[nick_charles]

.....

 

[panda]

i could care less of the technical differences, vinyl simply sounds better when i listen. on a purity standpoint, wouldn't vinyl win out because it never has to be converted? even digital recordings need to be converted back to analog for amplification.

 

[Publius]

On the basis of "purity" like that, digital beats all analog technologies handily. After all, the word "analog" itself refers to a transduced analogue of the original signal - not the original signal itself.

For vinyl specifically, how exactly is "convert from analog to digital, then convert digital to analog" any less "pure" than "equalize signal, convert from electrical signal to groove, read groove with stylus, convert back to electrical signal, de-equalize signal"? I'd say CDs are far more pure than vinyl under that sort of thinking.

 

[Publius]

Quote:

Originally Posted by Golden Ears /img/forum/go_quote.gif Actually this is not true. I designed a record cleaning machine that has played albums over 250 times without any additional noise added to the record. The lead in groove is silent. I suppose I could make a few more of the machines.

I will say I agree with you in that a large number of plays like that may not cause increased noise on a well-cleaned and well-maintained system. In fact some people think that repeated plays can help improve the noise levels by scooping dirt out of the way and polishing the groove walls.

 

[chesebert]

Nyquist theory says analog signal that has been sampled can be perfectly reconstructed from the samples if the sampling rate exceeds 2B samples per second, where B is the highest frequency in the original signal.

That graph, while not per se wrong, is misleading to say the least.

Honestly I don't remember anything from DSP anymore. but if you build the perfect anti alias filter and a ridiculous order butterworth
filter you can get perfect 22khz wave out it (that much I know....since it worked in Matlab

 

[jp11801]

First both vinyl and digital can sound great, personally I think it much harder to get good sounding digital as many dacs are designed with pure specs in mind and not tuned to ear. Just because something measures well does not mean it sounds good. Good measurements are just a start.
CDs resolution is pretty darn good and a well mastered (somewhat or a rarity) cd should give a well mastered chunk of vinyl a run for the money. I have a personal preference for vinyl but I grew up in the era of the LP.

Early CDs were not mastered using EQ designed for LPs. This demonstrates a lack of understanding of how music is produced. Most early CDs were flat transfers from the master tape or early version production master tape. This often whey they were sought out. Where they sucked they sucked because the master did not sound good and later versions were goosed with EQ, compression or other post production goodies. If you were to listen to a recording with the RIAA curve applied with no RIAA eq to reverse it you would hear alot of treble and little bass. If the RIAA curve were not applied to LPs the groove required to reproduce bass content would be too wide for the needle. Also the needle would jump out of the groove if LPs are not done right.

Digital is a stepped reproduction much in the same way analog film for movies is a stepped approximation of motion. The greater number of images or in music the higher the sample rate the greater number of snapshots. Given a high enough number of samples and it will be indistinguishable from analog tape providing care was taken with the recording.

To my ears well done LPs sound better have a better realized dynamic range (for rock and jazz). This is not because the LP is superior but because the person who mastered the LP set the level low enough to reproduce more of the dynamic range on the tape. Most cds have s higher volume starting point and go into digital clipping. The best explanation of this for a modern record is to get a some computer recording software and open a cd file in it what you will see for rock, pop and some jazz is the brick wall. Fully saturated waves with little or no low points. I did this with the past Red Hot Chili Peppers CD and Dana California is horribly brick walled but the RTI mastered LP is great and had plenty of dynamic range when I did my needledrop. The digitized LP sounds leagues better than the cd for the simple reason there is more dynamic range, less compression is used and the volume is not so loud that it clips. You can stand to listen to the LP much louder than the CD because of this fact. Anyone that wants to examine this for themselves just needs to download a piece of recording software and you can check out your cds wave files all day long.

Someone earlier stated the because Stac Ricker (mofi as well as others) likes cds then cds are better, well to offer a counterbalance Steve Hoffman stated that to his ear LPs are closer to the master tape sound than cds. It sounds like a preference thing more than anything else.

 

[chesebert]

Quote:

Originally Posted by jp11801 /img/forum/go_quote.gif Digital is a stepped reproduction much in the same way analog film for movies is a stepped approximation of motion. The greater number of images or in music the higher the sample rate the greater number of snapshots. Given a high enough number of samples and it will be indistinguishable from analog tape providing care was taken with the recording.

Helpful but your analogy is inherently wrong. It's possible to produce perfect analog signal from mere samples whereas it's impossible to do the same with films. One is perceptive, the other is substantive.

 

[jp11801]

when Nyquist's theory becomes Nyquist's law LMK until then it's theory

 

[chesebert]

Quote:

Originally Posted by jp11801 /img/forum/go_quote.gif when Nyquist's theory becomes Nyquist's law LMK until then it's theory

I think you are being a little disingenuous with your argument here. It proves true both in simulation and in the lab. It's theory because the theorem is based on a set of assumptions, one being the bandwidth of the signal is finite (brick), which is not quite true in the real life.

I am not a DSP engineer... my knowledge only extends as far as building filters in Matlab (and diminishing as I am getting older).. perhaps some current practitioner with a little more DSP background can chime in.

 

[jp11801]

Quote:

Originally Posted by chesebert /img/forum/go_quote.gif I think you are being a little disingenuous with your argument here. It proves true both in simulation and in the lab. It's theory because the theorem is based on a set of assumptions, one being the bandwidth of the signal is finite (brick), which is not quite true in the real life. I am not a DSP engineer... my knowledge only extends as far as building filters in Matlab (and diminishing as I am getting older).. perhaps some current practitioner with a little more DSP background can chime in.

you are correct in that I was having fun with you, but both analog and digital are but approximations of the event that is captured on tape or memory.
People have preference for either and I am not here to try and take anything away from folks in either camp just trying to state neither is perfect.

 

[jp11801]

[/IMG]

RHCP Dana Cal screen shot from Peak LE 3/4 of the way through the track brick walled like crazy

much better mastering of RHCP Suck My Kiss from Blood Sugar Sex Magic

I don't have any theorems to recite but when digital clips it causes what I will deem the cringe effect, it makes me cringe and reach for the volume. When analog tapes clip particularly with tube gear it is a much softer clip that while still problematic not as cringe worthy

 

[joe_cool]

On the subject of dynamic range and compression, I submit this comparison of "Ride Across The River" by Dire Straits from "Brothers In Arms".

The top graphic is of a "remastered" version (ASIN: B0013AZTR0) and the bottom graphic is of the 1985 Warner-Polygram version (UPC 07599252642).

As a demonstration, the top graphic has been normalized (reduced in volume by -12db) to match the lower volume of the lower graphic.

 

[jp11801]

the top one is truncated all over the place

 

[joe_cool]

Yes the top ~12db is clipped. Of course it sounds louder.

 

[jp11801]

Quote:

Originally Posted by joe_cool /img/forum/go_quote.gif Yes the top ~12db is clipped. Of course it sounds louder.

sadly mine is the crap version