maybe you should think about what double blind testing does say (everyone jumps on what it can't tell you - proving the negative proposition that no one can hear x)
DBT testing shows that frequency response matching to within 0.1 dB is required to reduce the ability of a large number of listeners to discriminate with statistically significant reliability
some indicate that differences of 0.1 dB over more than 1 octave are sufficient clue to discriminate between systems
So >0.1 dB frequency response differences can be discriminated and contribute to "voicing", some amp reviews have commented on a particular designer’s "signature" x dB lift/rolloff/"presence peak", ect. shaping of amplifier response
DBT testing shows that frequency response matching to within 0.1 dB is required to reduce the ability of a large number of listeners to discriminate with statistically significant reliability
some indicate that differences of 0.1 dB over more than 1 octave are sufficient clue to discriminate between systems
So >0.1 dB frequency response differences can be discriminated and contribute to "voicing", some amp reviews have commented on a particular designer’s "signature" x dB lift/rolloff/"presence peak", ect. shaping of amplifier response
It's not a joke. An equalizer is the right tool in order to obtain these effects. By moving a few sliders you can instantly make the music dark or brigth, punchy or bassy, forward or recessed, thin or thick, articulate or muddy...
You wont't be able to distinguish whatever little modification you do to an amplifier in a blind test, but you will easily distinguish equalisation.
I did live audio mixing for several years (vocals, guitars, wind), so I had to use equalization and compresion extensively. Some people may seem puzzled at what I say, but it's because they have never done serious listening while mixing and equalising. On the other hand, when you are in a venue mixing for a few hundred people you are forced to do *serious* listening and to create effects in order to make the mix "listenable"
You wont't be able to distinguish whatever little modification you do to an amplifier in a blind test, but you will easily distinguish equalisation.
I did live audio mixing for several years (vocals, guitars, wind), so I had to use equalization and compresion extensively. Some people may seem puzzled at what I say, but it's because they have never done serious listening while mixing and equalising. On the other hand, when you are in a venue mixing for a few hundred people you are forced to do *serious* listening and to create effects in order to make the mix "listenable"
Hi beppe,
Varying the feedback ratio changes many things at once. The ratio you choose depends on the open loop gain and how much closed loop gain you want. There are too many factors to come up with a simple statement.
I have found that as you increase the feedback, at some point the music loses it's "dynamics". This may or may no be the case with some amps. I haven't played with everything (hardly any compared to how many designs there are). There are designers that are far more experienced than I am who could better comment here.
The very best thing you could do is play with a design you are experienced with and vary tail currents, feedback and so many other things. Then once you have the amp stabilized again, have a listen.
-Chris
Varying the feedback ratio changes many things at once. The ratio you choose depends on the open loop gain and how much closed loop gain you want. There are too many factors to come up with a simple statement.
I have found that as you increase the feedback, at some point the music loses it's "dynamics". This may or may no be the case with some amps. I haven't played with everything (hardly any compared to how many designs there are). There are designers that are far more experienced than I am who could better comment here.
The very best thing you could do is play with a design you are experienced with and vary tail currents, feedback and so many other things. Then once you have the amp stabilized again, have a listen.
-Chris
Could you back those statements up with any papers? Here, they say the following: "One decibel is close to the Just Noticeable Difference (JND) for sound level."
http://www.phys.unsw.edu.au/~jw/dB.html
Also, note the phon graph toward the bottom, which illustrates perceived loudness over frequency. I know from my own use of a RS meter when adjusting my HT system that values of 0.5 dB are barely audible using the noise tones provided by Avia. 0.1 dB would not be discernable.
I personally think something other than frequency response is what causes perceived differences between amps.
http://www.phys.unsw.edu.au/~jw/dB.html
Also, note the phon graph toward the bottom, which illustrates perceived loudness over frequency. I know from my own use of a RS meter when adjusting my HT system that values of 0.5 dB are barely audible using the noise tones provided by Avia. 0.1 dB would not be discernable.
I personally think something other than frequency response is what causes perceived differences between amps.
Level Matching Thresholds and refs in DBT:
http://www.provide.net/~djcarlst/abx_crit.htm
http://www.provide.net/~djcarlst/abx_peri.htm
maybe you could google: double blind test audio
only 8 million hits but I pulled these from the 1st few pages
http://www.provide.net/~djcarlst/abx_crit.htm
http://www.provide.net/~djcarlst/abx_peri.htm
maybe you could google: double blind test audio
only 8 million hits but I pulled these from the 1st few pages
anatech said:
I prefer to use specific tools when it comes to tune sound. I don't consider amplifiers as "sound editing" tools, but rather the opposite.
Crossovers are also quite fun to play with. Everyone should try seriously speaker building and tuning, instead of focusing only in amplifiers because that's a barely 1% of the story.
ctviggen said:
You need to be very careful with statements like that. Whilst it's true that that it is difficult to hear a change in level of 1dB, it is not true to say that a change in the frequency response of 1dB is similarly difficult to hear - when tweaking active crossovers, I have found differences of 0.2dB to be audible.
Hi Eva,
I took lt cdr data's question to be a question about the "sound" of an amplifier and how to design one, or change an existing one.
-Chris
I agree, but I have little experience with this compared to others.
I took lt cdr data's question to be a question about the "sound" of an amplifier and how to design one, or change an existing one.
I have, but with a given pair of speakers you can still have large differences in the way a system may sound just by replacing the amplifier. The speakers and room have a rather large effect on the sound. That's a given.
-Chris
ctviggen said:
It's always a matter of frequency content. The effects that I mentioned may be achieved both by boosting or cutting existing frequencies or by adding more frequency components to the signal. However, due to the harmonic nature of music signals, any added component is very likely to be already present (so you may just boost or cut it instead).
These effects are based in relative energy contents between different frequency bands rather than absolute values. We hear in a very interactive way, as our sensitivity to each frequency band is continuously modulated by the amount of energy from all the adjacent bands reaching us. I don't know any serious paper about that, but these effects are quite evident and repeatable when equalizing loud music.
Also, in my experience playing with crossover networks and equalisation while other (not involved in audio) people was listening, normal people has a *lot* of trouble perceiving changes in music material smaller than 3dB over narrow frequency bands (1/2 octave or so). In other words, most of them can't determine which is the right tweeter polarity in a 2-way system with a 12db/oct x-over (except when the filter is so well tuned that it produces a huge notch when the polarity is wrong).
The context that I keep hearing voicing being mentioned in is one where the final design is tweeked to obtain a particular sound or spacial characturistic...
It's more of a marketing line that makes the designer sound like his command of his creation is absolute. Actually the process is most likely more trying to keep the design stable and dial out the nasties that that different types of music and loads point out.
That being said, There are a lot of things that affect the sound of a design, but they are fundamental of the design and occur earlier than the final step. Specifically:
The execution of the power supply will determine determine how effectively the power can pass to the output. The ground network creates a reference for the input and feedback network as well as the return path for all of the bypassing added to try and stablize the amp. The layout of these makes a large difference, once you start considering the interface impedance's between stages and the ability of wires and traces to act as antennas.
I don't believe you should be attempting to voice an amplifier, the real efforts should be to experiment with the design to minimise interactions to allow different parts of the music to co-exist with minimal interaction. also to improve stability without relying on bandaid fixes. The sound of a tiny bell coming through clearly in the presence of more powerful parts of the music is just one goal. Impact. Clarity at low listening levels. Lack of congestion on vocals. These are all easily affected by the execution of the design and the environment the circuit functions in.
Build a basic amp as a test bed, find a good source, and start experimenting.
It's more of a marketing line that makes the designer sound like his command of his creation is absolute. Actually the process is most likely more trying to keep the design stable and dial out the nasties that that different types of music and loads point out.
That being said, There are a lot of things that affect the sound of a design, but they are fundamental of the design and occur earlier than the final step. Specifically:
The execution of the power supply will determine determine how effectively the power can pass to the output. The ground network creates a reference for the input and feedback network as well as the return path for all of the bypassing added to try and stablize the amp. The layout of these makes a large difference, once you start considering the interface impedance's between stages and the ability of wires and traces to act as antennas.
I don't believe you should be attempting to voice an amplifier, the real efforts should be to experiment with the design to minimise interactions to allow different parts of the music to co-exist with minimal interaction. also to improve stability without relying on bandaid fixes. The sound of a tiny bell coming through clearly in the presence of more powerful parts of the music is just one goal. Impact. Clarity at low listening levels. Lack of congestion on vocals. These are all easily affected by the execution of the design and the environment the circuit functions in.
Build a basic amp as a test bed, find a good source, and start experimenting.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.