This is just a load of sophistry. Plausible sounding but essentially fallacious. Typical audio craziness, an argument to prove bent is straight.
Seriously, do you ever think about the implications of what you write?
“Sometimes I wonder about the karmic implications of these actions.” -Farmer Vincent
10/10 VladimirK. Go to top of the class.
I have no idea why Sy is feigning ignorance; this topic has been discussed many times before on the forum eg below by Tubelab:
http://www.diyaudio.com/forums/lounge/264298-blameless-standard-tube-amplifiers-5.html#post4114860
He seems to use this technique regularly to further his own agendas
Then why did you?
Your second paragraph is not even worthy of reply.
Re para 3: why did you mention single ended; I never alluded to this in my argument which was based purely on generic premises. Polarising to extremes to deflect from the validity of the central argument ?
I have no idea why Sy is feigning ignorance; this topic has been discussed many times before on the forum eg below by Tubelab:
http://www.diyaudio.com/forums/lounge/264298-blameless-standard-tube-amplifiers-5.html#post4114860
He seems to use this technique regularly to further his own agendas
Then why did you?
Your second paragraph is not even worthy of reply.
Re para 3: why did you mention single ended; I never alluded to this in my argument which was based purely on generic premises. Polarising to extremes to deflect from the validity of the central argument ?
The ignorance is not feigned- I don't know what you mean by the term "steady state conditions." I'm beginning to wonder if you do. Could you set my mind at peace by simply telling me what you mean by that phrase?
What is your belief based on, some objective data by any chance?
Once again, strawman argument in an attempt to bolster your preexisting belief (which stems probably from magazine reviews, subjective listening experience, marketing hype, sales pitch, online testimony, audiophile claims) and discredit the established industry standards because they don't pamper your own notion.
No instantaneous switching between the cords? There it is, the flawed method I mentioned. You didn't account for the aural memory span which is very short.
It seems that you are the type that prefers to be spoon fed. I already provided the link for you. Put some effort in helping yourself, it's good for you.
Last edited:
Did you get that from Robert Harley's article or from the ones written by the likes of him?
A simple example of the "steady state" issue: if feeding the DUT a test signal the PS stabilises into a particular waveform pattern. But when fed music the PS is continuously "destabilising" - the voltage is sagging and rebuilding in a very erratic, and if you look at it on a virtual scope, rather terrifying way. Many would say that PSRR solves everything, but closer examination in simulations, with real world parasitics in all the key places, demonstrates that the electronics are having a very unpleasant time dealing with it all ...
Your flippant attitude barely masks your ignorance. Refutation by labelling is a rather puerile and primitive excercise more suited to the schoolyard.
Bent is bent. By stating that I was alluding otherwise you have clearly demonstrated an inabilty or deliberate unwillingness to grasp the basic fundamentals of what I was stating.
Last edited:
No, he didn't, and that's not his fault, he wasn't trying to define terms for you, nor did he use the term "steady state." What exactly do you mean by the term "steady state"? Or is it just a mantra and actually has no specific meaning to you? That's OK, too, you just need to be clear about this.
Tedium ad nauseam. Steady state = static as used by Tubelab. You knew that. Goodbye; I'm going back to my popcorn
Last edited:
sine waves happen to have derivatives of all orders - how "dynamic" do you want?
even the supposition that power interactions are somehow too difficult for "conventional engineering" has a logical problem in that if the load current/test signal sine is relatively prime to the AC mains supply frequency you get plenty complex beats between charging pulses and load
and some want to lump in 2-tone IMD as "too simple" too - now you can get signal envelopes from peak to null of as many seconds, minutes, hours as you can control the 2-tones relative frequencies to
even the supposition that power interactions are somehow too difficult for "conventional engineering" has a logical problem in that if the load current/test signal sine is relatively prime to the AC mains supply frequency you get plenty complex beats between charging pulses and load
and some want to lump in 2-tone IMD as "too simple" too - now you can get signal envelopes from peak to null of as many seconds, minutes, hours as you can control the 2-tones relative frequencies to
A good solution is time domain analysis, but in the real world of expensive test gear they are struggling to be good enough, and to be set up in such a way for a designer to really appreciate what's going on. As an example, I'm playing with an ambitious class AB power amplifier design, which delivers remarkably impressive FFT figures even when highly "stressed" - yet, it is easy to see that crossover distortion is still the Achilles Heel, buried in the noise of those nice FFT displays: a nasty glitch is injected into the output, well within hearing sensitivity range. So what does one take notice of - "quality" FFT results, or a clearly visible artifact in the waveform shape ...?
Nothing is "too difficult", but it does have to be taken into account ... if I was asked to really test an amplifier, using conventional techniques, I would stress the power supply while at the same taking "delicate" measurements. As an example, for a unit with single supply and 2 channels, I would increasingly stress one channel, while monitoring the THD, say, of the other channel being fed only a relatively low level signal. The measurement of the latter should not vary one iota doing this, but who would put money on that happening, 😉? My guess is that "good sounding" amps will be those which show least variation doing such a test ...
since discrete time series and the fft are mathematical duals - have exactly the same information content your claim is strictly speaking false
the point of using both views is that human eyeballs on the data in either form can clearly see different features - not that the information isn't there, could be extracted by the right algorithm
and you're supposing a crap ADC/fft system/setup if you can't see "a nasty glitch is injected into the output, well within hearing sensitivity range" in the fft data
Last edited:
Strictly speaking, yes, false ... but who has the ability to translate the FFT "grass" into a time domain behaviour, at a glance?
Both views provide information, but my personal take is that what is very clear in the time domain is the more important, subjectively.
The nasty glitch is not visible by eyeballing the raw waveform, but by subtracting it from the pure input signal: differences everywhere else are in single mV's, but jump to over 100mV at the two crossovers ...
Both views provide information, but my personal take is that what is very clear in the time domain is the more important, subjectively.
The nasty glitch is not visible by eyeballing the raw waveform, but by subtracting it from the pure input signal: differences everywhere else are in single mV's, but jump to over 100mV at the two crossovers ...
- Status
- Not open for further replies.