What "THD" is considered "HiFi"?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
My viewpoint is that we can't make any statement about THD without talking about at what level it is measured at, what load it is driving. To me, this frames the discussion.
Otherwise it is truly a meaningless number.

Historically, THD+N was measured with a Notch filter and an AC voltmeter, with the DUT driven by a low distortion signal generator. Notch out the signal, the rest was bad.
Had a simplicity to it.

To me its main usage is to determine when an amp clips.

Sad in this day and age there isn't a useful way to talk about linearity,

Doug
 
AX tech editor
Joined 2002
Paid Member
Jan? say what?

I cited the two authors of the studies... I presume you are familiar with both??

_-_-

... backtracking....
OK, I didn't see the connection between your statement and those two names. You mean Geddes and Shorter did double blind studies that showed what you stated?
I was questioning that DBT for that had ever been conducted, but I may have missed it.

jd
 
Guys, did I really miss that?

Talking about THD without saying at which power level it was measured is 100% useless.

One can specify (almost) every amp for every THD-level, even Ovidius 0.000x% wonders can hit 1% THD if run hard enough into clipping.

A 10W tube amp with 5% THD may show 0.01% at 1W, so one could specify it as 0.01% THD amp if one can live with 1W output power.

Power is also a big factor, 1% THD at say 100W is certainly much less of a problem (as the 100W will likely never be reached in listening) in comparison to 1% at 10W (much closer to real power requirement, depending on speakers).

Have fun, Hannes

PS: nice to see this odd/even order component discussion is finally showing up again. I find it funny that this topic is of so much interest. There is so much information on this available, just read e.g. Nelson Pass' articles on what he finds in group listening tests...
 
Last edited:
In reality everyone is right. Distortion is not a simple measurement. It's not a case of how much but how offensive.

Most High-End Speakers, even those at many thousands of dollars, will produce distortion, but pleasant distortion.

I am currently using an Arcam AV350 which produces less than 0,01% distortion into a pair of B&W DM603 s3's which produce about 1% distortion. The result is pleasant to the ear, which is what is all important.

My Home-Brew 80W Class A amplifier produces about 0.1% distortion at 5W, which equates to about 90dB through the B&Ws but is far less stressful to listen to.

My advice can only be - Take your Ears Shopping. If you like the sound then buy it.
 
I like it. It implies listening to a solo vocalist shortly after your high intensity metal-fest will be different than the other way around. :D


In some cases, perhaps so.
But in the case of a Class A amplifier that never leaves Class A?

Maybe that is a good thing, better than a chip amp?

The devices will actually cool a bit during the big excursions, especially if they run into AB... so what should one expect then??

Ok, this is cheepz ampz... but still worthy of discussing other "amps"?

_-_-bear
 
Even 50 cent battery powered chip amps can produce low distortion (well under 1%).

On the other hand, I often find the source material to be an offender. I have been driven nuts during listening tests to new amps I built. I hear something nasty and my heart sinks. I unplug the amp and put on my Sennheisers and replay the passage at lower volume through a different source only to find the distortion is part of the source material.

I agree that speakers do distort the sound, but as long as it is not harsh, like clipping, it is more of a coloration to me.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.