Do cd's store a more exact copy of sound than LP's

Well, digital isn't so bad as before. And mass market LPs were much worse right when cds first came out than thirty years berore or twenty since. And digital should get much better and cheaper. But not today. Nothing is accurate. Every thing matters. Digital masters are better than before so it makes sense to stay with them in a good digital format. But 'more accurate' is a case by case call. It almost is a senseless question because the answer is so complex. Even the accuracy of a digital tape says nothing about what gets on the disc or what happens during playback. And so it was with LP. The essence of digital error correction is smearing by guessing. But in the last ten years we have gotten more out of LPs then we ever dreamed was in them lovely little walls. LP sales are up the last fourteen and CD is down. And it was for nine years before Napster. My daughter has hundreds of pieces on her computer, but just try and take away her records. Most of the best at the end of the day is still AAA. Even when the day comes that SACD or DAD surpasses analog, which might first come through surround sound, I will be long dead before the difference will have been good enough to have made me melt my records.

Aroc, the only medium that stores sound is our mind.
Asa buddy, where are you? This is the second time today that you've come to mind.

16 bits do not provide enough data space for complex music for orchestra or big band Jazz. To put all music into CD format simply compress data, and therefore lost some information for complex music. Therefore, playing CD seems like there are only 5 violins in orchestra, but you hear maybe 10 from good tape or LP, for example. Upsampling only make these 5 violin sounds more refined but it is still 5 violin (no way you can add data back! Upsampling can't add more violins but make existing ones clear). SACD with moer bits will help. For LP, maybe those 10 violins do not have very precice tone comapred to CD because of warp of LP or bad print, but the info was there due to the larger data base, namely dynamic range in audio(no way you can add violins by elecronic equipments first either!).
So, inpep..., trust your ears and there is solid science behind it too. Do you really think a machine know how to add music to Beethoven?
Your ears already told you the truth, don't reject it because being lazy to study science.

Pick some music with 100 people performing from low to high volume. Let yours ears tell you what is going on.

One violin or a hundred violins, there is only one signal (voltage variation over time) per channel. The fidelity with which this analog signal can be represented by a series of numbers (voltages at times) which is the digital method, is precisely known from the resolution (number of voltage values available, bits) and the frequency at which the value is updated. Fidelity of analog recording is much harder to quantify, because the errors are also analog and vary greatly. In addition to distortion of the signal, the analog media introduce various errors that are unrelated to the music, and overlay it. Turntable rumble, LP surface noise, tape hiss, etc. Theoretically the music exists in the analog signal down to infinitely low levels, and with infinitely fine resolution, but this cannot be heard because it is masked, at some level, by the noise. Analog technology strives to reduce the noise level so that masking of the music signal occurs at lower volume level. Digital technology is intended to provide resolution equal to or better than what can be obtained in the real world by analog equipment. 16 bit 44.1 KHz is OK for the typical mass market audio systems, but is really cutting it close, when compared against the best analog equipment. 24-bit 96KHz should leave analog in the dust, but this may not always be the case because mastering of the recording (disc or LP) plays such an important part in overall audio quality.

The idea that an analog recording includes very low level information that is masked by noise is supported by work that has been done to recover historic recordings made almost a hundred years ago. Of course, these are of terrible audio quality when listened to "straight". However, these recordings can be digitized and processed in various ways to separate the music signal from all the noise, and the results are quite astonishing. High frequency response is particularly surprising. The HF stuff actually is on those ancient recordings, but so greatly attenuated that you wouldn't think that it was there at all until the signal is reprocessed. All of this is working at a noise level that is much higher than that on the typical LP, but it does illustrate the point that music information is present in analog recordings, even if unheard, below the noise threshold.

Bluefin, I am not too lazy to study science as shown by Eldartford's response, but I certainly won't study what you call science! My lord, recording 100 violins needs a larger data base than 5! Please.