john curl said:
You guessed it JC, I'm pretty good at banging on iron though, I would like to think I can handle this. Not to worry about your jewel in the rough, Phil H has stepped up on the technical end and I will be getting some help from Rick over here at Cullen Circuits when I have the parts and instructions completed.
I have been very careful to compile the (priceless and very much appreciated) advice from you, Mr. Crump, and Phil H. and as you may have noticed, am taking my time and asking lots of questions to get it right.
Thanks again, It is always a pleasure to get a response from the man himself.
Matt
BTW coffee is for laptops.
mfc said:
No, it's not quite like that. I'll comment more here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=138477
janneman said:
I'd think that music files would be more difficult to hear, since they're not offensive. The louder music would mask the sound since it's similar. Like with some music compression, they take out the quieter information when during loud passages.
I know that when CD first came out, many audiophiles complained that they didn't enjoy listening as much, listened for shorter periods of time, and some complained of ear fatigue. I'd think these noises and distortions would be much quieter than a Sousa band in the background.
The ear perceives some sounds as much louder. And when you have aversion to a sound it limits the loudness and length of time you can listen to it. That's something short term listening tests don't take into account.
janneman said:
Jan,
I disagree completely with your conclusion(s) here.
I agree that the Sousa (his name, John Phillip Sousa) band is uncorrelated, but as such when rendered at such a low level is in effect, noise.
Noise of generally benign harmonic spectra - which is of critical importance.
Whereas the sound of a cap (for example) or an amp is correlated to the main program material, and in the case of a sufficiently clean system without masking taking place would make the correlated aberrations (can we call it that?) much much easier to detect as aberrations from "proper" or "natural" sound - IF that is what sort of aberration is generated, and is of sufficient level, and comes at a point in time when it is audible (time WRT to any given waveform's shape and how the ear perceives that).
What will determine if it is audible or not, and here I mean heard as an aberration, is mostly the nature of the spectum of said aberration. We know that if it is mainly low order, and not of particularly high level it is very easy to "ignore it", 'let it pass", or not be heard at all. Otoh, one MIGHT be able to notice that it is there, IF one got to hear the same passage of program (and let's be clear here, all program material is not equally "revealing" of said "aberrations" ) through an equivalent (or the same) system that is freed of "aberrations" of this particular sort, or perhaps merely of much lower amplitude but the same "aberrations".
WRT to the Diffmaker files, I am not yet clear on what significance they have or will have... they look like a tool that may have some value. Similar perhaps to the "null" test used for amplifiers?
I think the key idea here is that it is not just simple amplitude of "distortion" that will determine audibility or detectability.
I think it is also important to discriminate in discussing this matter between the ability to discern in a discrete way what the "signal in question" is - in this case "can you hear a Sousa band mixed N dB down?" - as compared to what I am calling detecting an aberration, which is merely hearing that something is not quite "as it ought to be."
_-_-bear
PS. it is a bit wierd that I keep quoting Dr. Geddes, as we have disagreed on quite a few things in the past... but his Gedlee metric certainly shows something of critical importance WRT what we hear and what is doing what. I think it explains a great deal, and has the "ring of truth." Fwiw...
OK, I'm not an expert here either. You may be right.
Tell you what, do me a favour. Download DiffMaker, and listen to the following tracks:
- The difference track between no cap and a film cap;
- The difference track between no cap and a ceramic cap;
- Test track # 1 (for convenience) difference track between track with sousa and without.
Now as you listen to the difference tracks, you in fact listen to the 'aberration' in isolation, which is present in one track of each pair, but not in the other. I would be interested in your view of which one you expect is easier to detect down in the track.
Jan Didden
Tell you what, do me a favour. Download DiffMaker, and listen to the following tracks:
- The difference track between no cap and a film cap;
- The difference track between no cap and a ceramic cap;
- Test track # 1 (for convenience) difference track between track with sousa and without.
Now as you listen to the difference tracks, you in fact listen to the 'aberration' in isolation, which is present in one track of each pair, but not in the other. I would be interested in your view of which one you expect is easier to detect down in the track.
Jan Didden
janneman said:
Jan
If have listened to the tracks you recommended, and I can hear a diff between the no cap/film cap and even clearer between no cap/ceramic cap
The test track #1 that should be with no Sousa / Sousa band I can not hear the Sousa band.
That said I think the recording is so bad and distorted that it would be possible to hide a F1 racing car at full throttle in there without hearing it.
The DiffMaker is a step in the right direction, but misused it is more like a toy made for entertainment.
I did the test with headset.
Cheers
Stinius
stinius said:
Did you listen to each of the two tracks in a pair or to the difference? If I listen to the difference for test track 1 I clearly hear the band to the point that I start tapping my foot...
Jan Didden
janneman said:
Jan
I listened to at them separately, not the difference.
As I said it is possible to hide a lot of sound in the difference between the two tracks.
Cheers
Stinius
That is what I was afraid of, Stinius. The test material must be pristine, to hear really subtle differences. ANYTHING coming off of my computer is suspect.
As far as ceramic caps are concerned, I am glad that people are coming to recognize this problem, and it is about time! I find it one of the biggest differences between 'good' and 'mid-fi' audio.
I discovered the problem, myself, with a tube Fisher tuner. I replaced the .1uf output cap with a 2 uf ceramic, thinking that I was improving the sound, back in 1972. I kept hearing this 'edginess' coming from the tuner, and I traced it back to the ceramic cap. Why had I missed this in measurement? About 1 year later, Tektronix showed me how to measure large value ceramic caps, AND why they were virtually awful for audio reproduction. Then, with newer test equipment, I could also measure harmonic and IM distortion from a single cap.
As far as ceramic caps are concerned, I am glad that people are coming to recognize this problem, and it is about time! I find it one of the biggest differences between 'good' and 'mid-fi' audio.
I discovered the problem, myself, with a tube Fisher tuner. I replaced the .1uf output cap with a 2 uf ceramic, thinking that I was improving the sound, back in 1972. I kept hearing this 'edginess' coming from the tuner, and I traced it back to the ceramic cap. Why had I missed this in measurement? About 1 year later, Tektronix showed me how to measure large value ceramic caps, AND why they were virtually awful for audio reproduction. Then, with newer test equipment, I could also measure harmonic and IM distortion from a single cap.
john curl said:
A typical wrong extrapolation, both time and technology wise.
While "large ceramic" capacitors (in particular in the uF range) can certainly be horrible, exhibiting a full range of problems starting with microphony and ending with huge tempcos, small ceramic NP0 caps are excellent. This was checked, published and confirmed all along in the last 20 years.
Most here would not use NPO caps for ANYTHING in audio, as they are too expensive, too small in value, and to large, to be practical, except as a replacement for mica. I just use mica, myself.
"Most here .... " the only caps I have in a signal path are Mundorf's MCap®ZN in a crossover, 20 EUR each.
And 2P7 NPOs in a FB.
And 2P7 NPOs in a FB.
john curl said:
Huh? Last time I've checked (and that was not 40 years ago
), silver micas used to be small in value, large and expensive compared to NP0.I'm not saying NP0 is better than silver mica, but it's just a convenient, for all practical purposes, alternative.
john curl said:
Yes, polypropylene 22OpF. And 1uF caps in a servo. Nippon Chemi-Cons in a PSU.
Hi John,
Ceramic caps are not a problem in audio. They are simply a good part in the wrong application. And I will also happily use NPO ceramics in some applications. I'll use the other ceramic types as filters across power supplies without a sonic penalty. Same as carbon composite resistors are great for RF work, but not the best for low noise, or circuits with large voltage fluctuations across them.
Many designers go ahead and use parts without first learning about them. It would be the same idea as if people started building engines without regard for bolt strengths and expansion properties. Then there is the entire friction thing on top of that. Chemical reactions between coolant types and metals, or oil chemistry. We would call someone who does not learn about these things, but goes ahead and builds internal combustion (or jet) engines completely irresponsible. In my book, the same holds true in audio. There's just less money on the line so society ignores the lack of education prevalent in audio engineering.
So the only problem in audio as I see it, is the lack of knowledge that exists on components. Blaming a mis-used part is like blaming the tools.
I find myself in agreement with syn08 and many others who have experience with testing these parts.
-Chris
Ceramic caps are not a problem in audio. They are simply a good part in the wrong application. And I will also happily use NPO ceramics in some applications. I'll use the other ceramic types as filters across power supplies without a sonic penalty. Same as carbon composite resistors are great for RF work, but not the best for low noise, or circuits with large voltage fluctuations across them.
Many designers go ahead and use parts without first learning about them. It would be the same idea as if people started building engines without regard for bolt strengths and expansion properties. Then there is the entire friction thing on top of that. Chemical reactions between coolant types and metals, or oil chemistry. We would call someone who does not learn about these things, but goes ahead and builds internal combustion (or jet) engines completely irresponsible. In my book, the same holds true in audio. There's just less money on the line so society ignores the lack of education prevalent in audio engineering.
So the only problem in audio as I see it, is the lack of knowledge that exists on components. Blaming a mis-used part is like blaming the tools.
I find myself in agreement with syn08 and many others who have experience with testing these parts.
-Chris
- Status
- Not open for further replies.