Laser Turntable

[turkster]

In theory analogue reproduction can be higher fidelity than digital, but in practice you need 15 or 30 inch per second tape to match up to 96KHz/24bit digital.

Yeah but thats not CD sampling rate. CDs= 44Khz.
If as you say LP sounds better because of the equalisation used then why not try and mimic this with CD amp stages?

The thing with digital is that the process arbitarily measures the sound whereas analogue even if it doesnt necessarily contain the same amount of imformation works on a process which is a lot similar to the original sound.

I dont know a lot about audio prosesses but I am a musician and have spent some time listening to a friend of mines Linn/Naim/Rega system and Tbh the LP (despite the noise) sounds more realistic than the CD.

[Butuz]

Hmm we are getting into dangerous waters here - ive seen many a thread explode into a 10 page monster thanks to the neverending cd vs vinyl argument.

At the end of the day the technologies are fundimentally different. Some like cd others like vinyl.

Either way - both can never actually 100% capture the music as it came out of the instruments, and both get bartewardised by recording studios in between the sound leaving the instruments and the sound getting put onto cd or vinyl.

Just go with what sounds good to you and sod everyone else, thats my motto.

Butuz

[Rave]

Hmm we are getting into dangerous waters here - ive seen many a thread explode into a 10 page monster thanks to the neverending cd vs vinyl argument.

10 pages? That's nothing. It took months for my friend to beat the idea that vinyl was inherently 'superior' out of me, and now I have the zeal of the converted.

Seriously, I don't particularly want to start off a big long circular discussion either, but this is interesting stuff, and by learning about the technology better, you can make a more informed choice when it comes to buying and using the right equiptment.

Just go with what sounds good to you and sod everyone else, thats my motto.

Pretty good advice there mate, but never forget that what sounds good to you is not necessarily what's technically best. Of all the five senses, I would say that hearing is probably the least reliable (followed by sight).

At the end of the day the technologies are fundimentally different.

The thing with digital is that the process arbitarily measures the sound whereas analogue even if it doesnt necessarily contain the same amount of imformation works on a process which is a lot similar to the original sound.

Well, it may seem like the technologies are different, but actually they have a great deal in common. There are several ways of recording an analogue signal, but the most common, for studio use at least, is magnetic tape. Tape is a base layer of plastic, coated in a magnetic substance, Chromium Dioxide for example. The CrO2 (or whatever) is deposited onto the tape as millions of tiny crystals, each of which is free to move under a magnetic force. When a recording is made, the recording signal is passed through an electromagnet in the tape head, which causes the crystals to align themselves in the direction of the signal. When the tape is played back, the moving magnetic field caused by the aligned crystals passing over the head creates an electrical signal, which we then amplify etc. Now, you can see from that the the resolution of the recording is ultimately affected by the number and size of the crystals. The smaller the crystals the higher the resolution, but also the harder the tape is to make. That's why SuperVHS tapes are more expensive than VHS, for example. You can also improve the resolution by increasing the thickness of the tape, and the speed at which the tape passes over the head, which is why studios use 1" tape travelling at 15 or 30 inches per second, rather than the 1/8" tape moving at 1 7/8" ips of the Phillips Minicassette.

Now digital recording OTOH looks at the analogue signal a certain number of times a second, and each time records the precise voltage level of the signal at that point. Upon playback, the voltage is recreated from the recorded data, and the signal created goes, via some smoothing capacitors, to be amplified.

Now, here's where they're similar: although tape is 'analogue' in theory, the signal is still being broken up into discrete 'bits' of information, because the precise signal level at any given point is recorded by the individual alignment of a set number of magnetic crystals. Just as you can increase the accuracy of tape by making the crystals smaller, so you can increase the resolution with which the ADC samples the signal, going from 16 bit to 24 or 32 bit sampling for example. Just as you can increase the resolution of tape by increasing the speed at which it passes the head, you can increase the ADC's sampling rate, going from 44.1KHz to 96 or 192KHz, for example.

Now, there comes a point where increasing tape speed (or the speed at which a record spins) becomes more and more expensive and impractical, whereas thanks to Moore's Law and the ever increasing density of data storage mediums, increasing sample and bit rates on digital systems is very easy. We passed the stage where digital outperforms analogue on a cost/accuracy basis a very long time ago.

Either way - both can never actually 100% capture the music as it came out of the instruments, and both get bartewardised by recording studios in between the sound leaving the instruments and the sound getting put onto cd or vinyl.

Very true. In my experience, most of the innacuracy comes from transducers anyway. I would much rather listen to CD through expensive headphones/speakers, than to vinyl through cheap ones, that's for sure. As ever, it's a case of getting one's priorities straight.

Rich :¬)

[Trash Man]

Can anyone else remember seeing this on Tomorrows Wolrd about 10 years ago? I can

[unrealrocks]

Don't forget Studios do and have for almost 20years now used DIGITAL Masters which are sampled - yes they are are very high sample rates (in exess of 100kHz) although it is still sampling and that is then stamped onto the LP.

Also it often depends on the quality of the vinal - depending on the plant and how much the artist pays it can be cheap vynal or expensive. I've got around 100 White label LPs and they will degrade in close on 20 listens as there just not upto the standard (there also normally very warped - but thats the way there sent). A Decent quality one will make the overall sound much better.

[turkster]

Nice explanation Rave. I understand where your coming from now.
But you didnt explain whether analogue magnetic recording has any limitation regarding equivalent bit depth?

How does LP compare to CD? You said before that you need 15-30 ips of magnetic tape to compare with 96khz/24bit but what to compare with 44/khz?

In the mean time we should all be saving up for good quality seperates

[Rave]

But you didnt explain whether analogue magnetic recording has any limitation regarding equivalent bit depth?

That's because I'm beginning to get out of my depth. Lowe doesn't seem to make an appearance in here very often, but I believe he has a degree in sound recording or something similar, so he'll probably be able to give you a better answer. I'll give it a go in the mean time though.

Basically, increasing the sampling accuracy on a digital recording increases the dynamic range, which I believe is pretty much the same as the signal/noise ratio. Basically the higher this is, the less likely it is that noise will have an effect on the quiet passages of music. Really, the only recordings where you see much benefit are probably classical music pieces. For example you might want to have the hi-fi cranked so the first few bars of Beethoven's Fifth come out with a bit of impact, and a low noise system means you won't hear too much hiss while you're waiting for it to start, or in the quiet passages. Most pop music recordings are compressed, so there really isn't a great deal of variation in sound volume all the way through. In fact most pop music would probably sound fine with only 12 bit resolution.

You would think that if you had too low a resolution, that the choppiness of the sound wave would be audible, but remember that it all goes through a smoothing capacitor; you probably don't need more than say 8 or 10 bit resolution to avoid that.

Now, the s/n ratio of 16 bit audio is approx. 90dB, which is about the same as a good transistor amp can manage. 24 bit takes you beyond 110dB, which is more than enough. Really, 20bit would be plenty, but I guess you might as well future proof a standard against future advances in transistor technology.

Now, reducing noise on a tape is helped by moving it past the head faster, having wider tape etc. 30 inches per second is about the limit of the speed you can practically run a tape at, and even at that speed I believe that you start to run into problems like the bass response beginning to tail off. A standard cassette tape has a s/n ratio of maybe 60dB, perhaps 65dB for the best tapes. You can cheat a bit with Dolby Noise reduction to cut out noise. I'm guessing that studio analogue tape systems can probably get into the 80-90dB signal to noise ratio, a guess is all that is.

How does LP compare to CD? You said before that you need 15-30 ips of magnetic tape to compare with 96khz/24bit but what to compare with 44/khz?

I have no idea to be perfectly honest, and I must admit that my comparison of 30ips tape to 96/24 is not based on any detailed scientific knowledge, just hearsay.

As I've said already, 16 bit resolution is good for 90dB s/n ratio already, so that's not really much of a limitation. The 44.1KHz sampling rate OTOH is a little bit of a problem. Humans can (generally) hear up to about 20KHz. Now theoretically 44.1KHz can sample signals up to 22.05KHz accurately; the problem is that if you have any signals higher than 22.05KHz going into the ADC you get aliasing, which is where phantom audible signals will be generated where the ADC is confused by the higher frequency signals. So, to keep a good response to signals up to 20KHz, but to make sure that no signals over 22.05 get in you need a 'brick wall' filter which has a very sudden rolloff. They're not ideal from a sonic point of view and can introduce audible artifacts. I'm pretty hazy about the specific effects and the workarounds to this problem. I'm guessing that if you record at a higher frequency and downsample to 44.1KHz, you can use a computer algorithm to take out all the high frequency signals without any nasty filter artifacts.

It's a bit hard to compare vinyl to CD because they're kind of apples and pears. As I've already mentioned, vinyl is completely unsuitable for recording unequalised music, you need to roll off the bass and boost the treble to get a record that can be played back properly on a conventional turntable. If you did not the bass signals would be huge, and very hard to track, and the treble signals would be too tiny to be picked up accurately (surface noise would be a big problem). I don't really know what the signal to noise ratio is either, but it varies greatly from record to record, depending on the quality of the vinyl used, how dirty it is etc. I would guess a well cut record on virgin vinyl might be capable of 70dB s/n ratio (comparing the noise from a record to that from a cassette tape). LPs have poor stereo separation too, they can only manage about 30dB difference between channels, but that isn't much of a problem as the human head only provides 15dB.

The final problem with vinyl is that it's very hard to reproduce higher frequencies without distortion, especially on the inner tracks of the record. As I mentioned earlier, the outer track of a 33 1/3 RPM record passes the stylus at approx 20 ips. At that speed, a 20KHz signal ends up printed on the disc as a wave form with peaks .025mm (25 microns) wide. You need an elliptical or line contact stylus to be able to track that properly. On the inner tracks of the record, where the speed will be 10ips or less, even a line contact stylus has great difficulty tracking the record at very high frequencies; a cheap spherical stylus like the one on my Bush player produces noticeable distortion, which is pretty intrusive through headphones (it's not so bad on my speakers).

In the mean time we should all be saving up for good quality seperates

Yep. I forgot to comemnt on your friend's Linn/Naim/Rega rig BTW.....drool! Can you say rhythm 'n timing?

Rich :¬)

[turkster]

Nice explanation. I'll have to read that a few tmes before i remember it all

Yep. I forgot to comemnt on your friend's Linn/Naim/Rega rig BTW.....drool!
Rich :¬)

Yeah drool indeed Tis lovely kit. Massive Linn speakers, Linn turntable, Naim amp and multiple preamps, Rega CD Player, Naim Radio. Worth an absolutley shocking amount.

I try and keep taking him CDs to show him so that he'll play them while i'm there

Got myself some seperates a few months ago Nad amp Mission speakers Phillips CD Denon Tuner but been pissed off because Nad amp never worked properly and have been trying to get Richer sounds to replace it ever since

[Crazy Cricket]

what?

is it made of baby panda's or something?

Don't be daft, everyone knows baby pandas are terrible at this sort of thing, they introduce all sorts of harmonics


There was one turntable made out of 17th century timber. And there is that french table, or tables as it consists of two large floor standing units, one the turntable, the other the motor. Betwix the two you have the drive belt running in free air

[Rave]

The 17th century timber one is still available isn't it? Isn't it Michelle's top model?

The Linn LP12 as owned by myself, and presumably turkster's mate, is not particularly exotic, but the platter, after it's cast and machined, is left in a hot room for a month so the residual stresses from the manufacturing process can 'relax'. Welcome to the weird and wonderful world of high end turntable design.

Rich :¬)