Best of both worlds

[JerryLove]

Frequency amplitude is not the only type of audible coloration. You can have two units with identical frequency response and measured sine/sweep THD and they can still sound (and measure) very different when playing music.

So they measure identically on all frequencies, but measure differently playing frequencies?!?

This sounds like snake-oil, and is debunked by every electrician and amplifier builder I've felt any need to trust. It also fails to completely to jibe with my own experience with amplifiers. Please prove your claim by showing me the measurements you are referencing.

No. a wavefrom is a waveform. A well built amplifier makes no changes to the shape of the waveform (that come anywhere near audiability) and therefore has no sound.

http://www.roger-russell.com/truth/truth.htm#goodamplifiers

In doublie-blind tests: no one can tell the difference between well-built amplifiers.

 

[Josuah]

So they measure identically on all frequencies, but measure differently playing frequencies?!?

This sounds like snake-oil, and is debunked by every electrician and amplifier builder I've felt any need to trust. It also fails to completely to jibe with my own experience with amplifiers. Please prove your claim by showing me the measurements you are referencing.

I didn't say they would measure different amplitudes for different frequencies. I said that two designs with the same frequency response (white noise or sweep measurement) and THD measurements (sine wave or sweep measurement) can sound and measure different when playing music. But I did not mean frequency response or THD measurements, because those are done using white noise, sweep, or sine waves.

IMD or multi-tone THD measurements are more interesting. But often not done or even considered. A multi-tone THD of many tones which are not multiples of each other is very useful because it can help expose weaknesses in the feedback design or "response rate" (for lack of a better term) of the circuit.

Unfortunately I only know of a few audio gear two-tone IMD measurements that are sometimes done by JEJ over at Secrets. But maybe this paper will be of interest to you: A New Method for Measuring Distortion using a Multitone Stimulus and Non-Coherence.

From the paper (this is what I was getting at):

It is almost impossible to compare the different distortion measurement techniques (namely one tone, two tones, multitone, and noise) because the distortion obtained is very dependent on the crestfactor and the spectral content of the stimulus.

Another useful test would be pre- and post-ringing of an impulse signal. There is no standardized test/measurement for this, but you could measure it if you wished.

 

[JerryLove]

I didn't say they would measure different amplitudes for different frequencies. I said that two designs with the same frequency response (white noise or sweep measurement) and THD measurements (sine wave or sweep measurement) can sound and measure different when playing music. But I did not mean frequency response or THD measurements, because those are done using white noise, sweep, or sine waves.

Or pink noise. In fact randomized noise is a pretty common calibrating tool.

IMD or multi-tone THD measurements are more interesting. But often not done or even considered. A multi-tone THD of many tones which are not multiples of each other is very useful because it can help expose weaknesses in the feedback design or "response rate" (for lack of a better term) of the circuit.

It's really more simple than you make it sound. Coloration will cause the waveform out to look different from the waveform in. It simply doesn't happen in well-built solid-state amps (unless they are driven too high). That's why SS amps don't have a sound. That's why no DBT have ever found an ability to differentiate level-matched SS amps by ear.

 

[WmAx]

I may have spoken out of turn. I should have simply stated in IMHO. not to turn this into a guitar amp forum but I find it interesting the vast number of SS guitar amps that are "modeling amps". They all have the same "classic tube amp" emulators. Funny they are all seeking that tube amp sound. I've owned various ( and currently am using one) and they just don't get there. But that could be a product of my playing.
Back to the original topic. Thanks for the replys all. I am starting to get the sense from speaking people that, money spent on speakers will go farther to get better sound quality out of my turntable given my current set up. I'll most likely start there. Which I'm sure will lead to questions on what speakers to go with.

Most tube modelers, it seems, do not sound right because they are not fully emulating tube behavior. A digital tube model usually takes a 'snapshot' of the harmonic distortion profile (along with frequency response) of a tube they want to emulate, then applies that profile to the signal. But, here is the problem: a real tube does not have a 'single' profile. As you saturate the tube, the distortion profile actually changes the harmonic structure dynamically, with each tube having it's own specific 'curve' of change. So by not taking into account this property of real tubes, the models will sound 'flat', especially with blues/jazz, etc,. that the player plays a wide dynamic range near the distortion thresholds, varying the distortion as they play by substantial degree. Now, there are modelers that sound exactly, or virtually exactly, like real tubes, but they are rare. The newest rack mount Digitech modeler takes into account the full dynamic behavior of tubes, as well as there recently discontinued GNX3000 floor processor (this is not related to the GNX3, GNX4, etc,; these use totally different processing that does not take into account the dynamic tube variables). Now, to use these sophisticated modelers and have them sound like the real thing, you ironically must not use traditional guitar cabinets. You should use the most accurate PA monitors and high power solid state pro amps. Because these modelers can accurately replicate every facet of the chain (tube distortion, cabinet resonances, guitar speaker responses, etc.), but only if fed through a low distortion accurate chain. Very few modelers copy tube behavior accuratel6y. Last I checked, even the latest version of the very popular Line 6 Pod series did not yet do it. This is okay for heavier rock/heavy metal, etc.; that uses high gain/distortion all the time. Because the distortion profile does remain more or less similar in these cases. But with dynamic playing that uses a wide range, like jazz, blues, etc.; the inaccurate tube modeling is very obvious.

-Chris

 

[jostenmeat]

That is very interesting Chris, thanks for sharing. May I ask where you read/learned about this stuff? (Do AES papers talk about guitar gear or something?). Interesting!