Time smearing is the enemy of quality audio sound. We are very sensitive to minute differences in arrival times of sound as it reaches one ear and then the other. We are talking about microsecond. The brain uses these differences to localize the source of the sound. If we want a realistic stereo image of the musical performance, pinpointing each player in space, we need to eliminate as much time smear as possible.
The great Raymond Cooke of KEF helped pioneer impulse testing of loudspeakers. KEF drivers were known for the quality of their individual drivers. They were used in BBC monitor speakers. These drivers were quite smooth from a frequency standpoint, but they had good impulse response as well. The impulse test is like plucking the string of a guitar and then allowing the natural decay of the vibration to stop the vibration and thus end the sound.
In a speaker driver, however, we want the vibration to stop as quickly as possible. If it lingers then it smears the sound. The tiny differences in arrival time of the sound to the ears is lost, thus confusing the brain and ultimately the stereo image. BBC monitors were known for their stereo imaging, other speaker manufacturers followed suit with their own impulse testing.
Time smearing in our audio systems has many causes. One is speaker placement. Room reverberations can smear the sound. Careful placement of the speakers will help reduce inter-reaction with wall boundaries. In live recording of acoustical music we want to hear the hall ambiance. This information will be lost in the ambiance of our own listening rooms. And again, the separation of musical instruments will also be diminished with poor speaker placement.
In the early days of solid state amplifiers time smearing was not a concern, at least for some of the manufacturers. I owned a high-powered solid state amplifier which had great specs and was highly reviewed. A great deal of overall negative feedback was used in the design to lower the distortion ratings. But it sounded terrible on my Magneplanar Tympany 1’s. Then I heard one of David Berning’s prototype amps driving the Tympany 1’s. It was a night and day difference. With the Berning the music came alive and so did the stereo image. David’s amplifier used no negative feedback at all. Apparently this feedback messed up the timing of the reproduction. Others can explain it. The measurements sometimes have little to do with the perceived sound. The ears cannot be fooled.
Time smearing can be a problem with vinyl recordings as well. That is why people are will to pay large sums of money for rock steady turntables, vibration free toneams, and delicate, lightweight phono cartridges that are compliant, but properly damped. I loved the sound of my Decca Mk V but is sounded better when I put it in a damped tonearm. The Decca, itself, had practically no damping.
Now what about digital music? Timing is everything. The first CD’s and CD players had great specs. But the sound was just not right. It did not sound alive. Were was the hall ambiance? Timing. Jitter. Theoretically, a 44.1Khz was a high enough sampling rate to fool the ears into thinking that digital sound was actually analogue. But is was not, at least not in the beginning. There was little hall ambiance. Stereo imaging was not good. The sound seemed lifeless. What could go wrong? Bits are bits. We are just dealing with zero’s and one’s. Yes, but they had better be in the proper order. Time smearing occurs when they get out of sync so to speak.
High sampling rates, apparently, make it easier to keep the bits in line. The skeptics will say ask why should the sampling rate go out to 96 kHz or even 192 kHz. We can only hear frequencies to 20 kHz at best. Yes, but we are not talking about frequency response. We are talking about filtering out unwanted sound to that we can hear the sound that we want reproduced. It is easier to do the filtering at higher sampling rates. Not only do we want to eliminate unwanted artifacts from the analogue to digital conversion, we also want the timing of the bits to be correct. Low noise and low jitter. See the chart below:
We are looking at tiny differences, measured in microseconds. The ears can hear this difference. An impulse signal is sent through the system. The 96 kHz sampling rate rings longer and is less controlled than the 192 kHz sampling rate. This is a difference in the performance of the filtering.
The best performance, however, comes via MQA (Master Quality Authenticate) invented by John Robert (Bob) Stuart of Meridian Audio. It passes the input signal through but then it stops it short. Much of the unwanted ringing is removed which obscures the timing of the digital signal. In digital timing is everything.
More can be said about MQA. For now, suffice it to say that it has shown great promise in the converting an analogue through a digital time domain and then back to analogue. Ours ears want to hear analogue with as little time smearing as possible. Could this be the future of digital audio? We shall see. MQA music files are now being steamed over Tidal.
For me, the monthly streaming cost of $20 is too high. I would like to be able to download the files and keep them on my hard-drive. Fortunately there are some sites to do so: 2L Music Store and HighResAudio being two. Now if only I could buy a reasonably priced compatible DAC. Wait, what about the Dragonfly Red? And Audirvana Plus is now compatible with MQA. Somebody stop me.