You may have noticed the till teorema: the more THD the more audiophile. Using a chip like that will prevend me from spending hundreds of hours searching for a little more of 7th harmonic by swapping resistors, expensive electrolytics, or adding ferrite or maybe even real pearls in the digital grounding....
I never did and will no do. I have a problem concerning what we call "Produktästhetik" with commercial elektronics. The only products of the consumer range of electronic/electric devices i buy are tools : Hilti, Fein, Flex/porter cable... and of corse computer parts if absolutely necessary.
btw: i would suspect the specs of commercial gear are very good. Same as compare gainclone THD to aleph THD...
But why do you test them with a -60dB signal? Wouldn't full scale, or at least half scale, be more appropriate?
For the chip, it wasn't written made in taiwan but only taiwan, do france would also be enough 😉
For the chip, it wasn't written made in taiwan but only taiwan, do france would also be enough 😉
Back to serious for those who are interested
Todays work so far:
Tested 5 chips like a, b, c, d, e, f (only one channel).
Then tested 6 combinations of two chips parallel, like a&b, c&f, b&d ...
Distortion is random, everything is possible.
The result of throwing two chips together is unpredictable.
If the two chips have values -50 & -54, the result could be:
-48, -50, -52, -54, or -56, but mostly it will be in the middle like -52.
In my test it was better than each single chip 9x, like the better chip 2x, in the middle 17x, like the inferior chip 3x, worse than each single chip 5x.
With lots of chips, good combinations can be found.
Todays work so far:
Tested 5 chips like a, b, c, d, e, f (only one channel).
Then tested 6 combinations of two chips parallel, like a&b, c&f, b&d ...
Distortion is random, everything is possible.
The result of throwing two chips together is unpredictable.
If the two chips have values -50 & -54, the result could be:
-48, -50, -52, -54, or -56, but mostly it will be in the middle like -52.
In my test it was better than each single chip 9x, like the better chip 2x, in the middle 17x, like the inferior chip 3x, worse than each single chip 5x.
With lots of chips, good combinations can be found.
Bricolo said:
The analyzer can only see down to it's noise floor of -75dB.
With full scale signal it will not see much, because THD of the TDA is very low, like -90.
With the -60dB signal THD of the TDA goes up, to -40.
So it gets visible for the analyzer.
Philips used -60dB signal for selection too.
If you have the time, could you just show is the spectrum of a non os player, at a normal volume?
Just to see how big the 44kHz noise is
Just to see how big the 44kHz noise is
Yes, you already posted this
what I wanted to see is the amplitude of the 44khz compares to a full scale signal
what I wanted to see is the amplitude of the 44khz compares to a full scale signal
Bricolo said:
Yeah, send me a bottle of good french wine and I will see what I can do
Remember, this was with the analog filter that is inside the CD650, I did not remove it.
Bernhard said:
That's what I was fearing 🙁
And this is with your player's analog filter? If you didn't modify it, I don't think it changes something, since it was certainly designed to attenuate the 4*Fs garbage
thank you for the measurements!
hey Bricolo, you just safed one bottle of wine. Send it to me!
If you send together with 2 of the famous B3 it will also safe some shipping costs.
If you send together with 2 of the famous B3 it will also safe some shipping costs.
- Status
- Not open for further replies.