Unfortunately, THD is not a highly useful metric. One can argue that if THD is low enough it shouldn't matter, but there are other things to measure such as distortion spectrums at different frequencies and power levels, IMD, maybe jitter attenuation for DACs, and so on. Some people like to find the single tone sweet spot and measure a spectrum there, maybe -15dB FS@1kHz for a DAC, as though it were representative of real music waveforms.
Again, I just note there are a lot of projects here in the forum that never get comprehensibly measured. Why not? Nobody wants to know the truth?
Again, I just note there are a lot of projects here in the forum that never get comprehensibly measured. Why not? Nobody wants to know the truth?
For some people that may be a reasonable assumption. For other people who don't have a lot of trouble hearing undithered 16-bit audio and some other low level distortions, it may not be the best number.
Also, some people find that even good quality sound card sound different. For those people, I think notch filters and precision oscillators are probably appropriate.
By the way, Bruno Putzys said to believe audiophiles, that they aren't imagining things. I would probably amend that to say there are two types of audiophiles, those that can perform blind (if comfortable and practiced) and those who can't. He was probably thinking of the former.
Last edited:
Pick a number. I can't hear at -80dB so for me it works. I understand that what works for me (hearing loss) may not work for others with better hearing
I stated that I agree that THD is not a useful metric.
I also stated that the soundcard/pc/sw would be capable of displaying the harmonic profile, which I believe is useful, particularly when viewed at different signal levels and frequencies.
I stated that I agree that THD is not a useful metric.
I also stated that the soundcard/pc/sw would be capable of displaying the harmonic profile, which I believe is useful, particularly when viewed at different signal levels and frequencies.
The only effect that I can account for in the world of real world physics of a wire changing the sound quality as it moves around. and for the effect of "burn in" is the effect of a burnt and carbonized conductor.
Weather or not we should be using "resistor wire" instead of low ohmage wire in our audio cables remains to be seen. There might be some sonic benefit to that, who knows. We certainly use resistors in our spark plugs and engine bays to limit the amount of RFI produced.
Weather or not we should be using "resistor wire" instead of low ohmage wire in our audio cables remains to be seen. There might be some sonic benefit to that, who knows. We certainly use resistors in our spark plugs and engine bays to limit the amount of RFI produced.
Last edited:
Not true at all, a distortion magnifier a la Cordell is one way and twin tees are really trivial for anyone to make. You can do -140dB anyday with care. I did it when nothing was available.
Okay, I think we are mostly in agreement. I think for me I want a notch filter and precision oscillator, at a minimum.
I guess my wish is that we could make some progress on some of the disagreements around here if precision measurement capability were more widely available. But, I think a lot of the disagreements may be at a lower level than unaided sound cards may be good for. Also, we could have people who may not have access to the most suitable soundcards for accurate standalone measurements.
Last edited:
That's true, but you need more than a sound card to do it. That's why I said "only" a sound card.
I am all for making it easy for people do more measurements of projects they are working on here.
It would perhaps be valuable in a number of ways. For one thing, people could see how their layouts affect their results.
Last edited:
You are not going to get really good results with a $29 sound card.
You are going to have to spend some money, but realistically we are talking about a hobby. We are not talking about a $20,000 bass boat.
For less than $100 you can get the ASUS STRIX RAID PRO which claims 110 dB s/n on the input and 115 on the output.
Couple that with some software, a pc, a notch filter (if you are in this as a hobby you should be able to build one for less than $100) and the Cordell distortion reducer (<$200) and you should be able to get reasonable results.
Throw in an IG-18 and do "The Greening of the IG-18" mods (currently available on ebay for less than $100) and you have a reasonably low distortion variable signal source.
I have more problems with ground noise, EMI From CFLs & LED lamps, and RF pickup from nearby transmitter towers than I do with the noise floor of my equipment.
It's 20 Euros just to look at the schematic and see if you are interested. That probably doesn't help popularize it.
€2.99 actually Linear Audio | your tech audio resource
If you don't want to pay that then you are not serious about measuring anything.
If you don't want to pay that then you are not serious about measuring anything.
There are articles on the internet describing the function of the reducer. A competent engineer can design one from the block diagram.
130 Euro for a complete parts kit with PCB .
I'm too cheap to pay that much considering I have most of the parts.
I wish they still sold the PCB and critical parts as a sub-kit.
130 Euro for a complete parts kit with PCB .
I'm too cheap to pay that much considering I have most of the parts.
I wish they still sold the PCB and critical parts as a sub-kit.
Okay, useful for amplifiers. That's a fair amount of what goes on around here.
What about DACs? How shall we measure all the DAC projects going on? People say they sound different, some sound better than others, etc. Why?
Also, what about soundcard synchronous averaging FFTs for the masses?
What about DACs? How shall we measure all the DAC projects going on? People say they sound different, some sound better than others, etc. Why?
Also, what about soundcard synchronous averaging FFTs for the masses?
Well, I agree that some sound better than others.
I built a WM8741 and a CS8416/CS4398 dac and the CS8416/CS4398 sounds better to me when driven from the same source.
I have not tried to measure the differences to see what I could find.
I suspect that PS noise and brick wall filter have something to do with it.
I will close by saying the following:
At one time I thought a 1MHZ GBW op amp in unity gain could not effect an audio signal.
Someone posted comparison files with a straight wire and the op-amp playing the same music score
From an engineering standpoint, they should sound the same.
Yet, even with my limited hearing (hearing aids) I was able to distinguish the difference using FUBAR 2000 with an ABX plugin to a 90% confidence level.
This convinced me that (1) standard measurements may not be sufficient to differentiate audio files, and (2) even with damaged hearing, one may be able to hear differences which are below the accepted listening thresholds.
As a result of the listening experience, I endeavor to maintain an open mind to plausible explanations for phenomenon that I personally don't experience.
I built a WM8741 and a CS8416/CS4398 dac and the CS8416/CS4398 sounds better to me when driven from the same source.
I have not tried to measure the differences to see what I could find.
I suspect that PS noise and brick wall filter have something to do with it.
I will close by saying the following:
At one time I thought a 1MHZ GBW op amp in unity gain could not effect an audio signal.
Someone posted comparison files with a straight wire and the op-amp playing the same music score
From an engineering standpoint, they should sound the same.
Yet, even with my limited hearing (hearing aids) I was able to distinguish the difference using FUBAR 2000 with an ABX plugin to a 90% confidence level.
This convinced me that (1) standard measurements may not be sufficient to differentiate audio files, and (2) even with damaged hearing, one may be able to hear differences which are below the accepted listening thresholds.
As a result of the listening experience, I endeavor to maintain an open mind to plausible explanations for phenomenon that I personally don't experience.
- Status
- Not open for further replies.