I get what you are saying, Casey, but I think you aren't considering HOW the signal gets transferred in each device. The mechanism for electron transmission through a vacuum tube is inherently "messier" than solid state (again, I am not saying that it sounds bad, just that it is messy!). Even if there are fewer components (which is not always the case--take the gainclones for example[and yes, I know counting the IC as one component is cheating! 😀]), the characteristics of these components must be taken into account. As an example, one random inductor inline somewhere in your amp will cause massive amounts of havoc to your signal, though it may only be one component.valveitude said:
I think that any aurally attentive designer will come to realise that components affect the resultant sound quality and will eventually similarly conclude that less of them* in the signal path, would therefore be better.
But pulling back for a moment: this still means that they are altering the sound in someway (the "original waveform" for arguements sake). If that's the case, then we're still back at the original question of asking why the measurements that we're taking don't seem to adequately show this?
* = "less of them" in the sense of the quote by Einstein: "Everything should be made as simple as possible but no simpler"
But pulling back for a moment: this still means that they are altering the sound in someway (the "original waveform" for arguements sake). If that's the case, then we're still back at the original question of asking why the measurements that we're taking don't seem to adequately show this?
* = "less of them" in the sense of the quote by Einstein: "Everything should be made as simple as possible but no simpler"
Majestic said:
That's exactly the way I see it. In an attempt to express the problem in a way that leads to a solution I might say that one measuring system (our hearing) detects something while the other doesn't. That's why it seems to me that studying the measurement system that detects something might be more fruitful than studying the measurement system that detects nothing.
Ok, it's unfair to say it detects nothing, but you get my point.
-- Dave
But then we get back to the question, what are we hearing, if anything? No offence intended, but the ear/brain system is very easily fooled.
Before we can come up with any accurate system for measuring equipment, we have to know how and if the human hearing system perceives these events.
Before we can come up with any accurate system for measuring equipment, we have to know how and if the human hearing system perceives these events.
pinkmouse said:
True, but given the fact that so many people hear the same thing, I think it's safe to assume that we're not being fooled, or that we're all being fooled in the same way, which is just as interesting.
Right.
Dave Cigna said:
Which goes right to the heart of Milgram's work on peer pressure and authority figures.
Not in my experience. I've never ever been fooled into thinking that reproduced sound was real except for a few times when I thought my mobile phone was ringing...but it was an ad on my car radio.
The eye/brain system on the other hand is easily fooled. Especially when looking at square waves!!!..Often giving the brain the illusion that the amp will sound good.😀
Of course a lot of people discovered tubes despite peer opinion and marketing.pinkmouse said:
(Granted, once we're here... 😛 )
pinkmouse said:
I disagree, and that's why I think these threads always get lost. Nobody distinguishes between quantitative analysis and qualitative analysis. EC starts a thread about quantitative analysis and people drift off into debating qualitative analysis.
The best example I know of for elucidating between the two is the old spy game. The spook agencies would have two departments of cryptography, qualitative analysis, and quantitive analysis. Qualitative analysis, obviously, was trying to decode all the coded messages. Quantitative analysis was about trying to correlate all the measurable things: "OK, everytime, there is a long message on this frequency with this pattern near the beginning, this spy we follow has a busy week, therefore this message is for him." They do that without ever knowing what the message said.
And that is the point. Let's focus on quantitative analysis. What measurements have people found to correlate with what they think is a good sounding amp. Let's not try to figure out the "code", the "how" or "why" it sounds better.
Hi dfdye,
[
Hmm…correct me if I’m wrong here, but isn’t a PN junction the heart of most random noise generators? I know there are countless mechanisms to consider in each devices method of plumbing electrons, but this statement seems a little absolute to me. For an admitted simpleton such as myself, heating atoms up so that the valence bonds weaken allowing the electrons to be sucked off by a opposite potential seems pretty straight forward to me.
I think this statement goes beyond “cheating”
.
You’ve arrived at the heart of my argument. EVERY passive component in the circuit is a “random inductor” to some degree( as well as a random capacitor, ect.) . Minimizing the number of said components is a good thing.
Again, I’m not saying that simple is ALWAYS the best approach. I have heard SS gear that I could live with (given the right associated gear…a Krell\ Apogee combo comes to mind), in fact the system I’m putting together now will have a SS amp on the sub.
[
Hmm…correct me if I’m wrong here, but isn’t a PN junction the heart of most random noise generators? I know there are countless mechanisms to consider in each devices method of plumbing electrons, but this statement seems a little absolute to me. For an admitted simpleton such as myself, heating atoms up so that the valence bonds weaken allowing the electrons to be sucked off by a opposite potential seems pretty straight forward to me.
I think this statement goes beyond “cheating”
.You’ve arrived at the heart of my argument. EVERY passive component in the circuit is a “random inductor” to some degree( as well as a random capacitor, ect.) . Minimizing the number of said components is a good thing.
Again, I’m not saying that simple is ALWAYS the best approach. I have heard SS gear that I could live with (given the right associated gear…a Krell\ Apogee combo comes to mind), in fact the system I’m putting together now will have a SS amp on the sub.
leadbelly said:
That has already been discussed for decades. Lots of good has come out of it, but by my assessment it has gone only so far. There are certain aspects of a good sounding system that don't seem to correlate with any of the measurements that have been made so far. It seems we need to find new measurements or ways to interpret the old measurements. That's the point as far as I can tell.
Why not? That seems to me to be the ultimate goal. If you mean that we should find the quantitative measurement first, then use the data to figure out the how and why, I understand that, but the how and why is ultimately what we're after, isn't it?
leadbelly said:
But we can't even agree on that - you might say SE valve, I might say Krell, and someone else might say Gainclone...
Dave Cigna said:
You're taking my statement out of context. Of course there's nothing wrong with finding the "how and why" and that is a great thing for everybody, like breaking the code in my analogy. The problem is when one thwarts attempts at quantitative analysis by stating that quantitative analysis is worthless without qualitative understanding. It's not, as you stated.
pinkmouse said:
The point is to identify what measurements correlate well with the sound of an individual amp. You have to try to keep as many variables as constant as possible.
Quoting SY here. We should probably start from here. Anyone anything QUANTITATIVE to add?
Gianluca
"First, the basic measurements, THD, IM, power bandwidth. If an amp can't perform OK on easy stuff like this, it will be a disaster on music.
Stability. I use square wave tests into resistive loads, RC combinations, and actual speakers. This is done at a few different levels and a few different frequencies.
Stability after overload. For this, I use gated sine wave bursts, set for a level a bit above clipping. Again, I try to use a few different loads. The silence following a burst will hash up if there's any dynamic stability issue. It does require a tone-burst generator or a very clever gating circuit following a sine wave generator.
Blocking. For this, I use a similar waveform, but with a sine wave below clipping filling in the "spaces." Amps with poor blocking behavior (and this is nearly universal in RC-coupled feedback amps) will show very evident choking between clipping bursts. The lower-level sine wave may show very obvious crossover distortion or an exponential envelope. Bad. I generate this waveform with a function generator and a tone burst generator. Very jury-rigged, but it works."
Gianluca
"First, the basic measurements, THD, IM, power bandwidth. If an amp can't perform OK on easy stuff like this, it will be a disaster on music.
Stability. I use square wave tests into resistive loads, RC combinations, and actual speakers. This is done at a few different levels and a few different frequencies.
Stability after overload. For this, I use gated sine wave bursts, set for a level a bit above clipping. Again, I try to use a few different loads. The silence following a burst will hash up if there's any dynamic stability issue. It does require a tone-burst generator or a very clever gating circuit following a sine wave generator.
Blocking. For this, I use a similar waveform, but with a sine wave below clipping filling in the "spaces." Amps with poor blocking behavior (and this is nearly universal in RC-coupled feedback amps) will show very evident choking between clipping bursts. The lower-level sine wave may show very obvious crossover distortion or an exponential envelope. Bad. I generate this waveform with a function generator and a tone burst generator. Very jury-rigged, but it works."
and
You know Cheever did a thesis on a measurement.
http://w3.mit.edu/cheever/www/cheever_thesis.pdf
And even if SY would have failed him on that thesis had he been on his commitee....It strikes me that allegedly his TAD I think it was called...did better a job than THD and IMD at predicting how humans would experience the amps.
However "shoddy" the thesis was it sounds like something to try and replicate. And if the experiment's results can't be reproduced... At least then someone can come out and say it's trash. If the results can be reproduced we have a new measurement method that better correlates with perceived sound quality...how hard can this be?
It is fairly simple, but electron transfer via solid state electronics is much more well regulated and doesn't create as many noise issues. That has been established electronics dogma since before I started dealing with any of these issues.valveitude said:
Of course it does! 😀 But let's keep in mind that the noise generated by good IC's is vanishingly low in all harmonics. With this assumption, let's look at your next argument. . .I think this statement goes beyond “cheating”
. . . . and this is where solid state components have a pretty good advantage (with regard to noise generation): the random noise generated by them is astoundingly small. As such, as you simplify to the point of a gainclone, as many noisy components have been removed from the system as possible.
I also am not saying that tube gear should be scrapped! I just think it is noisy and adds layers to the music that aren't there in the original (there are obviously exceptions to this that have been pointed out in the thread already, but as a general rule, the "tube sound" contains some level of audible distortion at some harmonic). Some people like it and some people don't. I happen to be on the fence, but to say that "tubes are more accurate than solid state amps" is by and large a false claim. Whether it is more MUSICAL is entirely subjective!
If noise is the issue, I'd recommend against taking in live music.
As far as the Gainclone goes, I use one and advise that everyone build one just to put boutique audio(tubed or SS) into perspective. However, it does not rival, by any means, the better tube amps that I have heard. It does better most commercial SS gear.
At the level of small signal gear, I can't imagine why anyone would chose SS over a good tubed circuit. That one is beyond me in my experience.
As far as the Gainclone goes, I use one and advise that everyone build one just to put boutique audio(tubed or SS) into perspective. However, it does not rival, by any means, the better tube amps that I have heard. It does better most commercial SS gear.
At the level of small signal gear, I can't imagine why anyone would chose SS over a good tubed circuit. That one is beyond me in my experience.
This is exactly the problem.
Whilst I completely appreciate, and share EC8010's wish to have a nice set of tests and universally applicable rules to build the perfect amp, it all falls down when we have to take into account the mushy grey stuff. As long as we have such definate opinions as expressed above, no measurement metric will be approved by all, and is therefore worthless.
Whilst I completely appreciate, and share EC8010's wish to have a nice set of tests and universally applicable rules to build the perfect amp, it all falls down when we have to take into account the mushy grey stuff. As long as we have such definate opinions as expressed above, no measurement metric will be approved by all, and is therefore worthless.
- Status
- Not open for further replies.