Just resetting the comment section for Summer Tour.
This entry was posted on Monday, June 27th, 2016 at 3:33 pm and is filed under Uncategorized.
You can follow any responses to this entry through the RSS 2.0 feed.
Both comments and pings are currently closed.
I’ll admit it. I’ve slept on Hooteroll! Hat tip to dusty for edifying me. Have been spinning ad nausea today. Uncle Martin’s South Side Strut. My kinda groove. Might be some of the sickest music Jer ever made. Feels like it’d be right at home late nt @ Maple Leaf during JF. Elevating.
so basics- we are sampling the sound X times per second. So at one instance, we have a measurement of the amplitude of the sound. The bit depth determines the amount of information captured for each measurement of the amplitude.
That information increases the gradation of the amplitudes captured… just like with digital video. Simple digital video might just have 16 colors… but modern digital video has more data and allows for millions of gradation of color… 50 shades of grey, if you will.
So, digital instruments are going to measure inputs (sounds waves, light waves) and then round that measurement to the nearest unit the data file will hold. All fifty shades of grey are just ‘grey’ with 8bit digital video.
With more bit depth, there is more gradation to get more accurate measurements of the amplitude at each sample. Now, that gradation gets calibrated by the recorder. You could spread it across a huge frequency range in which case a small percentage of the range would capture audible data. That is the headroom of which you speak. But if the recording artist sets the range of frequency near the audible range, then we get more precise measurements of amplitudes.
Now, this begs the question of whether that resolution does anything. At some point, like 1,000 bit, the extra gradation just doesn’t make much of a difference because the loss to rounding is so small.
So yes, the nyquist sampling theorem sets the floor for the data needed to recreate a wave form. If you use that floor, you will get the wave form but we know from early computer audio that it doesn’t sound good to our ears. The more data, that better it sounds. 24bit sounds better than 16bit to many. I doubt 32bit would be noticeable.
Mrs. Ren Here! Stepping out of the woodwork to join all you amazing people in the amazing rides Phish provides! Hopefully Miner puts this comment through, didn’t work last time…please allow this hopeful noob to join the circus!