Originally Posted by QueueCumber
Now to see if I can manage to hear the jitter issues... Anyone have new/additional input on what I can do to try and hear it?
. Nick actually provided an indirect clue: soundstage.
Let's examine what jitter does to sound. If you see the graph I posted, it creates new fequencies that did not exit. Those frequencies therefore, raise the noise floor of the system. What happens when you do that? You destroy anything that exists at those levels or lower. Signals that are much louder are not affected.
So what you want to listen for, is quiet, and faint sounds, not the loud signals. This is an important lesson in digital versus analog. Digital systems are fantastic at playing maximum value because then the signal is well above its distortion product (created due to jitter in this case). Analog systems hate this as the higher the volume, the more distortion they create. Being analog instruments ourselves
, we tend to test digital systems as if they were analog.
My test for jitter (or frankly, finding any other minute distortion) involves finding a sharp transient that decays into nothing gradually due to recording venue reflections. A single guitar note or symbol crash decaying over 2-5 seconds is perfect. Then using a headphone, I listen to that decay at elevated volumes (careful you don't go deaf when the rest of the music plays
). I have a set up with a headphone amp with dual input which makes it easy to do A/B tests (blindly if needed). But even without, you should be able to still peform the test reasonably well.
Now, listen to length of decay. A less ideal system will decay nicely as first and then there is a sharp end to the signal. The better system will decay smoothly into nothing. This decay becomes key to soundstage which was mentioned above. Without those reverbrations the soundstange compresses. In other words, soundstage is second order distortion.
Second effect to look for is accentuating high frequency. Recall that jitter creates new sidebands one of which is at higher frequency than the source. Therefore, it contributes to making the sound "brighter" (what audiophiles call "harsher").
Most importantly, you need to do this in A/B mode where the switchover hopefully is instantaneous. Anything over 4 seconds invalides the testing as your brain will not remember things that long. In my own testing, even 1 second switch over is too long.
To help with the test, I use the A-B mode of the source to keep repeating the segment I am interested in so that when I switch back and forth I can quickly hear the segment again.
Note that it will take you a while to hear jitter. It is a very illusive artifact as it comes and goes. And what it does is dynamic as it depends on what you are playing and masking effect.
My experience though is that once you "learn" to hear it, then it becomes much easier to find it, much like compression artifacts.