Grateful Dead

you too. So sorry.

Disc number two from the Omni Show (4/1/1990) will not play in my car. The car radio says "disc error" when I called Deadnet they told me they would not be able to replace the disc because it was over 30 days old. Can you help me? Who did you talk to when you called customer service? I am not very happy about this. Thanks!!

send me your order # and the details and I'll see what the Dr. can do.

I must say I am very impressed with the sound quality and strong performances of all of these shows. I have been listening off and on for the past couple of months. However when I got to disc two of the first Omni Show (April 1, 1990) I discovered the disc was defected and would not play. When I called DeadNet they told me there was nothing they could do for me because the purchase was over 30 days old. Well they did tell me to repurchase the box set and return it with the defective disc. I do not want to go through all of that. I payed close to $250.00 for this and Deadnet is not willing to replace a broken disc. Any advice?

Audio inspector is the name of the software I was using. It makes some quick general assessments of the file and then starts to deeply analyze from the beginning. It takes a couple of minutes just to get through 15 seconds of a track, which is all I let it do as I didn't have much time. So keep in mind that I think those numbers are for the first 15 seconds. However, I coukd see and zoom into the entire file. It was immediately clear that the HD file was significantly narrower from top to bottom, indicating no gain (I don't know the technical terms for most of this, so I'm assuming yours is correct) or else much less gain had been applied to that file. Since everything I read indicates that the primary purpose for applying dynamic compression is to make room for gain, I believe that little or no dynamic range compression was used on the HD file (at least compared to the 16-bit file). The CD file on the other hand appears to use almost all the available amplitude range from top to bottom. Keep in mind that the -10db and -15dn peak numbers (and the other numbers as well) I referred to may be for just the first 15 seconds.

Hi JMT2010 - I posted a few links that go into a lot of detail about the technical aspects of digital audio - you can find them below. You're close, but not quite there in what you described. for instance, at the very end, you refer "the human ear does not pick up ..... it just hears a continuum". The issue here is that it doesn't have to pick or not pick up the individual samples. The digital to analog converter (DAC) takes the stored digital information and converts it back to an analog wave. The Nyquist theorm, on which the very idea of digital audio is based, states that as long as the frequency of sampling is as least twice as high as the highest frequency of sound being reproduced, then the ORIGINAL analog sound wave, of any complexity, can be reproduced EXACTLY. That's why the "stair step" concept that hi res websites like to display is a deception. When you look at a graph of a waveform stored digitally, yes if you zoom way in you can see "stair step" looking (jagged) edges to the waveform. It's a deception, because the DAC recreates from this the original sound wave EXACTLY - as long as the frequencies are below half the sampling rate. Another thing that was not quite right was your interpretation of bit-depth. It's even simpler than your first sentence. What is actually contained in each "sample" is one amplitude measurement, just a number between 0 and 65,536 for 16-bit and between 0 and 16,777,216 for 24-bit, representing the amplitude of the wave at that moment. Forget about the noise floor for a moment. The ONLY thing stored in each sample is a number representing an instantaneous measurement of the amplitude of the sound wave at that moment. Quantization error is the difference between the ACTUAL amplitude of the sound wave at that point, and the measured amplitude using a discrete number of only 65,536 or 16,777,216 possible values. Dithering is the process which mathematically converts those errors to white noise, and noise shaping actually moves that noise to largely inaudible ranges of the sound frequency spectrum. Ultimately, it is the level of noise in a digital file that determines the "noise floor" of the file. This is the exact equivalent of the signal to noise ratio (SNR) of an analog recording (LP or analog tape). Keep in mind that the SNR of even a 16-bit recording is many times better than the SNR of LP OR analog tape. Most people don't understand that, either. So, taking your Pink Floyd "Time" example, a 16-bit recording can capture the quietest elements of the clocks ticking. Of course, THAT is a recording that was NOT originally recorded digitally - it was originally recorded to analog tape. So the SNR can NEVER be better than on the original analog tape - there is a minimum noise level already inherent in the recording to begin with. Modern recordings are recorded to 24/192 digital files, and then if converted to CD (or 16-bit downloads) they are converted to 16-bit using noise-shaped dithering. Done properly, the resulting 16-bit files have a slightly lower signal to noise ratio, however it is already below the level of human perception. The noise floor of your listening environment is ALWAYS (unless you're in outer space or something) higher than the noise floor of a properly dithered 16-bit recording. Noise you don't usually notice, the hum of the refrigerator, your breathing and heartbeat, the water heater, etc. - even the quietest of most rooms still has a noise floor that is above the noise floor of a 16-bit recording let alone a 24-bit one. This is nit-picking a bit, isn't it???? The other thing you referenced is HOW does a stream of amplitude measurements capture actual music. Take out a piece of paper. Let's say you're sampling at 10 times per second instead of 44,100 times per second. So, 1/10th of a second you capture an amplitude measurement (the height of a sound wave). On the piece of paper draw a dot at that height. It might be easier if you draw a rectangle with that height (just of like the rectangles under a curve in pictures of integration from a calculus textbook). When you connect the dots, you can see the sound wave shape. The more dots, the more exact the representation of the wave. This is where the Nyquist theorem comes in. Higher frequency sounds are going up and down across the x-axis in narrower bands than lower frequency sounds which take more time (stretch out farther along the x-axis) before coming back across the x-axis). The theorem states that as long as the sampling is rate is at least twice the highest frequency, the DAC can mathematically recreate the EXACT analog sound wave. So, 44,100 samples per second is enough to EXACTLY recreate any frequencies below 22,050Hz. This is above the range of hearing for human adults. So, some people who don't understand the technical aspects will pay more for a 24/192 file than a 24/96 file. Keep in mind what the actual difference is. A 24/192 file is taking 192,000 samples per second, and a 24/96 file is taking 96,000 samples per second. The Nyquist theorem states that the 192k/s file can PERFECTLY reproduce any frequencies below 96kHz. The Nyquist theorem states that the 96k/s file can reproduce any frequencies below 48kHz. Um, most adults can't even hear much beyond 16-18khz let alone 20khz. The ONLY difference between the fidelity of the 24/96 and 24/192 is that the 24/192 can encode frequencies from 48kHz to 96kHz and the 24/96 can't. Those frequencies are all and entirely WAY WAY WAY beyond the human hearing apparatus. But, go through some of these threads and watch some people saying things like, "are we paying for 24/96 or are we actually getting the full 24/192?" The question is nonsensical. NO ONE can hear ANYTHING in the 48-96khz range AT ALL. Not only that - none of the microphones used to record the music capture anything in those frequencies at all AND on the off-chance they did, they're filtered out for technical reasons. Just WHAT do people think they're missing in the 96 vs the 192 file? It shows that they just don't understand what they're spending their $$$ on. They are assuming that 192 has to be better than 96, and/or that if its more expensive (and larger) it must be better. Anyone who understands sound at all knows that a audio with or without frequencies between 48khz and 96khz is going to be identical unless you're a hummingbird or something. It's like thinking that a picture that has light going up to the x-ray range encoded in it is going to look better than a picture that only includes light in the spectrum our eyes actually have the hardware to respond to. And then, they will actually post about how much more depth there is to the music, how much more full and somehow realistic the experience is. It's clearly entirely in the realm of psychological expectations. Actually, properly dithered, a 16/44.1 digital file made from the EXACT SAME SOURCE as the 24/192 digital file is INDISTINGUISHABLE from each other by the human ear. ALL scientific studies done in controlled environments confirm this. You will NEVER convince some people of this, however. The idea that more bits and more samples must be better seems to make to much sense to most people, and marketing has done it's job. Lastly, as you can see in one of my last posts, I compared the 16-bit CD files to the hi res files that are being offered for Wake Up To Find Out. I compared them using Audio Inspector. That comparison confirmed that these two digital files are NOT from the same source. This has nothing to do with the inherent ability of a 16/44.1 file to be as perfect to human ears as a 24/192 file. What is being done is common in the practice of making CD's. They compressed the dynamic range (the range of softest to loudest sounds) so that they could then increase the amplitude across the entire range, making the CD louder at any given volume setting than it would have been. This was either not done to the 24/192 file, or not to the same extent, because the 24/192 file is not as loud, the amplitude of the sound waves at any given point is lower than on the 16/44.1 file. This was done INTENTIONALLY (I'd rather they didn't). It is probably done because people "expect" their CD's be play at a certain volume - they think something is wrong if they put another CD on, and it's way louder without turning the volume up - they ask, "why is this one so damn low!". So, they're dealing with consumer expectations. It has nothing to do with 16/44.1 versus 24/96 or 24/192.

From what I have read, the higher the bit depth, say 16 bit vs. 24 bit, the more decibels of signal is possible above a noise threshold. I get that. It is a metric of quietest sounds to loudest possible to be reproduced in fidelity perhaps. An analogy for that might be Pink Floyd's song 'Time' where you hear the clocks ticking very quietly in the beginning and then have the loudness of the alarms going off the next moment after. The loudness change is dramatic. OK, I am having a difficult time drawing analogies to the music we listen to on CD versus say cassettes or vinyl. The waveform for analog music is continuous if displayed on a graph. Music in the forms of ones and zeroes getting converted to analog is what escapes me. How doe the reproduction of the sound of a guitar and drums get unscrambled from the digital ones and zeroes? I get that the sampling rate captures 44,100 pieces of information per a second (44.1kHz rate) of a music passage, but what is the information stored in that 1/44,100th of second? Playback is at 44.1 kHz per a second I assume ( on a CD's WAV file format). The human ear doe not pick up the 1/44,100ths of a second "quantized" sound pulses. It just hears a continuum.

If anyone is a big collector like me and bought a few box sets/poster combinations.. do yourself a favor. Open the poster container and make sure what you ordered is in there. There's 1 poster left available to buy onlne... you can't add more than 1 to the cart. I decided to open mine tonight. Suffice to say, the 4" and 3" containers I have, which should have multiple posters, only had 1 each. Nervous, scared and terrified doesn't begin to describe the butterflies in the stomach or stomach acid reflux in my throat... since they've been sitting in my closet unopened and uninspected since July. Word to the wise... check to make sure you got what you ordered. :-( Called customer service. Suffice to say, this has to go higher for any hope of resolution. Not how I wanted to start Christmas... check what you ordered... at least I checked now and not 5 years from now. But still... my faith is w/ Dr. Rhino or someone, to help.

Right, "make-up gain" is a post-compression volume increase that presumably brings the peak up to 0 dB (or wherever the engineer chooses). It's really odd that they chose -15 dB and -10 dB for the HD and CD files, respectively. That headroom (relatively huge) serves no purpose. So, how did you know the CD files were more dynamically compressed than the HD files?

That's where my bet is too. Sadly. "Why do I have to turn *this* CD up louder than my other discs?" is a question that often (not always) can be answered "because it was better engineered."