Analog/Digital to CD or Vinyl

Snowman999

Active member
Vinyl is making a huge comeback. Why? I'm not really sure. But, music recorded analog does sound richer/fuller and purer on vinyl than they do on CD. But, CDs are portable and MP3s have music coming out of something that doesn't have a speaker. How does cell phone music work?

As I was told years ago (when my mind functioned properly) that analog sound waves are rounded, as digital soundwaves are flat.

Here's my question: If you record digitally and press it to vinyl is anything added to the recording that you won't get MP3 or CD?

I'm listening to vinyl that just arrived and it does sound richer than the MP3 versions. I just don't know if they recorded analog or digital.
 
Just had a long, drawn-out discussion about this last night.

[1] Vinyl sounded great because you didn't put something that sounded like crap on vinyl. It's complex, expensive, the signal needs to be "contained" (for lack of a better term) into certain guidelines and parameters based on the limitations of the medium.

[2] Tape can certainly add a pleasant sounding distortion and saturation to a signal (IMO/E, if there's anything in recording that's very difficult to duplicate digitally, it's tape saturation). But also IMO/E, it's most pleasant at the source and has very little benefit later (although I'll admit to doing a good amount of layback during mastering sessions maybe a decade or more ago).

[3] And this is the important part -- I can carefully digitize a vinyl recording and play it back for you and you will not know the difference between the two - the "warmth, character and tone" of the vinyl will be captured and reproduced with near perfection - even at 44.1kHz / 16-bit. On the other hand - If I were to "vinylize" a digital recording, you WILL be able to tell the difference. Digital is far more accurate to the source than vinyl (or tape for that matter).

The reason that a vinyl recording might sound better than a digital version of the same recording is probably the limitation of the source -- You can't crush the hell out of a recording and put it on vinyl and expect the needle to stay in the groove (so typically, a wider dynamic range). You can't push insane top and or leave problematic sibilance (same reason). You can't have phase issues in the low end. So the version cut to vinyl may very well be a better sounding recording. But it's not better because of the vinyl - It was just a more pleasant signal being cut.
 
Just because vinyl is outselling CDs right now doesn't mean that there are TONS of records being sold! There are audiophiles and collectors, but what percentage of people that you personally know play vinyl on a regular basis? I know one person who does. I've played half a dozen records in the last year - more than previous years for a long time, but I haven't bought any new vinyl.
 
I'm with MM. I just redid my turntable and have been playing some of my old records, and at the same time, dumping them to digital at 88/24. I set them up to play back at the same time (as closely as possible) and flipping back and forth, I could NOT hear any difference between the two. I can guarantee that if I transferred a good digital recording to vinyl, I could pick out the record 100% of the time.

I did do comparisons of my 20 yr old AT-121 cartridge and the new Ortofon OM10e Super. Massive difference between them. The Ortofon is higher compliance, which my tonearm needs, and the suspension of the AT was probably shot, so that wasn't unexpected. Still the OM10 had a better balance, especially with low level signals.

Digital sound waves are NOT flat. ADCs do not convert numbers into stairsteps. That's just a convenient (and very wrong) way for some people to visualize a digital signal. Play a digital wave on a oscilloscope, and you should see nice rounded waves. If not, you have a lousy ADC.

I've heard some phenomenally good analog systems over the years, but for me, the digital stuff was always better than the vinyl. Relatively high surface noise, channel separation of only 20-25dB, and frequency responses which are similar to microphones (2 or 3 dB variation from flat) means that its not as faithful to the original as a digital source. Just like you can get a mic which is dark or bright depending on where the variations are, you can get a cartridge that has a bump in the lower range to give a warm tone, one with a scooped midrange, or a bright top end depending on what you like.

People like the ritual of cleaning a record, then sitting down and reading that nice big jacket. Its involving. You can't load up a playlist of 10 vinyl records and go off to do other stuff. If forces you to sit down and listen...

As for cell phones, well, they are just today's version of the old AM transistor radio from 1960. Unless you use a good set of headphones or earbuds, they have horrible sound. If you have earbuds on, you're probably thinking more about your workout routine than listening to music. Its just something to help pass the time.
 
THANKS EVERYONE that explains it.

BTW: Talisman, I never thought of that. I have music on 90% of the time. But, you did have to stay stationary with vinyl. I have a record player/cassette/CD recorder next to this computer. I've started listening to some vinyl again, and it is kind of annoying having to flip the record over. But, I don't have CDs of many. I have records I didn't even know I had. I just found Rick Danko's first solo record and The Band's Rock of Ages. I always knew I had Grand Funk's American Band album. I just found out I put the 3D glasses in the sleeve. Great album.

Crows: I tried watching that video, and it might as well have been in Japanese. I got the visuals, but the explanation was foreign as foreign can be.
 
I got the visuals, but the explanation was foreign as foreign can be.

It's actually quite simple when you strip it down. It's not stair steps, it's points on a curve. Because everything above a cutoff frequency is removed, the result is that there's only one curve that fits the points.
 
It's actually quite simple when you strip it down. It's not stair steps, it's points on a curve. Because everything above a cutoff frequency is removed, the result is that there's only one curve that fits the points.

I understood that visually. I'm talking about the machines he was using, his explanations. If you have a general knowledge of these things, it would make perfect sense. I know when you pump sound into any of those things the waves are reading the sounds. Why? What for? How they do it? I have no idea. So, I'd first need lessons in those machines. With everything I work with, I'm a button pusher. I don't need to know "how or why", it just needs to do its job when I push the right button.

When I was told about round and flat it was when Pro Tools was only for Professional studios. I want to say I had the discussion in about 1990. I've had two Tascam 4 track cassette recorders. They were great until you started bouncing. I was most happy with my Korg D1600. I used that thing for years. In the early 2000s when I got my Mac, I bought Pro Tools 6 for pennies compared to what the studio I recorded at paid for Pro Tools 1 the professional system. That's what I've been using since. One day I'll have to upgrade my computer system, and I'll get a new Pro Tools, Final Cut Pro and Final Draft. That'll keep me satisfied till I drop dead.
 
On one level an oscilloscope is fairly simple, like a TV with a narrow function. The beam scans repeatedly from left to right and the voltage of the signal deflects it up and down. You can adjust the scan rate to match the signal's frequency, so the wave appears to remain stationary, like when a car wheel appears to stop or slowly rotate backwards under artificial lights.
 
There's some really good stuff in this topic that I had not thought about. I mean the bit about people caring what they put on vinyl. The care in making sure the stylus didn't bounce out of the groove - the thought that went into how close the grooves were and what level the recording was to fit the music. It brought back memories of the terrible records - those that squeezed ten track on each side by lowering the level, which made them terrible sounding. We also had those super records where there was real space between the grooves and no crosstalk, and then of course the trend for recycled vinyl with chewed up paper labels included.

I'm also old enough to remember the first 16bit CD players - and we had some truly excellent releases, but we also had thousands of hastily released vinyl masters slapped straight onto CD that were thin and weak. The better turntables and cartridges had very good solid bass, compared to others, so those with decent systems had better bottom end, created in their replay system, NOT on the record, so that mix on CD would be lacking what the home hi-fis were putting back in.

All the lab testing back then revealed better audio quality potential with CD, and even the provable distortion in the D to A process sounded OK, and preferable to the analogue distortion found in most hifi systems. We even went through Technics/MNatsushita's one-bit oversample process and every step forward in digital processing has made it better and better. The resurgence in vinyl is an enthusiasts one - the same as those who maintain dark room processing and optical enlargements and printing, and the same as the real film emulsion enthusiasts. A niche market. Digital has immense benefits from capture to the ear drum. I've still a real want for a reel to reel. I want to buy the same make and model I had in the 70s. I really want it. I just don't have any logical reason for this. It will NOT sound better, but my brain would convince myself it was a sensible purchase. I have sufficient test gear and experience to know that in every measurable form, it would be inferior. Sad, but true.
 
rob aylestone said:
I'm also old enough to remember the first 16bit CD players - and we had some truly excellent releases, but we also had thousands of hastily released vinyl masters slapped straight onto CD that were thin and weak.

There was AAD, DAD, DDD and of course ADD. Things sounded better once the industry had a while to figure out how to use digital. Some of the earliest CD releases I had were mastered at line level, probably compliant with RIAA curves. Technology has improved since, but towards the latter end of the '80's there was a substantial increase in mastering levels with frequency curves that take advantage of the increased range of digital. Many of those releases had clipping going on that wasn't necessarily audible or detrimental. The mixes still had enough dynamics and punch to breathe. That was before the loudness wars escalated to the brick and mortar phase.

rob aylestone said:
All the lab testing back then revealed better audio quality potential with CD, and even the provable distortion in the D to A process sounded OK, and preferable to the analogue distortion found in most hifi systems.

Lab testing is one thing, and not all distortion is created equal. I think one of the biggest improvements in recent digital technology comes from some of the distortion and saturation plugins designed to approximate analog gear. Quantization distortion, aliasing, hard clipping and excessive brick wall limiting are not things that I find preferable. Headroom matters, analog or digital.
 
I've got a few CDs that were actually transfers from vinyl. My first copy of James Gang "Yer Album". Steppenwolf's 1st album. It was really easy to figure out. Besides a few clicks and pops that were missed, some residual surface noise, one had wow that was patently obvious during a nice B3 organ passage! I had an early CD of a Jefferson Airplane album that I sent back to the manufacturer along with a letter that listed about a half dozen defects in the transfer. Apparently they remastered the whole thing and sent me a new copy that came from a real master tape, not a vinyl record transfer. It was infinitely better!

I have a CD that is from a local band who had cut a couple of albums. The lead singer/guitarist did the transfers from unopened discs that he had put away 25 or 30 years earlier. His job for years has been in audio/video production, so he took great pains to do the transfers. They sound very good, considering that the original tracks were done in one of our local studios in the early 70s. No fancy 2 inch 24 tracks. I think they were done on a 1/2" 4 track machine.

My CD copy of Iron Butterfly's Metamorphosis has an interesting artifact. During a quiet passage of Butterfly Bleu, there is a sudden burst of 60 cycle hum that isn't on the original album, or on a later redone CD. Somebody wasn't taking great pains to make sure everything was optimized!

Another oddball thing about I have from the early days. My first CD copy of Electric Ladyland, which is my #1 all time desert island record, was issued with the first CD as the first record, the second CD as the other disc. The only problem is that when they put out the record in the 60s, a lot of people used record changers, so the first disc has sides 1 and 4, the other has 2 and 3. You could put them on your record player, and it would play side 1, drop down and play side 2. Flip over both discs and you get sides 3 and 4. So the CD plays Sides 1-4 then 2-3. The song Still Raining/Still Dreaming comes AFTER Rainy Day/Dream Away. Its a minor point, but clearly nobody paid attention to those minor details. You have to wonder what else they missed, eh?
 
Quantization distortion, aliasing, hard clipping and excessive brick wall limiting are not things that I find preferable. Headroom matters, analog or digital.

Thankfully quantization distortion can be eliminated with dither, aliasing and brick wall filtering are moved well above the audible range if you record at 88.1k or above, and hard clipping can be easily eliminated using 24bit depth. The noise floor is so low that there is no reason to even approach clipping a signal when doing the original tracking. That leaves jitter as the last major source of distortion, and that can be minimized with a good clocking system.

Eventually people have to realize that digital audio has improved in the 45 years since the Soundstream recording system was introduced.
 
I've never been convinced jitter is even an audible artefact, just one that can be measured. So many of the sound sources recorded were never sonically pure anyway, so when you are recording Hammond organs, and from the 70s those terrible nasty hissy Farfisas, what kind of quality are we trying to capture anyway? Sound quality has the potential for such excellent quality nowadays I don't really see what we are talking about anymore when we talk about 'quality'. We never resolved deliberate vs accidental distortion - one is always talked about as 100% bad and the other as the opposite.
 
There's no reason that CD can't handle low frequencies. That's the easiest part of the signal to reproduce. For highly modulated bass, a cartridge is subject to mistracking at high velocities. Look up the Telarc 1812 Overture where they put a cannon shot on the record that used to screw up LOTS of playback system. When they reviewed the CD they said "This was reviewed in its analog form in Vol.4 No.5, and there is little to add to that review except to say that the hard-to-track cannon shots on analog are absolutely lethal on CD—though not hard to track! If your power amp is capable of ripping your woofers apart, the cannonshots will give it the opportunity to do so."

You can play back up to 45k with the right setup. That's why the Shibata or micro ridge stylus was developed. CD4 uses a modulated signal from 30k to 45k carrier signal for the matrixed information to be mixed with the base frequencies from 20-15k. The two are mixed to give discrete 4 channel sound, much like an FM signal provides both stereo and mono signals with only two channels.

Setup of CD4 is pretty critical, but when done properly, it works.

As for anything in vinyl adding warmth, its a red herring. Its either true to the original source or its wrong. If its "warmer" or more spacious sounding than the original master source, then it can only be because of things like errors in frequency response or phase. There is no way it extracts information that isn't present in the first place.

If you like it better, then its because you like euphonic coloration, not fidelity.
 
TalismanRich said:
Thankfully quantization distortion can be eliminated with dither, aliasing and brick wall filtering are moved well above the audible range if you record at 88.1k or above, and hard clipping can be easily eliminated using 24bit depth. The noise floor is so low that there is no reason to even approach clipping a signal when doing the original tracking.


I remember reading a number of articles from over 20 years ago that explained dither very well. Somehow that got lost on some people. Dither randomizes and decorrelates the LSB from the signal which results in a miniscule amount of noise similar to white noise or tape hiss, but it also preserves low level information inside the noise that would otherwise be lost or truncated and it removes quantization distortion. A friend of mine said he took a course where he was told, "don't dither - it just adds noise". Not true. You can add a very low level of white noise to the signal and it won't preserve anything or remove distortion because dither is not simply just adding noise.

As to hard clipping and brick wall limiting, it has more to do with the loudness wars. I disagree that 16 bit recordings need to be slammed for the sake of the noise floor, and I think movies mastered closer to -20 LUFS sound vastly superior to CDs at -7. Older CDs that weren't slammed sound better than newer releases that were. Improving digital has as much to do with actual advancements in technology as it does with not abusing it. I think that might speak to Massive's point about vinyl simply not being physically capable of handling the abuse. The technolgy itself isn't responsible for that.
 
The massive compression on CDs today has nothing to do with CD's available dynamic range. Its entirely to try to make more impact on a listener and grab his attention.

TV was terrible about this. Shows would have a volume with just speaking. Commercials would come on and EVERYTHING WAS TURNED UP TO 11!, so you turn down the TV. The the show comes back on and all the dialog would get lost so you would turn things back up.. Very irritating. LUFS was actually developed first to normalize the audio on TV broadcasts. Since the system worked, it was easy to implement into streaming services where you have a diverse number of sources which may or may not have similar volume and range.

I have a system on my Taurus sound system that actually compresses depending on how fast your are driving. Go slow and you get more dynamic range. As you speed up, those low level passages will be lost in road noise, so it ups the volume. The loud passages get a little louder, but not as much as the quiet stuff. Slow down and the volume level drops.
 
Back
Top