or Connect
Head-Fi.org › Forums › Misc.-Category Forums › Music › vinyl rip vs cd
New Posts  All Forums:Forum Nav:

vinyl rip vs cd - Page 14

post #196 of 334
Quote:
Originally Posted by bigshot View Post

Mechanical noise in turntables is miniscule compared to the noise floor of the records themselves. Unless of course if the turntable is improperly grounded.

We are getting to the core of the problem - slowly, but we do.

 

First, I am fanatic about grounding. I do not tolerate any hum in the system that can be distinctly heard as such and will not cease until I track the source of it and unroot it for good. It is the most detrimental in turntables, as "good enough", not directly audible during record playback, is not nearly enough for best performance. Once one hears it removed, music shines and returning to "good enough" hum performance is painful indeed. Not always, but at times I use batteries to supply the really low noise preamps. And I make sure there are no ground loops possible.

 

I wrote mechanical noise in operation . That is big difference. Even if you use Thorens Rumpelmesskoppler , which is the best rumble measuring probe known to me, generally producing rumble figure(s) 10 to 20 dB better than even the best of test records,  it DOES NOT measure the noise of the turntable/arm/cartridge IN OPERATION. That is to say with recorded signal being played - and bouncing and echoing from the highly resonant stucture called Thorens turntable.

 

Accelerations of the stylus in the groove can reach 2000 G. That is possible in the treble, roughly 5 to 10 kHz, recording velocities around 100 cm/second. Check any brochure for Shure V15 series, they did provide trackability chart where theorethical and ACTUAL recorded velocities that exceeded theorethical were given. If we take average top quality cartridge available today, with an effective stylus tip mass of 0.25 mg , it is easy to calculate the force acting on the groove walls>record>mat>platter>main bearing>subchassis >etc>etc. 

 

F = m x a, therefore 0.25 mg x 2000 G = roughly 5 or so kilos acting on the groove walls at the high frequencies. This energy has to be dissiapated somewhere, as fast as possible and preferably not be capable of returning to the point of origin, in our case the stylus - or else the stylus will be reading its own echoes for the duration of the entire disc side, masking low level recoded signal. That was meant as mechanical noise in operation. It is here that Thorens in unmodified condition is very poor indeed.

 

From the point of mere rumble, wow & flutter, it is a very good turntable.

post #197 of 334
Quote:
Originally Posted by PraetorXyn View Post
 

The Accurate Rip database is maintained by the creator of dBpoweramp, but he makes it openly available to other programs.

EAC is harder to set up than dBpoweramp, so that could be why you're getting differing results. The reason item 2 lists the extension as wav is because EAC is adding the item to the log before it uses FLAC to encode the file.

 

I've performed the following actions and tests in regards to these rippers:

1. Find a CD with lots of gaps. In dBpoweramp, add the Gap column to the display.

2. Rip the CD with dBpoweramp.

3. Open foobar2000 and select the album you ripped. Use the foo_texttools component to copy the track titles.

4. Open EAC and go to Database > Get CD Information From > Clipboard. This will make the track titles match.

5. Press F4 to detect gaps. Press F3 to analyze gaps for silence.

6. Rip the disc with EAC.

7. Generate CUE sheet (Multiple WAV Files With Gaps (Noncompliant)) with EAC

 

If you use dBpoweramp Batch Converter to convert the rips to WAV (it will make copies, not delete the original), you can compare the WAV files one at a time in EAC or all at once in foobar2000 by creating a playlist, then add the dBpoweramp wav files first, and the EAC wav files second. The audio samples should be a perfect match.

 

I then used EAC to burn the CD from the cuesheet, and then ripped the burnt CD with dBpoweramp. I compared the WAV files once again, and they were a perfect match.

 

So all I use EAC for is the generation of CUE sheets.

 

My reasoning:

1. dBpoweramp will rip a CD in ~2 minutes, and EAC can take 15+ minutes to get the exact same result.

2. dBpoweramp has the best metadata support of any program I've worked with.

3. dBpoweramp lets you set the folder structure dynamically based on metadata. If you do this in EAC, the folder structure will be written in your CUE sheet where the track names are.

4. Both programs allow logging, and can automatically generate an m3u playlist. dBpoweramp does it with a DSP effect, and it lets you control thefile name based on metadata.

Thank you for the detailed explanation. I am a slow learner so I will look into the Cue sheet and other parameters you have covered.

 

I just ripped "Mylène Farmer-Monkey Me" by dBp (Flac) and Foobar returns a DR7 for the CD.

Ripped by EAC (Flac), Foobar returns DR5.

Is the difference because of the quality of the rip by the two softwares or is it Foobar playing up?

I would appreciate a simple explanation.

Thanks

post #198 of 334

I'm with you.  I used to have a Samsung CRT television that I could hear when it was on.  I remember doing an internet search about this and found that it not unusual for this particular set, and that frequency that I was hearing (and others could not hear -- some people thought I was crazy) was only about 16khz.

 

Quote:

Originally Posted by brunk View Post
 

I have an odd ability that only a couple people i've come across in my life have the same thing. When a monitor, TV, PC or some other devices is powered on with absolutely no sound or video, speakers disabled etc., i can "hear" the device operating. I have tested this with a friend in the living room and me at the complete opposite end of the house, I could tell him precisely when it was on or off. That's just one of many different times I have tested myself to make sure i wasn't crazy lol. Is there an explanation for this?

post #199 of 334

I have a question for you (or anyone).  If you're listening to speakers or headphones that only respond to 20khz or, say, 38 khz (which I believe is claimed by my AKGs), how are the harmonics above that frequency produced?  Would it matter if vinyl or a 192khz recording has information at 45khz if the speakers don't reproduce it?

 

Quote:

Originally Posted by ploppy666 View Post
 

My mistake about cymbals.

 

But many instruments have harmonics above 20kHz, including cymbals.

post #200 of 334
Quote:
Originally Posted by acs236 View Post
I'm with you.  I used to have a Samsung CRT television that I could hear when it was on.  I remember doing an internet search about this and found that it not unusual for this particular set, and that frequency that I was hearing (and others could not hear -- some people thought I was crazy) was only about 16khz.


+1. The sound of a TV with nothing playing drives me crazy. For some TVs anyway. 

post #201 of 334

For the record, I'd rather read you two going back and forth than read over subjective comments about how certain headphones sound warmer or more natural, etc..

 

Quote:

Originally Posted by bigshot View Post
 

 

I agree with you on this point. I think we all feel that you've wasted more than enough time arguing!

post #202 of 334
Quote:
Originally Posted by acs236 View Post
If you're listening to speakers or headphones that only respond to 20khz or, say, 38 khz (which I believe is claimed by my AKGs), how are the harmonics above that frequency produced?  Would it matter if vinyl or a 192khz recording has information at 45khz if the speakers don't reproduce it?

 

Super audible frequencies in equipment not designed to deal with them can lead to harmonic distortion back down in the audible range. I posted a link to an article about this earlier in the thread. Most recorded music doesn't contain super audible frequencies, so it isn't a problem in most cases.

post #203 of 334
Quote:
Originally Posted by bigshot View Post
 

 

Super audible frequencies in equipment not designed to deal with them can lead to harmonic distortion back down in the audible range. I posted a link to an article about this earlier in the thread. Most recorded music doesn't contain super audible frequencies, so it isn't a problem in most cases.

Yep, it causes non-linearities for the speakers, which come back down into the audible range.

post #204 of 334
Quote:
Originally Posted by acs236 View Post
I have a question for you (or anyone).  If you're listening to speakers or headphones that only respond to 20khz or, say, 38 khz (which I believe is claimed by my AKGs), how are the harmonics above that frequency produced?  Would it matter if vinyl or a 192khz recording has information at 45khz if the speakers don't reproduce it?

 

In the old days speakers were just spec'd from 20 - 20K as stuff above 20K was deemed irrelevant. Most high fidelity speakers will however respond to signals above 20K but at relatively lower amplitude i.e a few db down and getting further down as the frequency rises until the speaker cone can no longer respond at all. We've had speakers capable of some response to 35k or even 50k for a while but our best evidence to date is that this high frequency component is not normally detected and thus makes no impact on the perception of music. The Balinese Gamelan has harmonics to about 50K and trumpets can have harmonics out to 100K.

 

Vinyl may have response above 20K but it is normally relatively little, often just noise and not generally encouraged by those who master vinyl as it is both difficult to cut into the lacquer and for a stylus to track , at high amplitude especially towards the label. Some older LPs may have extended frequency responses such as the 1962 Firebird Suite (see below)

 

 

We do not know what the peaks are but from the text of the article I found this in my guess is that it was sampled to peak at 0db, the guy was trying to make a case about hf content on LPs. So the material above 22k is at least 50 but probably nearer 70db down. Let's be generous and say just 50db down. Any humanly audible content would be very effectively masked out. A few exceptionally lucky young humans can hear to maybe 25k at high levels but detecting a low level harmonic at say 25K in the midst of a much louder fundamental would be a real challenge.

 

Something I just did for a laugh. I generated a 2k signal at -50db and I merged it with a segment of pop music. A 2k signal is in the range of good sensitivity for human ears. normally pretty easy to hear, certainly easily audible to me when unmasked. However buried in a music signal that peaked at close to 0db it was completely undetectable. Very high frequency (above 15K ) is much harder to detect i.e you need an extra 25db just to detect it in the absence of anything else.

 

Then I lowered the level on the music track to peak at -20db, with the music segment at -20db I could easily pick out the 2k tone again (but it is then only 30db down not 50 or even 70). When I tried the same with a 10K tone just lowering to 10K tone to -50db made it wholly inaudible (at normal levels I can still hear 10K) , of course others have rather better hearing than I but you get the picture. There may be content above 20k but it is not normally terribly audible.


Edited by nick_charles - 11/11/13 at 12:31pm
post #205 of 334
Maybe I don't quite understand what it means to have a music file with a sampling rate of 192 kHz, and how that correlates with the Nyquist Theorem. It's common to have vinyl rips at 192 kHz, isn't it?
Quote:
If the same transducer reproduces ultrasonics along with audible content, any nonlinearity will shift some of the ultrasonic content down into the audible range as an uncontrolled spray of intermodulation distortion products covering the entire audible spectrum. Nonlinearity in a power amplifier will produce the same effect. The effect is very slight, but listening tests have confirmed that both effects can be audible.

Originally I thought having a larger sampling rate meant that you get more "points" defined from a given continuous-time signal. From this standpoint, it would make sense to most people that having more samples would create a more accurate representation of the original continuous signal, no?

From that article, it's suggested that a higher sampling rate captures frequencies higher than that of a lower sampling rate (i.e. 44.1 kHz / 2 = 22.05 kHz maximum frequency, 192 kHz / 2 = 96 kHz maximum frequency). Why is it that taking more samples per second captures "things that don't exist on the original continuous-time signal?"

What I'm trying to ask is if you take a sine wave with an amplitude of one and period of 22 kHz, why would a higher sampling rate capture ultrasonics if it's not there in the first place?



My family owns a bunch of vinyl records and I'm interested in digitizing them. I'm not sure how to go about doing this since vinyl rips tend to sound much different from a CD rip and I'm not sure if it's due to the nature of the ripping equipment or something along those lines.
post #206 of 334

The entire reason vinyl rips are often done in the highest resolution is because typically the same people who believe vinyl has some kind of special magic believe that high sample rates have magic, too.

post #207 of 334
Or the high sampling rate because more samples per second might give a better representation of the original continuous waveform...
post #208 of 334

I also think it's because of the dynamic-range compression (loudness war) of certain digital versions of albums.  Compression is generally not an issue with vinyl.

 

Quote:

Originally Posted by Tus-Chan View Post
 

The entire reason vinyl rips are often done in the highest resolution is because typically the same people who believe vinyl has some kind of special magic believe that high sample rates have magic, too.

post #209 of 334
Quote:
Originally Posted by miceblue View Post

Maybe I don't quite understand what it means to have a music file with a sampling rate of 192 kHz, and how that correlates with the Nyquist Theorem. It's common to have vinyl rips at 192 kHz, isn't it?
Originally I thought having a larger sampling rate meant that you get more "points" defined from a given continuous-time signal. From this standpoint, it would make sense to most people that having more samples would create a more accurate representation of the original continuous signal, no?

From that article, it's suggested that a higher sampling rate captures frequencies higher than that of a lower sampling rate (i.e. 44.1 kHz / 2 = 22.05 kHz maximum frequency, 192 kHz / 2 = 96 kHz maximum frequency). Why is it that taking more samples per second captures "things that don't exist on the original continuous-time signal?"

What I'm trying to ask is if you take a sine wave with an amplitude of one and period of 22 kHz, why would a higher sampling rate capture ultrasonics if it's not there in the first place?



My family owns a bunch of vinyl records and I'm interested in digitizing them. I'm not sure how to go about doing this since vinyl rips tend to sound much different from a CD rip and I'm not sure if it's due to the nature of the ripping equipment or something along those lines.

Just rip them in 24/96. The whole resolution/sampling rate comparison is small here compared to the factors before digitizing. Clean the records, set your VTA etc. for each record, figure out if you want analog or digital RIAA, don't clip the signal. These are the things to worry about before sampling rates and such.

 

Quote:
Originally Posted by Tus-Chan View Post
 

The entire reason vinyl rips are often done in the highest resolution is because typically the same people who believe vinyl has some kind of special magic believe that high sample rates have magic, too.

Quote:
Originally Posted by miceblue View Post

Or the high sampling rate because more samples per second might give a better representation of the original continuous waveform...
 
Both of you are correct in a way, but Tus-Chan carries the majority reasoning behind it.
post #210 of 334
Quote:
Originally Posted by miceblue View Post

Or the high sampling rate because more samples per second might give a better representation of the original continuous waveform...

 

This isn't true. The Nyquist theory states that 16/44.1 is enough to exactly reproduce any waveform within the range of human hearing. More samples just add higher frequencies above the limits of human hearing. The resolution in the audible spectrum of redbook and high bitrate is identical.


Edited by bigshot - 11/11/13 at 2:17pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Music
Head-Fi.org › Forums › Misc.-Category Forums › Music › vinyl rip vs cd