Propeller Head Plaza Technical and scientific discussion of amps, cables and other topics.

---

[ Follow Ups ] [ Post Followup ] Thread: [ Display  All  Email ] [ Propeller Head Plaza ]

---

[ Alert Moderator ]

---

RE: Do you realize

Posted by morricab (S) on January 2, 2008 at 10:33:07

In Reply to: RE: Do you realize posted by theaudiohobby on January 2, 2008 at 05:13:39:

"Why would it be valid?"
Its rather obvious isn't it? I mean the level at which different distortions are audible is rather the whole point in determining the ultimate quality of the product. Just making arbitrary specifications because "lower is better" regardless of the methods used to "lower" distortion is not really a very valid way of doing things, although it seems to be the standard way of doing them. It turns out that the most common method for "lowering" distortion has problems of its own just like digital, while having lower distortion in the conventional sense, introduced other kinds of distortion. There is no free lunch!

"Well.."

Well what? If there is some other reason distortion is important beyond being possibly audible then please tell me what that importance would be.

"It's the facts that count not what I believe. I am simply asking you to back your opinion with something more than just your belief."

Don't try to turn it around, you have made a conscious choice to buy a very expensive piece of electronics that was based on listening I assume. If you claim it was also based on measurements then I can point out many other amps with as good or better measurements that cost far less money. You coming on here and agreeing with Geddes that amplifier distortions are small effects yet you have spent a large sum of money on an amp?? Why didn't you simply spend that money on some Geddes speakers and use a $150 Pioneer receiver?? A bit hypocritical don't you think?

What "facts" do you actually have? You have data but not facts. What kind of correlation do you have between measurements and listening preferences? If you have tools that allow me to do this then by all means present them. Numbers don't mean much if they don't correlate with real experience in this case. Afterall, we are talking about the listener as the final arbiter of sound quality are we not??

Now, Geddes developed actually a pretty decent metric for correlating distortion with perceieved sound quality; however, when I questioned him about it he said that it was mainly for loudspeakers because he didn't "believe" that "well designed" (whatever that is supposed to mean and by what criteria is an amp considered well designed??) amps sounded essentially the same or rather their affects were so minor as to not be important. I told him that I thought his metric might be useful in determining what made a good sounding amp. What he fails to understand is that the distortions of speakers and electronics do not really mask each other as they are fundamentally different in nature. So while "small" they can nevertheless be quite annoyingly audible.


I used the metric from Cheever's paper once to determine amp quality. The ones that came up best were not surprisingly ones with the least high order distortion. Now I have heard most of the ones that I applied this metric to and the results from listening were similar to the metric. I got the measurements from Stereophile and Soundstage. No, I haven't gathered all of these amps together in a room and tried them one after the other and then measured them. I had to take the data where I could find it and I listened to these amps under different conditions; however, I remember if they sounded good or not and those I remember sounding good had the better numbers using the Cheever metric. For example, I am not a huge fan of Wilson speakers but I found that they sounded really good once paired with a Lamm amp. I heard other speakers with Lamm amps and it never failed to sound at least very good. The same is true of OTL amplifiers I have heard and/or owned.

In fact, I think you have nothing beyond your belief in what you hear as well. You have used no metric to correlate measurements to your own hearing, have you? Didn't think so. Again, measurements in and of themselves are meaningless unless you can correlate them with audible effects or determine for sure that they are inaudible.

When I take data in the laboratory we have to correlate that data with real chemical and physical phenomena or at least to a model that describes the behavior of those phenomena, otherwise the data could describe anything or nothing. It has to RELATE to something unless perhaps it is being used merely as a QC tool. Many engineers have this "lower is better" mentality without thinking carefully how it relates to the listening. Lowering one kind of distortion often has the effect of introuding other distorions, which are perhaps more audible than the original and thus defeating the purpose of lowering the original distortion in the first place.

Digital introduced us to new distortions, even though it was much lower in "traditional" analog distortions. Many find them more objectionable than the older, higher in level, distortions. Now Class D amps are introducing completely new and different noise and distortion into amplification and the result is a sound that is unlike either tube or SS! Some don't mind the dryness of tone that I hear but those of us with natural instrument tone high on the priority list find them unlistenable for the long term (I have owned and sold two different kinds).

I have also used Keith Howard's little programs (look for his article on "euphonic distortion" in stereophile) to add distortion to a musical track. Howard did it in a very generic way and found that ALL audible distortion was worse than no added distortion; however, he also found that a progressively reducing distortion level with even and odd harmonics sounded the least objectionable.

I took it one step further, I looked up the distortion patterns published in Stereophile and Soundstage for 5 or 6 "top" amplifiers and added their distortion patterns to a short piece of music. The music I chose was a solo violin as I know the sound well. The distortion was audible on all the tracks where it was added and worse than the undistorted track. The best sounding one was the Lamm amp followed closely by another SET. The worst was a SS amp with high negative feedback. Guess which one had lower distortion??

" small effect does not necessarily mean inaudible, at least not in the context used"

I agree but he is clearly meaning it as audibly insignificant. To this I would say "Maybe to his ears" but if I have learned one thing from these phase distortion discussions the audibility and significance of certain distortions is HIGHLY listener dependent.

An interesting website looking at distortion audibility:
http://www.klippel-listeningtest.de/lt/default.html

This post is made possible by the generous support of people like you and our sponsors:

Kimber Kable

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---