I find that many prefer digital because it is a more convienent than vinyl. Just pack up your comptuer and you are ready to go. This is much easier than setting up a turntable and changing records.
greg.
By the way, people who are saying that R2R tapes (at the right speed; 15"ps), and LPs (at the right speed; 45rpm) sound better than CDs, are they living on another planet called Analogue?
Are they too subjectivist?
Ethan, what is the very Best sounding audio medium? ...Objectively.
Are they too subjectivist?
Ethan, what is the very Best sounding audio medium? ...Objectively.
Last edited:
Yah, I probably misinterpreted this whole discussion. I started off thinking your way was under discussion, then got off track when it turned to comparing two different amplifiers. Not to each other, but to what the null test shows for each... I have not done enough audio testing recently enough to say what null modern amps might reach so will defer to you.
If I were doing null tests on audio circuits today, I'd probably use a digital-based system (ADCs and DACs) to collect samples and then post-process to see how deep the null. That would make level and phase matching much easier, and also allow one to calibrate out the conversion chain, at least to the noise floor.
Let's rewind a bit. Your original statement is that null testing (an objective measurement) can prove that two devices are audibly the same & audibly transparent. You cite 80dB as the value that will qualify! If you cannot achieve this null in your own tests of the playback system, you are therefore not able to objectively verify that two playback systems are identical sounding.
Your now retreat to subjective criteria to claim that two playbacks are indistinguishable from one another (for one particular example). So I take it that you are admitting by this very use of subjective evaluation that your claim of objective measurement is wrong! if you can show a 80db null being achieved from the speaker outputs then you have achieved your objective! Otherwise your statement remains in the land of speculation & proves nothing!
This is no different to using THD as an objective measure of audibility - sometimes two devices that have the same THD will sound the same & sometimes they won't. Therefore THD is an agreed useless objective measurement with regard to audibility.
Why cite null testing as "proof" of anything audible & why then cite a value of 80dB which you cannot achieve (for the output from speakers)?
Last edited:
Isn't that what I've said about eight times in this thread so far? Including the idea of averaging to get a null deeper than the noise floor. But no matter.
Such a strategy is flawed from the outset anyway. If everyone knows that digital sounds "steely", then how can it be used as proof of perfection?
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |