Why would you use parts that have no audible effect? I'm no expert, but where I use snubbers and Zobels in amps, they have a function, and the amp will be less stable and noisier without them. Measurably so, and instability and noise are quite audible.
The hucksterism is a function of discouraging people from actually determining audible and measurable differences. It's the lifeblood of fashion audio.
Which describes Joe Rasmussen not at all. He does not deserve being derisively alluded to as a "huckster". It's been no secret that Joe has a commercial audio operation, including having working with the now deceased Allen Wright, long before joining here. Joe has been nothing but open and generous with his time, effort and experience here. Joe has discouraed no one from determining anything, and has in fact done exactly the opposite of that. The increasingly self-righteous antagonism toward him is unseemly and unjustified.
If any regular member were to associate hucksterism with another member, it would surprise none of us should they find themselves under strong moderation. Here, we have a moderator conducting in such association. SY, should you wish to so frequently participate in the back-and-forth of member discussion, that's fine, but I suggest that you relinquish your moderator authority. No doubt, the moderation team could find someone else to pick up your duties. As it stands, at the minimum, you owe Joe Rasmussen an public apology.
Precisely. It is the fans of this technique who claim that it is doing something special, something more than just a filter. When we start talking about the actual workings of a circuit design they get jittery, as we may be about to destroy the magic.Mark Whitney said:
I have no problem with attempts at disproving the effect.
A more cynical person could think that Sy and Joe are on the same team.
A more cynical person could think that Sy and Joe are on the same team.
A realistic person would recognize that I'm not on any "team." The closest I come is being a member in good standing of the Axis of Evil. 😀
It's not a matter of "disproving" that the "effect" is real. It's a matter of determining whether it's real, and if it's real, whether or not it's simple filtering that can be applied anywhere in the signal chain. Simple experiments, and necessary first steps.
It's not a matter of "disproving" that the "effect" is real. It's a matter of determining whether it's real, and if it's real, whether or not it's simple filtering that can be applied anywhere in the signal chain. Simple experiments, and necessary first steps.
I am not aware that anyone is trying to disprove the effect. We are trying to explain the effect (if it indeed exists) and seek evidence that it does exist.Mark Whitney said:
Some are theorists; some are experimenters. I have been trying to explore exactly what a cap in that position might do. That is my contribution.
At present we are in the position of someone trying to test whether an athlete actually does run faster when he is wearing his lucky socks, but the athlete won't cooperate with the tests as he doesn't want the magic turned off.
What about you will try by yourself (DIY) the same type of lucky socks as the athlete wear (not exactly the same socks he have used...)? If you may feel as an athlete... If not, then you may further watch on TV the running of the athlete with the lucky socks...
I wrote an article about it, with some examples.
http://linearaudionet.solide-ict.nl/sites/linearaudio.net/files/LA Vol 2 Yaniger(1).pdf
If it's something that looks plausible and I'm interested, I test it. Likewise if someone puts out some data to validate their claims and it's something interesting to me (eg., compression levels in MP3, all pass filters...). If it looks questionable and is surrounded by hype and lacks any data, I don't waste my time until at least some data is forthcoming.
http://linearaudionet.solide-ict.nl/sites/linearaudio.net/files/LA Vol 2 Yaniger(1).pdf
If it's something that looks plausible and I'm interested, I test it. Likewise if someone puts out some data to validate their claims and it's something interesting to me (eg., compression levels in MP3, all pass filters...). If it looks questionable and is surrounded by hype and lacks any data, I don't waste my time until at least some data is forthcoming.
See this:
http://www.diyaudio.com/forums/loun...ch-preamplifier-part-ii-6262.html#post4242204
It's not about f response as that can be corrected pre (digital) or post filter.
At the end of the thread, interaction with fet and bjt opamp inputs was raised but didn't really get explored.
However, Joe also used transformers.
...
So, someone with a good ADC should setup 2 circuits, record them, post them and then we can ABX with foobar, or whatever.
😗
http://www.diyaudio.com/forums/loun...ch-preamplifier-part-ii-6262.html#post4242204
It's not about f response as that can be corrected pre (digital) or post filter.
At the end of the thread, interaction with fet and bjt opamp inputs was raised but didn't really get explored.
However, Joe also used transformers.
...
So, someone with a good ADC should setup 2 circuits, record them, post them and then we can ABX with foobar, or whatever.
😗
Last edited:
Oh, sometimes it's fun to hypothesize. I recently read a post, pretty sure it was somewhere else on DIYaudio, discussing audiophiles rolling opamps and getting significantly different sound, apparently because the higher bandwith of their newly installed opamps made them so unstable in the circuit that it oscillated on high-frequency content, thus "adding brilliance to the high end."
I wonder if putting a cap on the output of this DAC gives it a load impedance lower than the DAC is spec'ed for, causing some sort of "better sounding" distortion.
The value of 1uF mentioned in this thread seems obscenely large for something that goes across a line-level signal. It seems to me the rolloff would start in the mid or even bass end of the audible spectrum, and the rest of the audio band would be shorted out.
And seeing how the DAC is a low-power chip made to drive line-level, rather high impedance (at least 1k, maybe 10k and up) loads, the 1uF would be a near-dead-short for much of the audio range, and would likely damage a (voltage-mode) DAC.
On the other had if this is a current-output DAC, the output voltage will be mostly across the the capacitor's series resistance, which if it's an electrolytic may create some nice distortion and indeed sound different and "better" than without the cap.
"...the difference was staggering!" Yet another clickbait headline on the Internet! 😀
I expect that the 1uF caps are non electrolytic, and he does specify current output DAC.
The 1uF is across the balanced output lines.
The outputs are PWM/PDM waveforms which means effectively that supplies are low impedance switched to the output lines.
1uf does indeed seem an extraordinary value to place across the output lines.
This would seem to cause large circulating currents, and consequences perhaps.
It is stated that a specified 1F supercap across the supplies has further (improving) effect.
The Supercap specified quotes 40 ohms internal resistance, so in this application would constitute a handsome Zobel power supply snubber.
I don't doubt that inserting a 1uF cap directly across the output lines of a current output DAC is likely to have some audible effect.
I don't have anything suitable to experiment with this and get some measurements....anybody ?.
Dan.
The 1uF is across the balanced output lines.
The outputs are PWM/PDM waveforms which means effectively that supplies are low impedance switched to the output lines.
1uf does indeed seem an extraordinary value to place across the output lines.
This would seem to cause large circulating currents, and consequences perhaps.
It is stated that a specified 1F supercap across the supplies has further (improving) effect.
The Supercap specified quotes 40 ohms internal resistance, so in this application would constitute a handsome Zobel power supply snubber.
I don't doubt that inserting a 1uF cap directly across the output lines of a current output DAC is likely to have some audible effect.
I don't have anything suitable to experiment with this and get some measurements....anybody ?.
Dan.
The issue is not level but impedance.benb said:
(Apart from amateur NOS DACs) a current-output DAC always feeds a very low impedance. The cap shunts this impedance. There will be very little signal voltage across the cap, so it cannot generate significant distortion however bad it is.
Even before you responded, I was sure that you felt you had cleverly camouflaged your rather nasty shot at Joe in a cloak of plausible deniability, but no one here is so dim to buy it. Everyone knows who you were alluding to. Perhaps, you missed a calling as a politician?
Now, I could ask you the obvious; if you weren't making insulting insinuations toward Joe - on a thread about Joe - then about who, you had to have had someone in mind, and why then make them on a thread anout Joe? But let's bypass all the cheeky B.S., shall we? If you would be so kind as to state that, at least as far as you can personally see, none of what you were writing applies to Joe Rasmussen, then I will retract my calling you out for that.
Do we have an agreement?
It doesn´t seem to me like the cap would cause transient currents, the dac being a current output type. It will provide a better operating condition for the dac by approximating what it wants to see at its outputs: zero ohms. In this case not all the way down to DC, but for a certain range of frequencies, depending on the value and type of cap.
Assuming that the current outputs are symmetrical in amplitude and well timed, you can imagine a virtual ground created right in the middle of the reactance of the capacitor, and each current output of the dac actually sees half the reactance in parallel with whatever follows, I would think.
Also not to be overlooked is the much shorter return path for the HF currents, they don´t need to go through ground anymore - it´s out of the I+ and right back to I- through the cap.
Assuming that the current outputs are symmetrical in amplitude and well timed, you can imagine a virtual ground created right in the middle of the reactance of the capacitor, and each current output of the dac actually sees half the reactance in parallel with whatever follows, I would think.
Also not to be overlooked is the much shorter return path for the HF currents, they don´t need to go through ground anymore - it´s out of the I+ and right back to I- through the cap.
In my case, I use film SMD caps, 1µ at the DAC (ES9018) outputs, and lower value film caps at the preamp inputs. This effect is to be observed at different cap values (under 1µ), but 1µ it was an optimum for my circuits. The effect quality is in my opinion correlated to the cap value.
Last edited:
I note that you've yet to either admit or deny that you were alluding to Joe Rasmussen. So please, take the opportunity to clearly do so. If you do not feel what you wrote to applies to Joe, simply say so. That should be painless enough to say, yes? Plus you get the bonus of having me admit I was mistaken.
Agreed?
Last edited:
- Status
- Not open for further replies.