When you don't understand, of course it appears as obfuscation to you.Scott,
From the moment I get up till when I collapse exhausted at night, I am constantly experimenting. It - is - what - I - do.
So, this is why I am on Merrill’s case, as his constant obfuscation simply does not add up as standard modus operandi for someone genuinely interested in audio. Anyway, I am getting bored with this.
ToS
Good, so maybe you & ScottJ can leave out the constant abuse of me & my posts?
Merrill,
And you are kicking off - again. You had your chance and blew it - again.
Look kiddo, the floor is yours and I am done with you.
ToS
I shall do you the favour of presuming you are paranoid or just enjoy trolling and not a habitual liar.
Thanks Scott - I'm thinking of a random modulation of amplitude not AM at a specific frequency which will indeed create sidebands.
Would a MT test signal modified in this way be more revealing of modulated noise floor?
That's all I ask - leave me alone - put me on your ignore list, please
If you do the search I told you to do, you will see Lynn Olsen uses the term "technicolor sound" here
"Besides, I wanted an amplifier with the gorgeous Technicolor sound of the Reichert but a bit less coloration, and some of the Ongaku's remarkable silence-between-notes presence and insight into musical textures. No such amplifier was on the market, at any price."
All modulation creates sidebands, but that is not a serious problem here.
Lynn Olsen is not a person to ask about accuracy and neutrality. I don't get it sometimes these folks would have a rich infill of all possible distortions in their signal chain, I doubt they would have any interest in the Benchmark amp.
Last edited:
Yes & if the amplitude is random, then random sidebands = noise
Multitones all varying in amplitude will create noise floor fluctuation. Add in IMD from various frequencies being played together & we have further noise floor modulation
Relate this to music signal
I didn't use the term "technicolor sound" to describe something that was accurate - quite the opposite but ToS got himself in a twist about my use of the phrase - go figure
A recurring theme in some forums is that someone promotes an unusual idea or hypothesis (usually of the “extraordinary claims require extraordinary evidence” variety), that needs some sort of testing... Then they complain when others decline to do that testing for them.... repeatedly....
Discussing the viability of a test & what might be the pitfalls/considerations seems a reasonable thing to do.
Asking if such a test had been done before also seems entirely reasonable.
I don't see anyone complaining, do you?
Asking if such a test had been done before also seems entirely reasonable.
I don't see anyone complaining, do you?
Martin Mallinson wrote about it briefly in an ESS white paper. Since he is VP of engineering at ESS, he probably knows all about it.
Also, Howie Hoyt has described listening for noise floor modulation. Apparently, he is one of those people that finds it annoying.
The effect is caused by state variable settling following volume level transients in sigma-delta dacs.
Last edited:
At what levels, though, and how much?
Just thinking about the noise floor of your average recording compared to the noise floor of a well-done DAC that has 120 dB+ SNR... I have serious doubts about the audibility of this phenomenon.
Just thinking about the noise floor of your average recording compared to the noise floor of a well-done DAC that has 120 dB+ SNR... I have serious doubts about the audibility of this phenomenon.
Audibility is likely the result of the modulation of the noise rather than being static noise - we are far more sensitive to modulation - that's why seeing the dynamic changes in noise floor would be interesting & this won't show with steady state signals
I imagine that's one reason they keep pushing noise levels down in dacs, often well below the distortion. I'm not sure unmasking distortion is necessarily a good idea, but if there is a good purpose for minimizing noise well below distortion it would probably stem from concern it could bother some people under some conditions.
" independent from the "blind" or "sighted" condition,"......ummm, what other conditions are there, I'm not quite following you there ?. I find that noise embedded in a signal drives subtle behaviours of the replay system and there is a short 'running in period' before the system stabilises to new source signal, usually determined by the first large amplitude peak in the 'new' recording, ie B recording or B upstream component. I can explain more offline/PM if you like.
Dan.
On vintage music I often easily hear tape hiss above even vinyl surface noise. How do you know what you are hearing is not some artifact of some of the elaborate noise removal algorithms people are using? I find some of these things quite annoying when overdone.
Likely? The only thing that's likely is this is this effect occurs at levels that no one can hear unless you have so much gain that a full scale signal would destroy your speakers.
The purpose is the SNR/DNR spec. There is no real need for mono mode, for example, but it lets them claim a better headline figure.
Last edited:
When it's said that the "noise floor of a well-done DAC that has 120 dB+ SNR" - what test signals are used to establish this - nice steady state test signals?
Don't know about speaker destruction, but being deafened when attempting near field critical listening could be an issue. I tend to listen at much lower levels than the system is capable of. Turn it up to impress visitors.
whilst I am aware that UK restorers such as Dutton try and remove as much noise as they can I've not found examples in the CDs he has produced of it sounding overdone. But my sample points are limited here.
What does annoy me is noise gates on pop recordings of a certain era and the fading out of noise between tracks. But that's just me.
Of course the exercise of extracting the noise floor of a recording is not that difficult in most cases, but no one wants to see the answer. I suspect some recording where these supposed effects are audible are two orders of magnitude worse than -120dB.
Yeah, that is the usual reason given for why it isn't a problem - Mallinson didn't say his people heard this issue ONLY when vol was cranked up - I wonder does Howie?
- Status
- Not open for further replies.