What speakers for $2,000?

[johnt]

If you don't need the utmost accuracy, even this will do:

http://www.amazon.com/Dayton-Audio-iMM-6-Calibrated-Measurement/dp/B00ADR2B84

 

It comes with a calibration profile so that its response is compensated for. I use it on my phone and tablet, but it also works on PCs with an extension.

 

This is what I use on my PC:

http://www.amazon.com/miniDSP-UMIK-1-Measurement-Calibrated-Microphone/dp/B00N4Q25R8

 

It is a very cheap investment for something that replaced this:

http://www.amazon.com/AudioControl-SA3055-Real-Analyzer-Meter/dp/B001UDM3LA

 

The iMM-6 looks very reasonable. Is there anyway I can do the calibration without a phone? I do not want to tie anything to a mobile device.

[Stagea]

The iMM-6 looks very reasonable. Is there anyway I can do the calibration without a phone? I do not want to tie anything to a mobile device.

 

You can use your PC. Every mic comes with a calibration file that you can input into REW when you're using the said microphone.

 

You can buy something like this so that you can plug it to your notebook and position the mic as needed:

http://www.amazon.com/StarTech-com-MUHSMF2M-Position-Headset-Extension/dp/B008DWGLLO

 

One done, get your pink noise ready. Your target response curve should fall by 3dB for every increase in octave (that is just how pink noise is). This will equate to a flat response with white noise.

 

There are many ways to skin a cat, but here's among the simplest ways to use the tool correctly on a 2ch system:

1. Do the measurement when it is very quiet (early mornings normally work).

2. Run the speaker to be measured in direct (disable other sources of sound).

3. Play pink noise and set the volume control to a nominal measurement level with only one channel running (approximately 75dB-80dB at your listening position in most cases; equivalent to a phone's dial tone with the handset on your ear).

4. Get rid of anything rattling. Fix speaker positioning (placement and toe) and room acoustics to get closest to your target response (no big peaks and/or deep suck-outs); measure each channel individually and get a good compromise.  

5. Fade the output to the channel that's closer to correct.

6. Cut the sharp peaks with the PEQ using high Q filters. Prioritize the tallest and widest ones.

7. Level out the wide curve undulations by attenuating the bumps using low Q filters.

8. Fade to the other channel then validate if the same PEQ curves can apply (validate if it's worth EQing independently). 

9. Enable both channels and listen to the result using a familar track. If the sound is obviously uneven, go to step 10 (otherwise, proceed to Step 11).

10. Measure with both channels on and mildly boost wide dips in response using low Q filters. Use sparingly as this eats into your headroom.

11. Ignore the effects of comb filtering as this is normal with both channels on.

12. EQ your sub independently using steps 3, 4, 5, 6 and 7.

13. Enable all channels (with sub) and turn on your crossover (switch out of direct mode).

14. Tweak the levels and crossover points to achieve a smooth curve, then validate with a frequency sweep.

15. Tweak the tilt of the curve to your liking with tone controls (very low Q filters). You can also build-in your "house curve" at this point.

16. Enjoy the music.

[Whit3Rav3n]

You can use your PC. Every mic comes with a calibration file that you can input into REW when you're using the said microphone.

 

You can buy something like this so that you can plug it to your notebook and position the mic as needed:

http://www.amazon.com/StarTech-com-MUHSMF2M-Position-Headset-Extension/dp/B008DWGLLO

 

One done, get your pink noise ready. Your target response curve should fall by 3dB for every increase in octave (that is just how pink noise is). This will equate to a flat response with white noise.

 

 

There are many ways to skin a cat, but here's among the simplest ways to use the tool correctly on a 2ch system:

1. Do the measurement when it is very quiet (early mornings normally work).

2. Run the speaker to be measured in direct (disable other sources of sound).

3. Play pink noise and set the volume control to a nominal measurement level with only one channel running (approximately 75dB-80dB at your listening position in most cases; equivalent to a phone's dial tone with the handset on your ear).

4. Get rid of anything rattling. Fix speaker positioning (placement and toe) and room acoustics to get closest to your target response (no big peaks and/or deep suck-outs); measure each channel individually and get a good compromise.  

5. Fade the output to the channel that's closer to correct.

6. Cut the sharp peaks with the PEQ using high Q filters. Prioritize the tallest and widest ones.

7. Level out the wide curve undulations by attenuating the bumps using low Q filters.

8. Fade to the other channel then validate if the same PEQ curves can apply (validate if it's worth EQing independently). 

9. Enable both channels and listen to the result using a familar track. If the sound is obviously uneven, go to step 10 (otherwise, proceed to Step 11).

10. Measure with both channels on and mildly boost wide dips in response using low Q filters. Use sparingly as this eats into your headroom.

11. Ignore the effects of comb filtering as this is normal with both channels on.

12. EQ your sub independently using steps 3, 4, 5, 6 and 7.

13. Enable all channels (with sub) and turn on your crossover (switch out of direct mode).

14. Tweak the levels and crossover points to achieve a smooth curve, then validate with a frequency sweep.

15. Tweak the tilt of the curve to your liking with tone controls (very low Q filters). You can also build-in your "house curve" at this point.

16. Enjoy the music.

Although I would recommend for him to just use REW. It already has a built in EQ function where a desired target response can be chosen and it does the job pretty well. Also noise signals are not really used anymore for measurements since the only advantage over sine sweep was the lower computational complexity, but these days it is not an issue anymore.

 

So my advice would be to do a measurement in REW, set a desired target response in the eq window, create the filter and export the filter impulse response. Preferably measure left and right side separately then create a stereo filter impulse response.

[Stagea]

Although I would recommend for him to just use REW. It already has a built in EQ function where a desired target response can be chosen and it does the job pretty well. Also noise signals are not really used anymore for measurements since the only advantage over sine sweep was the lower computational complexity, but these days it is not an issue anymore.

 

So my advice would be to do a measurement in REW, set a desired target response in the eq window, create the filter and export the filter impulse response. Preferably measure left and right side separately then create a stereo filter impulse response.

I completely forgot about that feature. Thanks for bringing that up. 

 

My old habits die hard. I normally work on fixed function and programmable external processors that don't normally take in correction files, so I generally disregard most of REW's functions and do things manually (I mostly use REW for frequency response and spectral decay measurements).

 

 

I do not want to tie anything to a mobile device.

 

 

I've never used Equalizer APO, but here's a GUI for it. It might make it nicer to use.

http://sourceforge.net/projects/peace-equalizer-apo-extension/

 

I just searched, and here's someone's before and after measurement after using REW's correction file on Equalizer APO (flat target curve):

http://techtalk.parts-express.com/attachment.php?attachmentid=41734&d=1384110403

 

I probably would've EQed the system a little differently if I did it manually (I care a lot about peaks as they stick out more than averages, so I'd likely apply less boost on that 1.7 - 3.5kHz range), but the results are pretty good.

[johnt]

Yesterday was a hot day. Temps in my part of California were over 110°F or a very uncomfortable 43°C for the European crowd. It was also an "energy savings day," which means we are encouraged not to use air conditioners. I couldn't go anywhere near my heat generating gear that includes my tv, pc, and space heater (receiver). I'll have to play around with REW tonight if I can.

 

  All I know is that it's going to be a loud weekend.

[Stagea]

Yesterday was a hot day. Temps in my part of California were over 110°F or a very uncomfortable 43°C for the European crowd. It was also an "energy savings day," which means we are encouraged not to use air conditioners. I couldn't go anywhere near my heat generating gear that includes my tv, pc, and space heater (receiver). I'll have to play around with REW tonight if I can.

 

All I know is that it's going to be a loud weekend.

 

Where in California are you at? I used to live in Benicia. 

 

Enjoy your toys over the weekend. That'd be a great preparation for your new speakers (though you'd recalibrate once they arrive).

[johnt]

Where in California are you at? I used to live in Benicia. 

 

Enjoy your toys over the weekend. That'd be a great preparation for your new speakers (though you'd recalibrate once they arrive).

 

My wife works in Benicia! What a small world.

 

I live on the outskirts of Vacaville. It was a quiet little town 10-15 years ago. It's gotten so congested. 

[Stagea]

My wife works in Benicia! What a small world.

 

I live on the outskirts of Vacaville. It was a quiet little town 10-15 years ago. It's gotten so congested. 

 

Neat!

 

Yes, I remember Vacaville for the outlet stores along I-80. Younger folks like going there. That was a long time ago, and I am sure that a lot had changed.

[johnt]

Neat!

Yes, I remember Vacaville for the outlet stores along I-80. Younger folks like going there. That was a long time ago, and I am sure that a lot had changed.

Yeah. They've added a Best Buy and like ten fancier fast food shops. PetCo with hecka cute kittens.

Where have you moved to?

[Stagea]

Yeah. They've added a Best Buy and like ten fancier fast food shops. PetCo with hecka cute kittens.

Where have you moved to?

 

Very nice. I am sure property prices have gone up quite a bit too.

 

I moved to the Philippines, though I am frequently in the US. I will be in Texas from October to January, for example.

[johnt]

Very nice. I am sure property prices have gone up quite a bit too.

 

I moved to the Philippines, though I am frequently in the US. I will be in Texas from October to January, for example.

 

I shouldn't complain then. 113 degrees in Texas is a cold day 

 

Let me know if you're ever in Benicia again. We can have a loudspeaker party in my backyard and upset the neighbors! Plus a BBQ!

[Stagea]

I shouldn't complain then. 113 degrees in Texas is a cold day 

 

Let me know if you're ever in Benicia again. We can have a loudspeaker party in my backyard and upset the neighbors! Plus a BBQ!

 

Thanks for the invite, John. That sounds delicious and rebellious. I will let you know if I'm in town (probably not anytime soon though).

 

Ivan

[Stagea]

I can no longer find your old thread, but here's a comparison between typical amplifiers when it comes to noise floor:

 

AV Receiver:

 

HiFi Amplifier:

 

Typical PC Speaker amps are normally in the -70db to -90db range in noise floor (hence the frequently audible hum or hiss).

[johnt]

I can no longer find your old thread, but here's a comparison between typical amplifiers when it comes to noise floor:

 

AV Receiver:

 

 

HiFi Amplifier:

 

 

Typical PC Speaker amps are normally in the -70db to -90db range in noise floor (hence the frequently audible hum or hiss).

 

Why are there massive spikes every 1,000kHz or so? I'm also surprised to see about a 10 dB difference between channels. Does the source say just how expensive each amplifier was... or are these your graphs?

 

I generally get a little humming from my subwoofers when I put my ear very close to the dust cover. I thought I had some sort of ground loop issue at first, but it isn't audible more than a foot away, and my subs only use 2 prong plugs without the ground... so it can't really be a ground loop. I just figured the magnets are massive and charged, there has to be some noise. Otherwise, my Klipsch speakers are shockingly quiet at "idle." 

[Stagea]

Why are there massive spikes every 1,000kHz or so? I'm also surprised to see about a 10 dB difference between channels. Does the source say just how expensive each amplifier was... or are these your graphs?

 

I generally get a little humming from my subwoofers when I put my ear very close to the dust cover. I thought I had some sort of ground loop issue at first, but it isn't audible more than a foot away, and my subs only use 2 prong plugs without the ground... so it can't really be a ground loop. I just figured the magnets are massive and charged, there has to be some noise. Otherwise, my Klipsch speakers are shockingly quiet at "idle." 

 

Both amplifiers are fed a 1kHz sine wave (so both will show 0dB at 1kHz... as that is the source signal). The succeeding peaks are harmonics. A perfectly clean amplifier will have a low noise floor (the level between the peaks) and have zero harmonic (no peaks aside from 1kHz).

 

We are comparing an $800 AV receiver and a $95,000 HiFi Separates (price for a 2 channel preamp and power amp). AV Receivers are clean enough for most applications. 

 

The cleanest Direct Digital gear can show noise floors in the -160dB range. These things can theoretically take advantage of 32-bit (or higher) recordings. Unfortunately, recordings generally don't resolve that much information. It does allow for the least amount of loss from the point of content creation to reproduction, however.

[johnt]

Both amplifiers are fed a 1kHz sine wave (so both will show 0dB at 1kHz... as that is the source signal). The succeeding peaks are harmonics. A perfectly clean amplifier will have a low noise floor (the level between the peaks) and have zero harmonic (no peaks aside from 1kHz).

 

We are comparing an $800 AV receiver and a $95,000 HiFi Separates (price for a 2 channel preamp and power amp). AV Receivers are clean enough for most applications. 

 

The cleanest Direct Digital gear can show noise floors in the -160dB range. These things can theoretically take advantage of 32-bit (or higher) recordings. Unfortunately, recordings generally don't resolve that much information. It does allow for the least amount of loss from the point of content creation to reproduction, however.

 

Holy smokes man. I can't believe it takes that much money for a quieter noise floor.

 

Just to make sure I understand, the noise floor is measured when an amp is turned on but not sending anything to the speakers, correct? Why is it important?

[Stagea]

Holy smokes man. I can't believe it takes that much money for a quieter noise floor.

 

Just to make sure I understand, the noise floor is measured when an amp is turned on but not sending anything to the speakers, correct? Why is it important?

 

When you look at THD + N ratings, that is often done by feeding a sine wave input. All the output aside from the amplified sine wave will be either noise or harmonic distortion. This however does not represent the audibility of waveform distortion very well (such as gain compression, crossover distortion, etc.).

 

Aside from clean numbers in terms of THD + N, higher end amplifiers also do not compress as much under load. Gain compression can be very evident in less capable amplifiers (especially at higher volumes, or when driving more difficult speakers). The distortion of the peaks noticeably reduces the lifelikeness of a recording. The amount of this distortion is also not consistent, as it varies by signal strength and frequency depending on the speaker's resistance and reactance. 

 

Crossover distortion is another thing to look at as this often results in a perceived "harshness" of the sound. This is because of higher order harmonic content that's highly typical of this phenomenon.

 

And as with headphone amps, the input and output impedance are factors for speaker amplifiers (to get sufficient damping). Unlike headphone amplifiers however, slew rate is a bigger factor in home amplifiers (as we are working with larger output signals).

 

Noise floor is important as the faintest signals get lost in the noise floor. A signal has to be higher than the noise floor to be perceptible. The range between your lowest perceptible signal and your maximum level is what is used to determine your dynamic range. This is what some people refer to as sonic contrast, it is equivalent to your display's contrast ratio. A low noise floor is equivalent to a deep black level.

[SSL]

Holy smokes man. I can't believe it takes that much money for a quieter noise floor.

 

Just to make sure I understand, the noise floor is measured when an amp is turned on but not sending anything to the speakers, correct? Why is it important?

 

Noise floor is always there, even when a signal is present. If it is too high, it will be audible, and detract from sound quality; any part of the signal that is below (quieter than) the noise floor is lost.

 

Exceptionally low noise floor is unnecessary, as is 24-bit or higher music. Actually utilizing such dynamic range would require turning up the listening volume past the threshold of death.

[Stagea]

Noise floor is always there, even when a signal is present. If it is too high, it will be audible, and detract from sound quality; any part of the signal that is below (quieter than) the noise floor is lost.

 

Exceptionally low noise floor is unnecessary, as is 24-bit or higher music. Actually utilizing such dynamic range would require turning up the listening volume past the threshold of death.

 

Most people do not listen to material at a level that takes advantage of higher bit depth recordings, nor do we have the hearing ability to take advantage of the bandwidth from higher sampling rates.

 

However, higher res recordings can help produce cleaner and more accurate waveforms well below those limits. The increased number of data points per unit time helps the DAC in plotting the right output waveform.

[SSL]

Most people do not listen to material at a level that takes advantage of higher bit depth recordings, nor do we have the hearing ability to take advantage of the bandwidth from higher sampling rates.

 

However, higher res recordings can help produce cleaner and more accurate waveforms well below those limits. The increased number of data points per unit time helps the DAC in plotting the right output waveform.

 

Does that have to do with "brick wall" filtering? Re-sampling? I'm not real clear on the audible effects of those factors, but I hear their avoidance often touted as an advantage of high bitrate audio.

[Stagea]

Does that have to do with "brick wall" filtering? Re-sampling? I'm not real clear on the audible effects of those factors, but I hear their avoidance often touted as an advantage of high bitrate audio.

Brick wall filtering is just applying a low pass filter on the output before the Nyquist frequency to filter out the upper range aliasing. Though this may have an effect in output, it is not the main reason for high res audio.

It is true that 44.1kHz / 48kHz sampling already captures all the data points that the typical ear can decipher in terms of frequency range. It is also true that 16-bit already captures all the data points needed for typical playback levels. DACs however cannot reliably connect those data points across the entire frequency range and across different amplitudes.

A DAC is often compromised to make a decent connection between the data points to approximate the intended signal. The curve and slope of that connection between the data points should vary depending on the amplitude and the frequency of the signal (to get close to the original waveform), and that is where standard resolution falls behind high res. DACs normally connect those points at the steepest slope possible to not cause any bandwidth restriction at the top end (yes, most DACs do not exactly follow the nyquist-shannon sampling theorem for reconstruction). At lower amplitudes and/or higher frequencies, this results in a jagged waveform. High res recordings offer enough data points to highly reduce those estimation (interpolation) errors, resulting in a clean waveform that is closer to the original. This is why DACs measure better when fed with high res recordings.

1kHz standard res data (16-bit 44kHz); output as seen with an oscilloscope:

Increasing the sample rate will not change the waveform in this case, as 16 bit is limited in precision for this measurement (though it can straighten the jaggies).

The same dac fed with high res (24-bit 96kHz) 1kHz:

The distortion generated by processing standard res had increased unintended output all over the frequency band by about 30dB:

This is actually a pretty good result. Some DACs do much worse. As you can see, this DAC enjoys a 40dB separation between the output waveform and the noise floor when running at 16 bit (and thus SNR for a 6dB signal). At the same output level (54dB in a 24-bit system normalized to the same output peak), the output wave form gets separated by 70dB from the noise floor.

Do note that this measurement is done on the last bit of resolution for 16 bit (which is 6dB of its 96dB range). Generally, the lower the intensity of the passage, the worse 16-bit fares. The higher the frequency of the sound (or the more complex the sound), the worse lower sampling rate recordings fare. Highly compressed mainstream materials do not have the dynamic range or the complexity to reveal the advantage of high resolution formats.

Here's what 20kHz looks like when comparing CD sample rate (44.1kHz) and high sample rate (192kHz) at the same bit depth (32 bit float):

The DAC has to be really good to recognize the sequence and try to recreate the original waveform if it's only getting 2.2 samples per cycle (44.1 divided by 20). Oftentimes, the DAC will not do it right. Even if you increase the resolution to 64-bit, if the sampling frequency is not sufficient then it will not do any good (you still get 2.2 samples per cycle).

Fewer samples also mean worse impulse/transient response (because data and peaks escape between samples). We don't listen to continuous test tones...

Digitization of audio data (analog to digital conversion) is always a "lossy compression" of the source (I'm not talking about digital to digital data compression). The source is an infinite stream of data (that is infinitely large), and it is just being "sampled" to be stored/manipulated digitally. The conversion from digital to analogue is an estimation of the original based on the sample points provided. It is similar to surveying an infinite population to get data. A higher bit depth is having a greater precision of measurement per sample, while a higher sampling rate is having more samples. The better your measurement per sample and the more samples you have, the more reliable your data (and the better you can recreate how the population is like). Even a perfect digital system (which obviously does not exist) will only recreate a perfect waveform from a perfect sinewave source; since real-world sound outside of a tone generator does not follow a perfect sine wave (but a very complex waveform)... additional data points help if you're after reproducing the source sound. PCM Audio playback is just interpolation (estimation) to recreate the waveform from saved data points, like how a JPEG file works to recreate the original image. The fewer the data points you have (higher JPEG compression), the further you may get from the original. Signal bandwidth is not equal to signal content.

On the content creation end, the extra resolution allows less dynamic compression to be used (less compromise in quality as lower level signals become more usable, since they stay above the reduced noise floor).

Some are also easy to dismiss DSD as inferior to PCM. Though PCM is easier and cheaper to manipulate and can offer a very high mathematical resolution, DSD has its own share of advantages (aside from impulse response, which is obvious) -- particularly when it comes to fewer conversion losses from recording to playback (no need for decimation during recording and delta sigma modulation during playback). This is what's driving the higher order DSDs in some applications (now reaching DSD 2048).

I see no excuse to not move up to higher resolution formats (aside from production cost and storage). Detractors of high res audio that claim that the audio information beyond 20kHz degrades audio quality can always apply the same low pass filter used on lower resolution digital audio data. There is nothing stopping them from doing just that.

The saying that high res does not make sense because we can only hear so much frequency range does not hold much water for me; it's as good as saying that the Xbox One is as good as a high end PC for gaming when you're just using a 1080p screen. The quality of each of those pixels matter the same way the quality of each wave that gets outputted matter. Our ears don't suddenly stop hearing at a certain frequency; we are just less sensitive to sounds as we get further up or down the accepted frequency range. Lower res digital is simply a lossy optimization that is marketed as an imperceptible change, banking on our reduced ability to decipher waveform detail at higher frequencies and/or lower intensities.

Round Earther and Flat Earther Cold Sufferers...

Round earther: I will eat spam because my body can't tell the difference

Flat earther: I will splurge on caviar because it makes my mouth water

[johnt]

@Stagea

 

That must have taken you some time to put together. It's great information. Thank you for sharing your thoughts and knowledge.

 

So you are saying that higher bit depth and sampling rates, or even higher quality methods of converting analog to digital, are not just beneficial to chase the ends of the hearing spectrum, but they also help improve the quality of waves within the limits of the human ear.

 

What about some of us who can't stand Spam, but we don't really like caviar either? I've heard you mention the round vs flat argument before. I'm still not sure I completely understand it. It sounds like creation vs evolution, but not necessarily related to religion. 

[Stagea]

@Stagea

 

That must have taken you some time to put together. It's great information. Thank you for sharing your thoughts and knowledge.

 

So you are saying that higher bit depth and sampling rates, or even higher quality methods of converting analog to digital, are not just beneficial to chase the ends of the hearing spectrum, but they also help improve the quality of waves within the limits of the human ear.

 

What about some of us who can't stand Spam, but we don't really like caviar either? I've heard you mention the round vs flat argument before. I'm still not sure I completely understand it. It sounds like creation vs evolution, but not necessarily related to religion. 

 

You're very welcome, JohnT.

 

When used correctly, higher bit depths and sampling rates is likely to result in an output that is closer to the input (whether within or outside of the audible range). This is the same way uncompressed UHD will render more detail than FHD. It is simply a matter of precision (granted that the devices used provide consistent performance). It is still a debate whether a lot of it is perceptible, but I'd rather have the detail in there until it is conclusively proven that it does not matter (which is the stand of those that prefer standard res audio). 

 

A perfect standard res digital system (that doesn't exist) which uses the correct decoding theorem (that normally does not happen) will result in a clean sine wave when fed 20kHz. Though real-world output is far from it, let's assume that it is. Despite this "perfection," minute changes in amplitude or frequency will not be captured by just 2 sample points per wave (oftentimes the source form may not be even close to a clean wave). This makes the system miss information within the target bandwidth of 20Hz-20kHz. Its only saving grace is the widely trumpeted premise that the human ear's mechanism cannot respond fast enough to decipher that detail (most people will accept this as fact, unless possibly scientifically proven otherwise). This premise is only a mathematical assumption based on the current medical understanding of how the ear works, and has not been measured. The medical understanding is derived from relating how we hear and how other animals respond to sound, and comparing how our ears are versus theirs.

 

What we do know is that it has been validated that most human ears can decipher upto around 140dB of dynamic range (as good in this realm as the best recording microphones). This is why the Fletcher-Munson curve spans 140dB. This much dynamic range cannot be contained in a 16-bit recording without compression and subsequent expansion (this is how 16-bit Dolby Digital can get 105dB of dynamic range, versus 96dB for plain old 16-bit PCM). This is what pushed 24-bit 48kHz to become the most common Bluray Audio resolution (since it covers the entire audible bandwidth and dynamic range at a convenient data rate), and the most common AVR DSP internal processing resolution. 

 

Round Earther = Follows conventional measurement and judgement. Most people think this way as they believe that they are supported by science. In audio, they typically spend the most on speakers as that is conventionally regarded as the greatest source of distortion (and highly disregard the upstream components because they believe that these produce no or minimal audible benefit as long as there is sufficient gain and definition). Spec geeks tend to fall under this umbrella. Round Earthers generally look for the "weakest link" in a system (say room acoustics or speaker performance).

 

Flat Earther = It's a derogatory term given by the round earthers to the "less scientific" audiophile. It's supposed to mean that they are adhering to a belief that is proven wrong. Flat Earthers eventually accepted the term as a differentiator, as they believe that current measurement systems do not correctly quantify component performance. They often look after subjective things based on experience with a system, instead of measuring performance or looking at spec sheets. These people tend to readily spend more on front end equipment and amplification, together with interconnects (like those exotic cables) and accessories (such as power filtration or even double-conversion). Their general belief is that your component is only as good as its signal source... so they focus on the source, following down the signal chain to the speakers.

 

Though I generally consider my approach closer to Round Earth, I accept both mindsets. All of Round Earth's newer measurements stemmed to explain experiences that Flat Earth used to just describe (what used to be left uncaptured by the tools of the time). It happened over a long time, as measurement methods improved. A lot of the numbers now like step response were just mathematical assumptions based on subjective descriptions until the tools with enough resolution to measure this existed to prove its effect (high speed digitizers are a fairly recent phenomenon). History had vindicated Flat Earth. 

 

Are there other things to scientifically "discover" and measure in audio? Time will tell, but I am not siding with either camp. I'd wait it out and do what I can with what I have (or can have).

 

I believe there is much to be understood and improved in the current state of audio recording and playback. Until we get to the point that we can recreate every type of live sound convincingly, there is work to be done. We are getting there in tech capability (I have heard uncompressed recordings of live performances that are eerily close to it with the right equipment), but far from it from an affordability/feasibility standpoint.

[johnt]

Uh oh... I think something is waiting at home for me 

 

[Stagea]

Uh oh... I think something is waiting at home for me 

 

 

 

 

 

Congrats man! Enjoy the weekend.